GB2626158A

GB2626158A - Methods of converging charged particles, reacting substances and separating substances, and devices therefor

Info

Publication number: GB2626158A
Application number: GB2300434.4A
Authority: GB
Inventors: Sideris Dimitrios
Original assignee: Genetic Microdevices Ltd
Current assignee: Genetic Microdevices Ltd
Priority date: 2023-01-11
Filing date: 2023-01-11
Publication date: 2024-07-17
Also published as: WO2024149979A1

Abstract

A method of converging charged particles 10, comprises: a) injecting the particles into a chamber 2 having a main axis; and b) applying, along the main axis, an electric field sequence. This comprises the steps of: b1) applying a first electric field gradient in a first direction of the main axis; and then b2) applying a second electric field gradient in a second direction of the main axis which is opposite to the first direction. During each of steps (b1) and (b2) the front edge of a group 12 of charged particles moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of the group is successively reduced. A method of separating substances is also described, comprising: applying an electric field gradient to charged particles of at least two substances in a chamber containing a sieving medium, the front edge of a group of the particles moving more slowly than the trailing edge of the group, to reduce the width of the group; and then performing electrophoresis on the group of charged particles along a separation channel containing a sieving medium.

Description

METHODS OF CONVERGING CHARGED PARTICLES, REACTING SUBSTANCES AND SEPARATING SUBSTANCES, AND DEVICES

THEREFOR

FIELD OF THE INVENTION

This invention relates to the field of manipulating charged particles of substances, and particularly methods of converging, that is spatially focussing, such particles. Also provided are related methods of carrying out a reaction between at least two substances, and of separating at least two substances. The invention also provides corresponding devices for performing such methods.

BACKGROUND

The ability to converge substance(s), such that the volume which the substance(s) occupies is reduced, finds benefit in a wide range of applications across many industries. One manner in which substances can be converged, is to form charged particles of the or each substance, and then manipulate the position of the charged particles using electric and/or magnetic fields. Optionally other forces such as hydrodynamic forces might be applied also. This spatial focussing can be performed on very small scales -in a so-called microfluidic environment -or on a much larger scale.

One example of an application in which a degree of focussing takes place is disclosed in WO-A-2006/070176. This describes a method of separating charged objects, specifically by electrophoresis, and a device capable of performing such a method. Examples of particles which may be separated in such a process include polymers such as proteins, DNA molecules, RNA molecules or other types of biomolecules such as biological cells. The same principles can be applied to many other types of substances. Here, an electric field gradient is applied to a sample containing a mixture of charged particles disposed in a fluid within a separation channel. The electric field profile is moved along the channel causing the sample to separate into bands of like particles, each band being located at an equilibrium position where the electric force on the particle due to the electric field gradient is balanced by the hydrodynamic force on the particle due to the fluid. The equilibrium position for each band moves along the channel due to the controlled movement of the electric field gradient. The shape of the electric field gradient (typically a substantially linear slope) also leads to some narrowing (focussing) of each band as the movement progresses, since the trailing edge of the band experiences a greater electric force than does the leading edge, causing the particles at the trailing edge to "catch up" with those at the leading edge to an extent. However, the focussing achieved is relatively weak and so the different bands at least partially overlap one another during the first part of the separation process for a significant period. If the rate of focussing could be increased, this would be of benefit since a clear separation between the respective bands could be achieved more quickly and in a shorter distance along the separation channel.

Many other applications exist in which enhanced particle convergence would be useful and each has correspondingly different benefits. Further examples will be given below.

SUMMARY OF INVENTION

In accordance with a first aspect of the present invention, a method of converging charged particles is provided, comprising: a) injecting charged particles of at least one substance into a chamber having a main axis; b) applying, along at least a first portion of the main axis of the chamber,

an electric field sequence, comprising the steps:

b1) applying a first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points within a first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a first direction of the main axis; and then b2) applying a second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points within the first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a second direction of the main axis which is opposite to the first direction; whereby, during step (b1), a group of the charged particles moves under the influence of the first electric field gradient in the first direction along the main axis, and in step (b2) the group of charged particles moves under the influence of the second electric field gradient in the second, opposite, direction along the main axis of the chamber, and during each of steps (b1) and (b2) the front edge of the group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is successively reduced.

By applying an electric field sequence of this sort to the charged particles in the chamber, a much greater rate of focussing can be achieved than was previously possible -that is, the width of the group of charged particles can be reduced more quickly. As such, the present technique can be used for example to either achieve a similar proportional reduction in the width of the group as in prior methods but in a shorter period of time or it can be used to achieve a greater proportional width reduction given the same length of time. In the first step of the sequence (b1), the group moves in the first direction and the trailing edge "catches-up" somewhat with the leading edge, and then in the second step of the sequence (b2), the same group of charged particles moves in the opposite direction and again its trailing edge (which was previously the leading edge) "catches-up". Since each, opposite, edge of the group is forced towards the other, in turn, the result is a strong converging (focussing) effect on the group of particles as a whole. It should be noted that the term "group" is used to describe a plurality of particles which could be of different types, whereas the term "band" is used to describe a plurality of particles of the same type (at least in terms of their mobility -it is likely that such a band will contain particles of a single substance, but it is not impossible that two or more substances might be present which have particles of the same mobility which will therefore remain together in the same band). Thus a "band" is a more specific type of "group". If the particles injected into the chamber include particles of different type, they will form into bands which each narrow as a result of the convergence process. The bands may focus at different rates and may move differently from one another, with the result that they may overlap with each other or separate from each other.

As it narrows, the group of particles subject to the electric field sequence will move back and forth (at least once, preferably more -in which case the movement can be described as an oscillation) along the main axis of the chamber about some effective average position. It should be noted that the converging effect is independent of the presence of any fluid or other sieving matrix (which might or might not be present in the chamber) and does not involve the attainment of equilibrium between forces on a particle. Indeed, due to the switching from the first electric field gradient to the second electric field gradient, any equilibrium condition (should a fluid be present) is typically avoided -at most, in any one implementation, at any one time one band of like particles could feasibly meet the equilibrium condition, but other bands of respectively different particle types will not (due to their different charge/mass ratios). More typically, none of the particles will be in the equilibrium condition during the converging process.

Another result of this technique is that the average electric field magnitude experienced during step b by the moving group of particles is non-zero. Due to the switching from the first electric field gradient to the second electric field gradient, particles that have moved towards a lower electric field magnitude in step (b1) will then be subject to a high electric field magnitude (of opposite direction) in step (b2) and vice versa. As such, over the whole of step b, the electric field magnitude (independent of sign) experienced by the particles averages to a non-zero value. Also at the average position about which the particles oscillate, the field magnitude will not be zero. This leads to a strong focussing effect. It should be noted that the average field experienced (or "seen") by the group of particles is not the same as the applied field gradients, since the particles will only experience the electric field present at the particular location where they are positioned at any one instance, and they will take a finite amount of time to move to another location upon switching of the fields. The electric field magnitude experienced by a group of particles during any particular electric field sequence can be modelled if the particle mobility is known.

The chamber could be of any shape and might or might not be elongate: its main axis is the line along which the electric field gradients are applied and this could have any geometric relation to the layout of the chamber However, if the chamber is elongate, preferably the main axis will be parallel to the elongate axis of the chamber. The chamber may be a sub-section of a larger cavity that may optionally also include a separation channel, as discussed below. The chamber could for example take the form of (all or part of) a cavity formed into the surface of a substrate, and need not be enclosed on all sides (although this is preferred) provided it can contain the sample in use -for instance, the chamber may have an open top surface.

The first portion of the main axis includes all locations on the main axis along which an electric field is applied at some point in time during the electric field sequence. As such, at any one point in time during the sequence, an electric field may be applied to all or just part of the first portion. The first portion could extend across the whole chamber, or only part of it. The first region of the main axis is a region of space substantially fixed relative to the main axis at least during any one cycle of steps (b1) and (b2). Thus a group of particles located in the first region will be subject to the first electric field gradient during step (b1) and then to the second electric field gradient during step (b2). The position and/or lateral extent of the first region may be changed as the method progresses (discussed below), but at a substantially slower pace than the switching between electric field gradients. As explained below, in some implementations there may be one or more additional regions, laterally offset from the first region, in which other electric field profiles are also applied during the electric field sequence. It will be appreciated that the nature of any electric field profile applied outside the first region during the method is unrestricted -for example it could include a zero-crossing point at some location outside the first region which would have little impact on the particles in the first region. The possibility of electric fields being applied along directions other than that parallel to the main axis during the sequence is also unrestricted.

The electric field sequence may or may not comprise steps additional to (b1) and (b2). In some preferred embodiments, the electric field sequence switches between step (b1) and step (b2) with no intermediate transition steps and preferably substantially zero time elapsing between steps (b1) and (b2). In other words, the electric field applied switches from the first electric field gradient to the second field gradient (or vice versa if the sequence is repeated as discussed below) in a discrete manner and preferably instantaneously. This very sudden change of field direction has the benefit that even very high-mobility particles are unable to keep pace with the changing field and so will be subject to the sequential strong focussing effect described above. If the switch between steps (b1) and (b2) is too slow, the charged particles could reach the point of minimum electric field magnitude and follow it, in which case the focussing effect would be weaker.

That said, for many particle mobilities an instantaneous switch between steps (b1) and (b2) is not essential and so in some embodiments, the electric field sequence further comprises a transition step (b1') between step (b1) and step (b2) during which the first electric field gradient is changed to the second electric field gradient, the duration of the transition step (b1') being shorter than the duration of either step (b1) or step (b2). By keeping the transition step short compared with steps (bl) and (b2), a strong focussing effect can be achieved for a wide range of particle mobilities. In preferred cases the transition step (bl ') is at least 10 times shorter, more preferably at least 50 times shorter, most preferably at least 100 times shorter, than either step (b1) or step (b2).

Similarly, the electric field sequence may preferably further comprise a transition step (b2') after step (b2) during which the second electric field gradient is changed to the first electric field gradient, the duration of the transition step (b2') being shorter than the duration of either step (b1) or step (b2), preferably at least 10 times shorter, more preferably at least 50 times shorter, most preferably at least 100 times shorter.

The applied electric field may be changed from the first electric field gradient to the second electric field gradient (or vice versa) in the transition steps (b1') and/or (b2') by translating an electric field profile along the main axis, for example. Either transition step could alternatively or in addition include a period in which no electric

field is applied.

The first and/or second electric field gradients may be stationary during steps (b1) and/or (b2) respectively. However, in a particularly preferred embodiment, during at least part of step (b1), the first electric field gradient is translated within the first region along the main axis of the chamber in the first direction, and/or during at least part of step (b2) the second electric field gradient is translated within the first region along the main axis of the chamber in the second direction. It should be noted that the whole of the first or second electric field gradient does not need to stay within the first region during this translation -part of it could move past the first region. By translating the electric field gradient in this way, the particles experience a greater average electric field magnitude than they would if the electric field gradient remained static (and hence a greater electric force). This is because the lowest magnitude part of the gradient (towards which the particles are moving) is itself translated forwards ahead of the moving particle group, with the result that a higher magnitude part of the gradient is always "seen" by the particles than would be the case with a static gradient. As such, the converging effect of the electric field sequence is further increased. In preferred cases, the translation of the first and/or second electric field gradient takes place throughout at least half of the duration of the respective step (b1) and/or (b2) -preferably the second half of the step. More preferably the translation of the first and/or second electric field gradient takes place throughout the whole duration of the respective step (b1) and/or (b2).

The electric field sequence (e.g. step b1 followed by step b2, optionally plus transition steps b1' and/or b2') could be performed a single time. However, in preferred embodiments wherein the electric field sequence is repeated a plurality of times, preferably at least 3 times, more preferably at least 10 times, still preferably at least 30 times, most preferably at least 50 times. By repeating the sequence -in effect, oscillating between the first and second electric field gradients -the converging effect will be progressed further and thus the amount of convergence increased. The number of repetitions (oscillations) will depend on the desired outcome which will in turn depend on the application in which the method is deployed. Examples will be given below. In some cases many more repetitions of the sequence, e.g. at least 100, and perhaps up to 200, may be desirable. In some applications there may also be reason to maintain the narrowed width of the group of particles after the desired amount of convergence has been achieved, e.g. for a period of time or indefinitely. As such, the repeats of the sequence may be continued for as long as desired, in order to preserve the converged group and avoid diffusion causing it to spread.

Repeating the electric field sequence causes the particles to oscillate along the main axis about some average position in the chamber. If there are multiple different particle types, they will form into bands which all oscillate about the same average position but with different amplitudes of oscillation depending on their molecular mobility or mass. Each band will focus as it oscillates.

The first and second electric field gradients could each be different, independent electric field profiles, one applied in step (b1) and the other in step (b2). However in some preferred embodiments, as alluded to above, one and the same electric field profile could be utilised to apply both gradients. For example, in step (b) the first and second electric field gradients may each be applied by applying a composite electric field profile along the first portion of the main axis of the chamber, the composite electric field profile including: a first part corresponding to the first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points and the magnitude of the electric field decreases in the first direction of the main axis; and a second part corresponding to the second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points and the magnitude of the electric field decreases in the second direction of the main axis which is opposite to the first direction; and applying the electric field sequence comprises changing the position of the first portion relative to the first axis by applying the composite electric field profile at different positions along the main axis of the chamber in steps (b1) and (b2) respectively, such that during step (b1) the first part of the composite electric field profile is applied along the first region of the main axis of the chamber, and during step (b2) the second part of the composite electric field profile is applied along the first region of the main axis of the chamber.

This approach simplifies control of the method, since just one electric field profile (the composite electric field profile) need be stored in memory and it can be applied to different positions along the main axis of the chamber in the respective steps, to achieve the necessary effect. It should be noted that such a composite electric field profile may typically include a zero-crossing point (i.e. a point on the profile having zero electric field magnitude, the profile having values which spatially change continuously from positive to negative from one side of the point to the other), although it could instead be discontinuous.

Such a composite electric field profile can be used whether or not the sequence includes transition steps (b1') and/or (b2'). If transition steps are included, preferably during transition step (b1') the first electric field gradient is changed to the second electric field gradient in the first region by moving the composite electric field profile along the main axis. Similarly, it is preferable that during transition step (b2') the second electric field gradient is changed to the first electric field gradient in the first region by moving the composite electric field profile along the main axis. In both cases the moving may be a translation. In particularly preferred implementations, the electric field sequence may be repeated a plurality of times by sequentially moving the composite electric field profile along the main axis in alternating directions.

In many implementations, all of the charged particles injected in step (a) will have the same charge sign. That is, either all of the charged particles injected in step (a) are positively charged or all of the charged particles injected in step (a) are negatively charged. This may be preferable in order to simplify handling. In the case of an aqueous solution, the (apparent) charge on a particle can be controlled by controlling the pH of the solution carrying the particle. The apparent charge of a particle in a solution depends on the isoelectric point of the particle and the pH of the solution. If the pH is set exactly at the isoelectric point, the charge is zero. If the pH of the solution is lower than the isoelectric point then the particle carries a net positive charge, and vice versa. Hence in some preferred implementations, the pH of a solution carrying particles of mixed type may be selected so that all of the particles have a positive charge or all have a negative charge. In this way, all of the particles injected into the chamber will move in the same direction as one another in the presence of an electric field and so can form part of one group of particles which is converged in the above-mentioned manner.

However, where the injected particles include particles of different substances, it is also possible to have a mixture of positive and negative charges at the same time. For instance, if the pH of the solution is above the isoelectric point of a first particle type but below that of a second particle type, some of the charged particles will be positive and others negative. In such a case, only the first particle type or the second particle type (not both) will be focussed by the sequential first and second electric field gradients described above. The particles of the opposite sign will experience a diverging effect. If the diverging particles are not of interest, this may not be problematic and so the method could be performed as already described in this scenario, with only some of the injected particles forming a group which is converged as the method progresses.

In other preferred implementations, however, both positive and negative charged particles can be converged simultaneously, in respective groups which will become located in different regions of the chamber along the main axis. This can be achieved as follows: in step b1) simultaneously with applying the first electric field gradient in the first region, a third electric field gradient is applied in a second region of the main axis laterally offset from the first region, in which the sign of the electric field is the same as that of the first electric field gradient at substantially all points within the second region and the magnitude of the electric field increases along the main axis in the second region in the first direction; and in step b2) simultaneously with applying the second electric field gradient in the first region, a fourth electric field gradient is applied in the second region, in which the sign of the electric field is the same as that of the second electric field gradient at substantially all points within the second region and the magnitude of the electric field increases along the main axis in the second region in the second direction; whereby, during step (b1), in the first region a first group of the charged particles having a first charge sign moves under the influence of the first electric field gradient in the first direction along the main axis while in the second region a second group of the charged particles having the opposite charge sign moves under the influence of the third electric field gradient in the second, opposite, direction along the main axis, and in step (b2) in the first region the first group of charged particles moves under the influence of the second electric field gradient in the second direction along the main axis, while in the second region the second group of charged particles moves under the influence of the fourth electric field gradient in the first direction along the main axis, and during each of steps (b1) and (b2) the front edge of each group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of each group of charged particles is successively reduced.

In essence, this preferred implementation of the method involves performing the already-described technique simultaneously in both first and second regions along the first axis, the direction of the gradients being reversed in the second region relative to those in the first region. Particles of a first charge sign (e.g. positive) located initially in the first region will form a first group which converges in the first region while those located initially in the second region will be dispersed (some moving into the first region). Meanwhile particles of the opposite charge sign (e.g. negative) located initially in the second region will form a second group which converges in the second region and those located initially in the first region will be dispersed (some moving into the second region). In this way, two groups of particles (one positive and the other negative) will be formed, and each will converge as the method proceeds.

Preferably the first and third electric field gradients will have substantially the same spatial rate of change of electric field with distance along the main axis as one another (but in opposite directions), and likewise preferably the second and fourth electric field gradients will have substantially the same spatial rate of change of electric field with distance along the main axis as one another (but in opposite directions). In other words, the gradients applied in the second region will be substantially mirror images of those applied in the first region. However this is not essential and in other cases, it may be appropriate to apply differently sloped gradients to each region, for instance if it is known that the particles of one charge sign are of significantly different mobility from those of the opposite charge sign.

The first and second regions are each substantially fixed relative to the main axis, although as described above in relation to the first region either or both could move and/or change in size slowly relative to the steps of the electric field sequence. The first and second regions are distinct regions of space and do not overlap one another. They could abut one another but preferably have a non-zero space between them, forming a third region which is preferably smaller than the first or second regions. The size of the third region can also be varied as the method progresses (at a slower pace than the steps of the sequence). Between the two regions in which particle convergence takes place, the third region acts as a buffer zone, beneficially reducing the likelihood of a particle of the first group moving out of the first region and reaching the second region (where, having the opposite charge sign from the second group, it would be dispersed and hence lost from the first group). The third region could be a region in which no electric field is applied during the electric field sequence. However, in more preferred embodiments, a non-zero electric field may be applied in the third region which is preferably configured to move particles back towards their respective group. Since both positive and negative particles can move into the third region, mixing (and optionally reaction) of such particles can take place here.

Hence, preferably: in step b1), simultaneously with applying the first and third electric field gradients, a non-zero electric field of the same sign as that of the first and third electric field gradients is applied in the third region, the non-zero electric field preferably having a spatial rate of change of electric field with distance along the main axis which is less than those of the first and third electric field gradients, most preferably being substantially constant across the third region; and in step b2), simultaneously with applying the second and fourth electric field gradients, a non-zero electric field of the same sign as that of the second and fourth electric field gradients is applied in the third region, the non-zero electric field preferably having a spatial rate of change of electric field with distance along the main axis which is less than those of the second and fourth electric field gradients, most preferably being substantially constant across the third region.

By arranging the non-zero electric field in the third region to be of the same sign as the gradients on either side of it in each respective step of the sequence, negative particles which reach it will be moved in one direction along the main axis (towards the region in which that group is being converged) and positive particles will be moved in the opposite direction. This reduces loss of particles from either group and thus enhances the converging effect. In preferred examples the nonzero electric field in the third region has a spatial rate of change of electric field with distance along the main axis which is less than those of the gradients on either side of it (i.e. a shallower slope), most preferably being flat (substantially constant). This means that any converging or diverging effect of the field in the third region will be weaker than in the first or second regions and is preferably absent, making the region substantially agnostic to charge sign -neither positive nor negative particles will experience any significant diverging effect in the third region.

It should be noted that, while the charge sign of a particle has been described above as dependent on the pH of a solution in which the particle is carried, it is not essential for the method to be performed in a solution. The method could, for instance, be implemented as a gaseous phase system in which case the changed particles could be generated by ionising the one or more substances. This will usually result in all-positive particles (although it is dependent on the process by which ionisation is carried out).

Preferably, the first and second electric field gradients have substantially the same spatial rate of change of electric field with distance along the main axis as one another. That is, the slope of the first electric field gradient applied in step (b1) in the first region is substantially the same as that of the second electric field gradient applied subsequently in step (b2) in the same region. This produces a symmetric converging effect with the group of particles in the first region being caused to oscillate around an average position approximately in the centre of the first region.

In other cases the slopes of the two gradients could differ from one another if a less symmetric outcome were desired. If a second region is provided with third and fourth sequential field gradients as described above, it is also preferably that the third and fourth electric field gradients have substantially the same spatial rate of change of electric field with distance along the main axis as one another, for the same reason.

It may also be desirable for the first and second electric field gradients to have substantially the same maximum and minimum electric field magnitudes as one another. In this way the electric forces which the particles are subject will be of similar magnitude in each step (but in opposite directions) which again contributes to a symmetrical oscillation of the group of particles. The same consideration preferably applies to the third and fourth electric field gradients.

Similarly, it is preferred that the first and second electric field gradients have substantially the same profile shape as one another. As such the difference between the electric field experienced by particles at the leading edge of the group and those at the trailing edge of the group will be substantially the same in step (b1) as in step (b2) and this improves the converging effect. The same requirement preferably applies to the third and fourth electric field gradients.

The electric field gradients could have various shapes provided in each case the electric field magnitude is greater at one end of the gradient than at the other. However, preferably the first and/or second electric field gradient is monotonic (i.e. continuously increasing or continuously decreasing, in a spatial rather than temporal sense) at least along the first region of the main axis. In preferred examples, the first and/or second electric field gradient is substantially rectilinear or curved, such as parabolic or exponential. It will be appreciated that on a very small scale the electric field may be a stepwise approximation to such shapes, as a result of the discrete nature of some electric field generating means (e.g. electrode arrays). All of these preferred features apply equally to the third and/or fourth electric field gradients in the second region.

It is preferred that steps (b1) and (b2) have substantially the same duration as one another. This has the result that the particles experience the opposing forces for substantially equal durations, which if the first and second electric field gradients are of substantially the same slope, will help to achieve a more symmetric converging effect. This is not essential but the durations of the two steps should be at least the same order of magnitude as one another.

It will be appreciated that the electric field sequence could potentially include one or more additional steps without detracting from the overall converging process described above. However, preferably the steps (b1) and (b2) together form the major part of the sequence and any additional steps occupy a minor portion of the process time. In preferred cases, where the electric field sequence is repeated a plurality of times, steps (b1) and (b2) together occupy at least 80%, preferably at least 90%, more preferably at least 99%, of the duration of each cycle. (A "cycle" means all the steps in the sequence being performed in turn, once).

In some particularly preferred implementations, the electric field sequence may consist of steps (b1) and (b2), plus optional transition steps (b1') and/or (b2'), the electric field sequence preferably being repeated a plurality of times. In other words, the electric field sequence preferably includes only step (b1) followed by step (b2), or preferably includes only steps (b1), (b1'), (b2) and (b2') in that order, and there are no additional steps in the sequence.

The cycle time will depend on the desired outcome and the mobilities of the particles involved. In preferred examples, the electric field sequence has a total duration (i.e. cycle time) of between 0.1 and 100 seconds, preferably between 0.1 and 10 seconds, more preferably between 0.1 and 5 seconds, most preferably between 1 and 3 seconds. Different cycle times will be appropriate for different applications of the method and examples will be given below. The cycle time can be user-set via control of the electric fields applied to the chamber.

The amount of movement experienced by the particles during the converging method will depend, inter alia, on their mobility, the electric field magnitudes and the durations for which the various fields are applied. This can be predicted by computer modelling of the process and, if necessary the various parameters can be adjusted to achieve the desired amount of movement -i.e. the spatial extent of the particles' oscillation. The desired oscillation width will depend on the application but in some preferred examples, in step (b) the electric field sequence is repeated a plurality of times, the first and second electric field gradients and the duration of each cycle being configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range 0.1 to 10 mm, preferably 0.5 to 5 mm, more preferably 1 to 3 mm. The width of the oscillation may be controlled to stay substantially constant over the duration of the method, or could be controlled to narrow as the method progresses, e.g. by shortening the cycle time (preferably gradually). Narrowing the oscillation width during the method reduces the time needed to achieve a certain level of group convergence.

The oscillation of the first group of particles will take place in the first region of the main axis of the chamber (i.e. where the first and second electric field gradients are sequentially applied). In some preferred examples, the first region of the main axis of the chamber has a length (along the main axis) in the range 0.1 to 5 cm, preferably 0.5 to 2.5 cm, more preferably 0.5 to 1.5 cm. The second region Of present) may also have a length in the same range.

VVhile some examples of oscillation width and region dimensions have been given above, in practice the utility of the disclosed method is not limited to any particular scale. The above examples are primarily suitable for small scale, e.g. microfluidic, implementations as might be used in scenarios where the converging method is part of a diagnostic test or assay involving biological cells for instance. However, in other applications the technique could be employed on a much larger scale, e.g. in reacting large amounts of substances to generate products in an industrial setting. The physical construction of the device and the manner in which the electric fields are generated may differ according to the scale (as discussed below) but the principle of the method remains the same.

Similarly, the nature of the charged particles injected in step (a) will depend on the application in question. In some cases, the particles could all be of the same type -for instance if the sole objective is to increase the concentration of the substance (i.e. reduce the volume which the amount of substance occupies). However in many applications there will be a mixture of particles, the aim being to react and/or separate particles of different type. Hence, preferably, in step (a) the charged particles are of a plurality of substances and include charged particles of different mobility, and during step (b) the group of charged particles additionally forms into a plurality of bands due to the influence of the electric field sequence, each band containing charged particles with the same mobility, and the width of each band is successively reduced. Thus, both the individual bands and the group of particles as a whole (i.e. the set of bands) converge as the method progresses, due to the already-described mechanism.

As noted above while the first region (and second, if provided) are generally fixed relative to the main axis at least during any one cycle of the sequence, it is possible to vary the size and/or location of the region(s) over a longer time frame. Therefore in some preferred embodiments, in step (b) the electric field sequence is repeated a plurality of times, the size of the first and/or second regions being reduced as the method proceeds by reducing the width of the main axis along which the respective electric field gradients are applied between successive cycles at least once, preferably every N cycles where N is an integer number greater than or equal to 1. Most preferably N is greater than 1. By reducing the size of the or each region as the method progresses, the oscillation with can be reduced and the converging effect further enhanced.

Additionally, at the start of the process, the injected particles (or "plug") may be spread over a relatively large volume within the chamber It can even be centimetres in length. In such cases, if an oscillation were implemented in which where the first and second field gradients extend over only a few mm, this could crop' the injected plug and essentially leave large parts of the injected sample outside of the oscillation.

Therefore the electric field gradients may need to be applied over a relatively wide region initially to capture a sufficient number of the particles. As the method progresses the particles will converge together and so the width of the region can be correspondingly reduced. Thus, in some implementations, step (b) may include a particle-collection phase and a subsequent particle-converging phase, the size of the first and/or second regions being greater in the particle-collection phase than in the particle-converging phase. The particle-collection phase could include a single cycle of the sequence, or multiple cycles. However typically the number of cycles in the particle-collection phase will be less than the number in the subsequent particle-converging phase.

For similar reasons it may be desirable to vary the slope of the electric field gradients as the method proceeds, optionally in line with any variation of region size. In a preferred case, the spatial rate of change of electric field magnitude with distance along the main axis of the respective electric field gradients is increased as the method proceeds, preferably being lower in a particle-collection phase than in a subsequent particle-converging phase. For instance, in an initial cycle of the sequence the slope of the first and second electric fields may be relative shallow (and optionally applied over a wider area) to capture particles and apply a weak focussing effect to start bringing them together. Then, the gradients may be increased (and optionally the regions narrowed) to strengthen the focussing effect and achieve the convergence desired on the captured particles. In an example, there may be a single oscillation period applied with a wider field that occupies about 5 cm and causes an oscillation of about 1.5 cm. This single back-forth step could shrink an initial 1 cm plug to about 5 mm width. Subsequent repeats of the electric field sequence can be performed over a much narrower first region to achieve the desired strong focusing effect.

Other methods of capturing the particles before beginning convergence are possible. For instance, in another embodiment the method further comprises, after step (a) and before step (b): a') collecting the charged particles together to place a major proportion of the group of charged particles, preferably substantially all, within the first portion of the main axis of the chamber.

Preferably, step (a') comprises: a'l) applying a static electric field gradient along a second portion of the main axis which is larger than and includes the first portion, the static electric field gradient including a zero crossing point begin which is located in the first portion, whereby the charged particles move under the influence of the static electric field gradient towards the zero crossing point.

This achieves a weak focussing effect, bringing the particles into the first portion of the chamber in order that they will be subject to the electric field sequence applied in step (b). For instance, this collection step could comprise the application of a wide and shallow field gradient that is static and crosses zero near the centre of the future oscillation. It is possible to include both a particle collection step (a') and subsequently narrow the region(s) and/or increase the slope of the field gradients as the repeats of step (b) progress if desired.

The first aspect of the invention further provides a device for converging charged particles, comprising: a chamber into which charged particles of at least one substance are injected in use, the chamber having a main axis; an electric field generator configured to apply electric fields along at least a first portion of the main axis of the chamber; and a controller configured to control the electric field generator, and programmed to control the electric field generator to apply an electric field sequence along at least the first portion of the main axis of the chamber, the

electric field sequence comprising the steps:

b1) applying a first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points within a first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a first direction of the main axis; and then b2) applying a second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points within the first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a second direction of the main axis which is opposite to the first direction; whereby in use, during step (b1), a group of the charged particles moves under the influence of the first electric field gradient in the first direction along the main axis, and in step (b2) the group of charged particles moves under the influence of the second electric field gradient in the second, opposite, direction along the main axis of the chamber, and during each of steps (b1) and (b2) the front edge of the group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is successively reduced.

The device provides all the benefits of the method already described. Suitable examples of physical constructions for the device include the electrophoresis devices disclosed in WO-A-2006/070176 or WO-A-2012/153108, although the controller will of course be programmed differently in the manner specified above. Preferably, the controller is further programmed to control the electric field generator to perform any of the preferred implementations of the method described above The electric field generator could take any form which enables the application of electric field gradients of the sort described along the main axis of the chamber, and the changing of one field gradient to another. In preferred examples, the electric field generator comprises an array of electrodes disposed along at least the first portion of the main axis of the chamber. In use, appropriate voltages will be applied to each electrode so as to give rise to the desired electric field gradients. Electrode arrays are particularly well suited to the generation of the electric field gradients in small scale (e.g. microfluidic) implementations of the device. The array of electrodes could have the same general form as disclosed in WO-A-2006/070176 or WO-A-2012/153108, for example.

Preferably, the chamber is provided with at least one input port for injecting a sample into the chamber, the sample comprising at least the charged particles of at least one substance. Desirably, the chamber is provided with at least one exit port for extraction of groups or bands of charged particles from the chamber. Additional electrodes may be associated with the input and/or exit ports to enable the application of further electric fields for injecting the particles into or drawing them out of the chamber.

In preferred examples the device further comprises a detector adapted to detect groups or bands of charged particles in the chamber, the detector preferably being adapted to image the groups or bands.

As mentioned above the disclosed method of converging charged particles finds use in many applications and industries. The second aspect of the invention relates to one such application Accelerating the rate of certain reactions (the term "reaction" is used here to encompass both chemical reactions and biological interactions, such as binding events) is of critical importance to many areas of science and industry. For example catalysts are used to transform heavy oil into gasoline or jet fuel. Reducing incubation times in immunoassays substantially affects the speed and throughput of important diagnostic tests for infectious diseases, dementia, cancer and cardiovascular conditions. These are only a few examples.

Traditionally there are four ways of increasing the speed of a reaction: 1) Use of catalysts: More than 80% of the chemical industry uses catalysts to increase the speed of reactions.

2) Increasing the temperature: By increasing temperature, the average speed of the particles and therefore the rate of reaction is increased.

3) Increasing the concentration of the reactants: Increasing the concentration increases the chances of collision between particles and therefore the rate of reaction.

4) Increasing surface area of the reactants: Increasing the surface area of the particles means increasing the possibility of reaction for any given molecule.

Electrokinetic mixing has also been proposed as an alternative to (2). This involves applying alternating current through a fluid to cause local heating and thus buoyancy-driven flow of the fluid, resulting in mixing. Techniques for further accelerating reactions would be desirable.

A second aspect of the invention therefore provides a method of carrying out a reaction between at least two substances, comprising: (i) providing charged particles of the at least two substances; and (ii) converging the charged particles within a chamber using the method of the first aspect of the invention, whereby the charged particles of each substance form a respective band, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting the reaction between the substances.

Thus the reaction method uses the method of converging charged particles described above to accelerate the reaction between the at least two substances, which could be carried in a liquid or exist in gaseous form. As noted above, the term "reaction" (and related words such as "react") refers here to both chemical reactions and biological interactions, such as binding events. The technique forces the particles of the different substances to coincide on the same spatial position over a period of time, as well as increasing the concentration of each substance (i.e. the amount of substance per unit volume). As a result of both effects, the reaction proceeds much faster than in previous methods. Depending on the nature of the various particles (e.g. their respective mobilities) and on the field parameters, the respective bands of particles may cross over one another as they oscillate in the chamber, or may simply occupy the same (increasingly small) volume. The close proximity of the different particles enhances the likelihood of collisions and therefore promotes the reaction. In addition, the reaction may also be enhanced by the average speed (and therefore momentum) of the particles being increased by the oscillation, causing increased collisions as the bands overlap one another in the chamber In tests, the disclosed method has successfully reduced the time taken for example reaction processes (including SARS-CoV-2 antigen tests, Lassa fever assays and tests for the influenza virus) very substantially.

An additional benefit of the disclosed method that has been observed in some cases is a significant increase in reaction efficiency -that is, the amount of analyte required for a successful reaction is substantially reduced. This in turn uses less resources and reduces costs. For example, in the case of an immunoassay, a probing antibody (or antigen) is used. In conventional assay methods typically the target analyte (antigen) is expected to be in low concentration (or rather, the interesting samples are those that have low abundance of target molecules -for example in a covid-19 antigen test, early detection is critical and there, the antigens are present in minute quantities). In order to ensure that incubation (i.e. the reaction) takes place, typically very high concentrations of probing antibody (or probing antigen in the reverse reaction) are used to ensure that each target antigen will find a matching antibody to react. If the concentration of antibodies is low then the chances of matching are small and incubation rate is extremely low.

It is also usually necessary, when using conventional techniques, to incubate the sample prior to the assay being performed, in order to increase the amount of target antibody/antigen complex present so as to reach detectable amounts. Sufficient incubation can take hours or even days.

However, since the presently disclosed method inherently concentrates and collides both probing and target molecules, the starting amount of the probing analyte can be greatly reduced. Even at very low initial concentrations when injected, both probing and target molecules get concentrated by the electric field sequence and, due to the repeated collisions, the reaction efficiency may be higher than with conventional incubation, even when in the conventional incubation the probing molecule is at very high concentrations. For example, the present inventor has found that for a SARS-CoV-2 antigen test, where the target molecule is nucleocapsid protein, in order to achieve incubation in one hour using conventional techniques, it was necessary to use concentrations of the order of 500pg/pL. The incubation was volatile with a significant failure rate. In contrast, using the presently disclosed method, and no prior incubation, the probe concentration was reduced to 50 pg/pL yet it was possible to achieve much higher signals from the reacted complex (Ab+Ag).

It will be appreciated that while the benefits of the reaction method have been illustrated by reference to an immunoassay scenario, the same principles apply to all types of chemical (or biochemical) reactions The method may include any of the preferred features of the first aspect of the invention discussed above. In the context of performing a reaction it is preferred that the particles of all substances which are required to react are arranged to have the same charge sign, so that the resulting bands converge in the same region of the chamber It is possible to react particles of opposite sign (since they will mix in some parts of the chamber) but the efficiency will be lower Preferably, to better promote the reaction, the electric field sequence will be repeated many times so that there are many oscillations of the bands (and hence more collisions). For example, in some advantageous implementations, in step (ii) the electric field sequence is repeated a plurality of times, preferably at least times, more preferably at least 50 times, still preferably at least 100 times. Alternatively or in addition the electric field sequence may be repeated (i.e. the oscillations continued) until the reaction has progressed a desired amount. For instance, in step (ii), the electric field sequence may be repeated until a predetermined amount of one of the at least two substances has reacted, the predetermined amount preferably being 30%, 50%, 70%, 80%, 90%, 99% or 100%. The percentages here relate to the proportion of reacted analyte: 100% corresponds to the stage where all molecules of one of the reactants have reacted. The percentage refers to the primary reactant (substance), i.e. the one that would ideally react 100%. Excess secondary reactant(s) may be added to increase the amount of primary reactant that actually reacts. For example, in an antigen test, the primary reactant is the antigen and the secondary reactant is the antibody. In order to establish the correspondence between number of oscillations and reaction progress, one or more runs of the process may be performed beforehand (e.g. as part of a calibration process on the same or a corresponding device). For instance, a test run may be performed in which N oscillations are performed, after which the resulting contents of the chamber are analysed, e.g. by performing a separation and using a detector to look for the presence or absence of a particular component, preferably the primary reactant. Multiple runs with different values of N may be performed in order to establish the minimum value of N at which there is substantially no primary reactant detected (meaning that 100% of the primary reactant has reacted). Subsequently, the method can be performed on the real samples with N (or more) oscillations if it is desired to react 100% of the primary component, or with a corresponding fraction of N oscillations should some lesser degree of reaction progress be desired (e.g. 0.5N if only 50% of the reactant is to be reacted).

In the context of performing a reaction, the cycle time of the electric field sequence will be selected with the primary aim of converging the particles into as small a volume as possible, in dependence on the particles' mobilities and the field parameters. In preferred examples, the cycle time is between 0.1 and 100 seconds, preferably between 0.1 and 10 seconds, still preferably between 1 and 5 seconds. Generally, shorter cycle times are preferred since this results in a narrower oscillation width of the bands, thereby keeping the substances in closer proximity for more of each cycle. However, if the particles are of low mobility the cycle time can be lengthened since the amount of motion will be relatively small.

Longer cycle times can be advantageous in the context of performing a reaction since this helps to collect sample particles which may initially be spread wide apart in the chamber, giving them more time to collate together under the influence of the applied field profiles. In some types of reaction there many only be a small amount of the sample present and so it is important to collect as much of it as possible.

Preferably, in small scale implementations of the reaction method (e.g. in a microfluidic environment), the first and second electric field gradients and the duration of each cycle are configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range of less than 10 mm, preferably less than 3 mm, more preferably less than 2 mm. Such small dimensions ensure that the particles of each substance are highly localised together thereby enhancing the reaction. Of course, in large scale implementations, e.g. where a reaction is performed in order to manufacture a chemical product on an industrial scale, the oscillation width may be substantially larger (and the cycle time may also need to be correspondingly longer).

Generally the objective when accelerating a reaction is to achieve maximum sample concentration and maximum band overlap during the oscillation. This maximises the reaction rate. Therefore oscillations with a small width and the use of steep electric field gradients, e.g. changing from 0 to 800V/cm over a first region having a width of 1 cm or less, will be advantageous in this scenario.

As noted above, it is preferable that the parameters of the electric field sequence and the respective motilities of the charged particles of each substance are such that the respective bands of each substance cross over one another within the chamber during step b, preferably repeatedly. For example, if a first band of high mobility particles is moving in the first direction followed by a second band of lower mobility particles, when the first electric field gradient is switched to the second electric field gradient, both bands will start to move in the opposite direction. The first band will overtake the second band, due to its higher mobility, crossing over the second band as it does so. During the crossing-over, collisions between the particles of each substance will take place.

In an advantageous development of the reaction method, after the electric field sequence has been performed a desired number of times (or the reaction has reached a desired state), the result of the reaction may be interrogated. Since some of the original unreacted substance(s) may remain, in a preferred implementation this may involve separating the various particle types now present in the chamber. Therefore the method preferably further comprises, after step (ii): (iii) separating the charged particles according to their type by performing electrophoresis on the content of the reaction chamber, whereby a reaction product is separated from any remaining amounts of the at least two substances. There may be more than one reaction happening in parallel so, in the general case, all reaction products will preferably be separated from one another (and any residual reactants). Example electrophoresis processes suitable for this will be described in more detail in relation to the third and fourth aspects of the invention.

Another further advantage of the disclosed reaction method is that when using it as a first step to a subsequent separation step (for example electrophoresis), using a lower concentration of secondary reactant (as is possible for the reasons explained above), reduces the chances of the unreacted signal swamping the small reaction-product signal.

In the present aspect, the electrophoresis process may be carried out in a separation channel which may or may not be fluidically connected to the chamber. For instance the separation channel could be separate from the chamber, the chamber contents being removed from the chamber and transferred to the separation chamber e.g. by pipette. Alternatively the separation channel could be connected to the chamber via a suitable conduit, or it could be contiguous with the channel (e.g. forming part of one and the same cavity). The separation could optionally be carried out at the same time as another batch of reactants is being reacted in the chamber.

In some implementations it may alternatively or additionally be desirable to output the product made by the reaction which may be a useful substance in itself. For example, the product of an enzymatic reaction. Such reaction may be facilitated by the methods of the invention. The product may be separated using the separation methods of the invention, or alternative methods as would be understood by a person of skill in the art. As such the method of the invention preferably further comprises, after step (ii) or after step (iii) (if performed): (iv) extracting a reaction product, preferably by applying an electric field configured to cause charged particles of the reaction product to move to a port through which the reaction product is removed.

In one scenario where the primary aim of the reaction is to produce quantities of the reaction product, the method may be run either multiple times or continuously. In the former case, for example, a certain amount of each substance may be injected into the chamber, the electric field sequence applied to perform the reaction, and then the product extracted before the process begins again with more of the reactants being injected. In the latter case the electric field sequence could be repeated substantially continuously with reactants being injected at intervals and the contents being extracted at certain points.

Accordingly, the invention also provides a method of detecting an analyte (such as a viral and/or bacterial antigen) with a binding protein that specifically binds the analyte, said method comprising: (i) providing charged particles of the analyte and the binding protein; and (ii) converging the charged particles within a chamber using the method of the first aspect of the invention, whereby the charged particles of each of the analyte and the binding protein form a respective band, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting interaction (such as a reaction or binding) between the analyte and the binding protein.

The term "binding protein" refers to any protein that can specifically bind to the target analyte (i.e. antigen). Preferably, the binding protein is an immunoglobulin molecule (i.e. antibody) or antigen-binding portion thereof which specifically binds to the target analyte (i.e. antigen). The term "specifically binds", "specific binding" or "binds" are used interchangeably and refer to an immunoglobulin molecule, or antigen-binding portion thereof, binding to an antigen or an epitope of said antigen with greater affinity than for other antigens or epitopes. Such affinity is measured via the equilibrium dissociation constant (Kd) and may be measured using standard procedures known in the art. The antibody is preferably an IgG antibody or fragment or derivative thereof but may alternatively be an IgA, IgM or IgE antibody or fragment or derivative thereof The binding protein may be a fragment, derivative or analog of said antibody. The terms "fragment", "derivative" and "analog" refer to a polypeptide that substantially retains the same biological function or activity of the full-size immunoglobulin molecule, for example the same antigen binding specificity. The immunoglobulin molecule may for example be a monoclonal antibody, a polyclonal antibody, a single heavy-chain variable domain, a Fab fragment, an scFV fragment, a diabody or any other antibody fragment or fusion as would be understood by a person of skill in the art. Preferably, the immunoglobulin (antibody) is a monoclonal antibody.

Accordingly, the method herein disclosed may be used in the detection of a variety of viral and/or bacterial infections. Examples of viral infections that may be detected include, but are not limited to, human immunodeficiency virus (HIV), ebola virus, hepatitis virus (A, B, or C), herpes virus (e.g, VZV, HSV-I, HAV-6, HSV-II, and CMV, Epstein Barr virus), adenovirus, influenza virus, flaviviruses, echovirus, rhinovirus, coxsackie virus, coronavirus (e.g. SARS-CoV, SARS-CoV2, MERS-CoV), respiratory syncytial virus, mumps virus, rotavirus, measles virus, rubella virus, parvovirus, vaccinia virus, HTLV virus, dengue virus, papillomavirus, molluscum virus, poliovirus, rabiesvirus, JO virus or arboviral encephalitis virus.

Examples of possible bacterial infections include, but are not limited to, Chlamydia, rickettsial bacteria, mycobacteria spp., staphylococci, streptococci, pneumonococci, meningococci and gonococci, klebsiella spp., proteus, Serratia, pseudomonas spp., Legionella, Corynebacterium diphtheriae, Salmonella spp., bacilli, Vibrio cholerae, Clostridium tetani, Clostridium botulinum, Bacillus anthracis, Yersinia pestis, Mycobacterium leprae, Mycobacterium lepromatosis, and Borrelia. Examples of possible parasitic infections to be detected include Entamoeba histolytica, Balantidium coli, Naegleria fowleri, Acanthamoeba, Giardia lambia, Cryptosporidium, Pneumocystis carinii, Plasmodium falciparum, Plasmodium vivax, Babesia microti, Totpanosoma brucei, Trypanosoma cruzi, Leishmania donovani, Toxoplasma gondli, Nippostrongylus bras/liens/s. Examples of possible fungal infections to be detected include Candida (albicans, krusei, glabrata, tropicalis, etc.), Corptococcus neoformans, Aspergillus (fumigatus, niger, etc.), Genus Mucorales (mucor, absidia, rhizopus), Sporothrix schenkii, Blastomyces dermatitidis, Paracoccidioides brasiliensis, Coccidioides immitis and Histoplasma capsulatum.

The second aspect of the invention also provides a reactor for performing a reaction between at least two substances, the reactor comprising a device for converging charged particles according to the first aspect of the invention, whereby in use charged particles of each substance form a respective band in the chamber, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting the reaction between the substances.

The reactor achieves all the benefits already described in relation to the second aspect and the controller can be further programmed to control the electric field generator to perform any of the preferred methods mentioned. The reactor can have any of the preferred features already described in relation to the device of the first aspect of the invention.

In preferred embodiments, the reactor further comprises a separation channel fluidicly connected to the chamber, for performing electrophoresis therein, the electric field generator being further adapted to apply electric fields along the separation channel and the controller being further configured to control the electric field generator to apply a particle-separation electric field along the separation channel. As explained above, in some implementations it may be desirable to perform a separation on the content of the reaction chamber and this is a convenient arrangement for doing so.

Alternatively or in addition to an extraction port from the chamber being provided, the reactor may further comprise an extraction port for extracting a separated band of particles from the separation channel. Typically the band to be extracted will be the reaction product.

Preferably, the separation channel is connected to an exit port of the chamber or the separation channel is contiguous with the chamber and aligned along the main axis of the chamber. Benefits of this arrangement will be described in relation to the third aspect of the invention.

Advantageously the reactor may further comprise a detector adapted to detect separated bands of charged particles in the separation channel, the detector preferably being adapted to image the bands.

A third aspect of the invention makes use of the increased particle convergence achieved by the first aspect in the context of separating substances. As mentioned above, WO-A-2006/070176 discloses an electrophoresis technique for separating charged objects using a shifting field gradient. The objective of the disclosed method is for the analyte molecules to achieve equilibrium between the frictional force as effected by the sieving medium in which the particles are carried and the electric force on them caused by the electric field. The method is effective but can be slow.

The third aspect of the invention therefore provides a method of separating at least two substances, comprising: (I) providing charged particles of the at least two substances; (II) converging the charged particles within a chamber containing a sieving medium using the method of the first aspect of the invention, whereby the width of the group of charged particles is reduced; and then (Ill) performing electrophoresis on the group of charged particles along a separation channel containing a sieving medium, the separation channel being contiguous with the chamber and being aligned with the main axis of the chamber.

By converging the charged particles so that the group occupies a smaller volume (relative to the initially injected "plug") before carrying out separation, the electrophoresis process completes faster than in conventional methods. This is because like particles are already closer together at the start of the separation and thus distinct bands of like particles emerge from the group of particles (i.e. each band can be distinguished from the next) than would otherwise be the case. In essence, a narrower group of mixed particles will inevitably give rise narrower bands of like particles, and those narrow bands will become spaced from one another more quickly than wider bands would (because the trailing edge of a first wide band will overlap the leading edge of the next wide band for a longer period than will be the case with narrower bands). As a result the bands can be distinguished from one another (and hence the substances separated) more quickly than was previously possible. A consequence of this is that the separation channel can be made correspondingly shorter than in conventional techniques, and so the size of the device performing the separation can be reduced. For example this makes it possible to use short, straight, open-ended separation channels as opposed to closed loop channels with infinite length.

It will be appreciated that while the technique disclosed in WO-A-2006/070176 can achieve a degree of band focussing during separation, a major disadvantage of this method is that it treats the focussing mechanism and the band separation as a single combined process. The shifting field shape propagates through the analyte molecules which are induced into motion that eventually leads to them reaching equilibrium positions on the field gradient. The separation and focussing process happen at different rates and there is strong dependence of each on the characteristics of the electric field gradient. However, the dependence is very different for the focussing process as compared with the separation process, and since in WO-A-2006/070176 both have to be achieved simultaneously by the same shifting electric field gradient, there has to be a major trade-off. That is, the electric field gradient has to be a compromise between a gradient that would optimise band convergence vs. a gradient that would optimise band separation.

The presently disclosed concept instead treats the focussing (converging) and the molecular separation processes separately, performing them one after the other. As such, the focussing and the separation processes can be optimised separately in (at least) two different phrases. The optimisation of the focussing process is achieved by the electric field sequence described above, preferably performed multiple times in order to oscillate the particles. The fundamental purpose of the electric field sequence in this phase is to avoid reaching the equilibrium condition. Only after the focussing process reaches a satisfactory level, then the disclosed method moves to a separation step that optimises separation, while potentially preserving a weaker focussing force.

Another benefit of separating the focussing and separation steps in this way relates to limitations on what electric field profiles can be established by available hardware in practical terms. Typically, any particular hardware implementation will be limited in terms of the maximum accumulated voltage it can safely handle and also a maximum voltage limit its power module can apply between any two electrodes. For instance, a device might be provided which has 50 electrodes spaced at 2.5 mm intervals along the main axis which is 12.5 cm long. To generate an electric field of say 500 V/cm at a particular location along the channel, this will require a voltage difference of 125 V between neighbouring electrodes at that location. If this level of electric field were to be applied along the whole length of the device, this would accumulate to 50 x 125 V = 6250 V over the 12.5 cm channel. This is an extremely high voltage which would place problematic insulation demands on the whole high voltage circuit including cables, high voltage connectors and the insulation related to fluids in chip, fluidic short circuits in terms of spills etc. However, in the presently disclosed approach, high electric fields may be beneficially applied across only a small portion of the device initially -for example, a steep focussing gradient might be applied over a first region comprising say a quarter of the 12.5 cm device initially, resulting in an accumulated voltage of only 1500 V. A shallower separation gradient (requiring lower electric fields and hence voltages) can then be applied over a larger section of the device, as described below. This is much more manageable from a hardware perspective.

It will be appreciated that the different particles may start to form into bands of like particles, or even sort themselves completely into bands, during the converging phase of the method (step (II)). This is not essential but since particles of different charge and/or mobility will move differently through the sieving matrix under the influence of an electric field, it is likely that at least some separation will start to take place. During step (II), any bands formed will likely overlap one another at least partially and oscillate in the same region of space (defining the bounds of the "group") which narrows as the method proceeds (as do any bands). This is the same mechanism as already described in relation to the second aspect of the invention (although in the third aspect the substances may or may not react with one another).

As discussed in relation to the first aspect of the invention, the convergence method applied during step (II) exploits the transient state of the force fields as opposed to the equilibrium state. In fact, while optimising the focussing process, the equilibrium state is prohibited for at least n-1 analytes (bands) by the way the electric field is applied. In other words, at most and as a special case, the equilibrium condition may be met by only one band (which is not optimal, but it is not excluded as a possibility). All other bands (and typically all bands) are prevented reaching their equilibrium condition in order to maximise focussing.

The separation channel is contiguous with the chamber in which convergence occurs and is aligned with the main axis of the chamber In this way, the dimension in which focussing has occurred (that parallel to the main axis) is the same as that along which separation will take place. This is necessary in order to achieve the benefits described above: only enhanced localisation along the same direction as that which the bands are to become separated along will assist in the resolving the bands from one another more quickly. It should be appreciated that this does not mean that the separation channel needs to be straight. The oscillation chamber could for instance be located on a tangential offshoot from a curved or circular separation channel. The chamber and the separation channel preferably each form part of one continuous cavity, e.g. formed in the surface of a substrate.

Constructions such as those disclosed in WO-A-2006/070176 or WO-A2012/153108 are suitable. The sieving matrix (which is typically the same in both the chamber and the separation channel) may be for example a fluid or a gel.

In step (II), the method of converging the particles may employ any of the preferred features discussed in relation to the first aspect of the invention above.

The electric field sequence applied in step (II) is preferably repeated a plurality of times in order to achieve a more significant degree of convergence before beginning the separation in step (Ill). The number of cycles required will depend on the mobilities of the particles and on the parameters of the electric field sequence but in preferred examples the sequence may be repeated at least 10 times, more preferably at least 30 times, still preferably at least 50 times. It will be noted that these preferences are towards fewer repeats than might be optimum when using the method to accelerate a reaction On the second aspect of the invention) and this is because in the context of a separation the primary objective is often speed and to begin the separation quickly (so as to minimise the overall running time), rather than to maximise the degree of convergence achieved. In contrast, when performing a reaction it is of primary importance to focus the particles as closely as possible so as to place them in close proximity with one another.

Alternatively or in addition, the electric field sequence may be repeated until the width of the group of charged particles reaches a predetermined threshold. For instance this could be monitored using a suitable detector which is able to image the group of particles in the chamber In preferred examples, the predetermined threshold may be 10mm or less, more preferably 1mm or less, still preferably 0.1mm or less. Again these are values suitable for a small-scale implementation. Larger-scale versions are also possible. Depending on the implementation, width thresholds relating to individual bands may be employed rather than those relating to the group as a whole.

The cycle time of the electric field sequence will likewise be selected to achieve a large number of oscillations within a short time frame. In some preferred examples, the cycle time is between 0.1 and 10 seconds, preferably between 0.1 and 5 seconds, still preferably between 1 and 3 seconds most preferably between 1.5 and 2.5 seconds.

In preferred implementations, the first and second electric field gradients and the duration of each cycle are configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range 0.1 to 10 mm, preferably 0.5 to 5 mm, more preferably 1 to 3 mm. It will be appreciated that these values are greater than the example group or band widths given above, since the group or band will typically be moving between extremes of the oscillation width rather than occupying the whole space.

The electrophoretic separation carried out in step (III) may be performed using any available electrophoresis process, such as the application of a static uniform electric field along the channel. However, more preferably, in step (III), electrophoresis is performed by: III') applying a particle-separation electric field along the separation channel, the particle-separation electric field having a field profile and thereby causing the charged particles to move relative to the sieving medium; and III") varying the applied particle-separation electric field so as to adjust the field profile relative to the separation channel, thereby causing the charged particles to separate into a respective band of each of the at least two substances under the combined influences of an electric force due to the electric field and a hydrodynamic force due to the sieving medium.

Examples of such processes are disclosed in WO-A-2006/070176, the contents of which are incorporated herein by reference.

As noted above, the characteristics of the applied electric field are different in step (II) as compared with step (Ill) in order to optimise it for performing the different functions of converging particles and then separating bands. In particular it is preferred that the field profile of the particle-separation electric field in step (111) has an electric field gradient of which the spatial rate of change of electric field magnitude with distance along the separation channel is less than that of the first and second electric field gradients in step (II). In other words, the electric field gradient in step (111) has a lesser slope than those applied in step (II). This has the result that the equilibrium point for each respective band of like particles (which the particles migrate to during step (110) becomes spaced from the next by a greater distance during separation. As such the separated bands are better resolved from one another. In contrast, a relatively steep gradient is beneficial during the convergence phase (step (10) in order to better focus each band prior to separation. It should be noted that the decrease in slope between steps (II) and (111) does not lead to the bands becoming defocussed (i.e. spreading). Rather, since there is still a slope in step (111), albeit a lesser one, the bands will continue to narrow in step (111) even though at a slower rate than in step (II). In other words, the flatter gradient used during the separation step does not cause de-focussing.

It should be noted that the change in gradient from step (II) to step (111) could be discrete or gradual. In the latter case, the slope of the gradient could for instance be gradually reduced (stepwise or continuously), either before it begins moving along the separation channel or during the movement.

More generally, the two phases (i.e. step (II) and step (III)) may be somewhat merged, without a distinct point at which the converging phase stops and the separation phase begins. For instance, the electric field gradient applied during the final occurrence of step (b2) might be controlled to transition in a substantially continuous manner into the gradient used to perform separation step (Ill).

It should be appreciated that the methods of the second and third aspects of the invention could be combined. That is, the at least two substances injected into the chamber may react with one another during step (II) of the separation method, resulting in at least one reaction product, in the manner described in relation to the second aspect of the invention. In this case the parameters of step (II), such as field gradient slopes, cycle times and/or number of cycles, will need to be chosen so as to balance the desire for a short overall running time of the separation, against the desire to maximise reaction yield. Each of the at least two substances (the reactants) may or may not also remain present (in a reduced amount) in the chamber once the reaction has taken place. Thus the separation performed in step (III) may be on the contents of the chamber after reaction, which will be different to the set of substances originally introduced. As such the substances separated into bands during step (Ill) may include one or more substances different from those injected in step (I), and may consist exclusively of such different substance(s).

The third aspect of the invention further provides an electrophoresis device for separating charged particles, comprising a device for converging charged particles according to any of the first aspect of the invention and a separation channel which is contiguous with the chamber and aligned along the main axis thereof, the electric field generator being further adapted to apply electric fields along the separation channel and the controller being further configured to control the electric field generator to apply a particle-separation electric field along the separation channel.

The electrophoresis device achieves all the benefits discussed above in connection with the method of the third aspect. The device for converging particles may have any of the preferred features discussed in relation to the device of the first aspect of the invention. The controller may also be further programmed to control the electric field generator to perform any of the preferred features of the method of the third aspect mentioned above.

Preferably, the electrophoresis device further comprises an extraction port for extracting a separated band of particles from the separation channel. If a reaction has taken place between the originally-injected substances, it is typically desirable to extract the reaction product as mentioned in connection with the second aspect.

The electrophoresis device may preferably further comprise a detector adapted to detect separated bands of charged particles in the separation channel, the detector preferably being adapted to image the bands.

In the third aspect of the invention, the electric field sequence disclosed in the first aspect of the invention is utilised to achieve strong focussing of the charged particles prior to separation. As discussed above, reducing the width of the group of charged particles increases the speed of the subsequent separation. This can also be achieved, albeit to a lesser extent, by alternative particle-converging techniques.

Hence a fourth aspect of the present invention provides a method of separating at least two substances, comprising: (I) preparing a sample containing the at least two substances to provide charged particles of the at least two substances; (II) collecting the charged particles within a chamber containing a sieving medium by applying a particle-collection electric field gradient along at least a portion of a main axis of the chamber, whereby the charged particles move under the influence of the electric field gradient to form at least one group of charged particles, the front edge of the group moving more slowly than the trailing edge of the group due to a smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is reduced; and then (III) performing electrophoresis on the group of charged particles along a separation channel containing a sieving medium, the separation channel being contiguous with the chamber and being aligned with the main axis of the chamber.

In this method, step (II) does not require a sequence of electric field gradients to be applied. Rather, just one electric field gradient can be applied for as long a duration as desired, with a profile designed to converge particles, before the applied electric field is changed to perform and optimise electrophoresis. This is a simpler approach than that of the third aspect of the invention, requiring less sophisticated control means, and will not achieve as great a focussing effect. However this will still lead to some improvement in separation time, which may be sufficient for many applications.

The particle-collection electric field gradient is preferably static during all or part of step (II) to collect particles across a wide area before separation begins. However in alternative embodiments the collection phase (step (ID) could overlap with the beginning of the separation (step (III)), in which case the particle-collection electric field gradient may be controlled to move along the main axis during all or at least the last part of step (II).

In some embodiments, the particle-collection electric field gradient may include a zero crossing point within the portion of the main axis of the chamber, whereby the at least one group of charged particles moves towards the zero crossing point.

A zero crossing point is a point on an electric field profile where the field magnitude is zero, the field having a positive sign immediately to one side of the point and a negative sign immediately to the other side of the point. Charged particles on both sides will experience a force moving them towards the zero-crossing point, causing a converging effect. Alternatively the particle-collection electric field gradient may be all-positive or all-negative. For instance, the particle-collection electric field gradient could have the same form as the first electric field gradient or the second electric field gradient described in relation to the first aspect.

Examples of such processes are disclosed in WO-A-2006/070176, the contents of which are incorporated herein by reference. It will be appreciated that the particle-separation electric field will have a different profile from that of the particle- collection electric field gradient. Preferably, the field profile of the particle-separation electric field in step (III) has an electric field gradient of which the spatial rate of change of electric field magnitude with distance along the separation channel is less than that of the particle-collection electric field gradient in step (II). In other words, the particle-separation electric field has a shallower slope than that of the particle-collection electric field gradient. This has the result that the equilibrium point for each respective band of like particles (which the particles migrate to during step (III)) becomes spaced from the next by a greater distance during separation. As such the separated bands are better resolved from one another. In contrast, a relative steep gradient is beneficial during the convergence phase (step (II)) in order to better focus each band prior to separation.

It should be noted that the change in gradient from step (II) to step (III) could be discrete or gradual. In the latter case, the slope of the gradient could for instance be gradually reduced (stepwise or continuously), either before it beings moving along the separation channel or during the movement. Thus, the electric field may be changed discretely or continuously from the particle-collection electric field gradient to the particle-separation electric field profile between step (II) and step (III), preferably while being moved from the chamber to the separation channel.

More generally, the two phases (i.e. step (II) and step (III)) may be somewhat merged, without a distinct point at which the converging phase stops and the separation phase begins. For instance, the electric field gradient applied during the final occurrence of step (b2) might be controlled transition in a substantially continuous manner into the gradient used to perform separation step (III).

It will be appreciated that the substances injected into the chamber in step (I) may or may not react with one another, although in this aspect of the invention the field applied in step (II) is not particularly designed to accelerate reaction. Thus the separation performed in step (III) may be on the contents of the chamber after reaction, which will be different to the set of substances originally introduced. As such the substances separated into bands during step (III) may include one or more substances different from those injected in step (I), and may consist exclusively of such different substance(s).

The fourth aspect of the invention also provides an electrophoresis device for separating charged particles of at least two substances, comprising: a chamber into which charged particles are injected in a sieving medium in use, the chamber having a main axis; a separation channel contiguous with the chamber and aligned with the main axis of the chamber; an electric field generator configured to apply electric fields along at least a first portion of the main axis of the chamber and along the separation channel; and a controller configured to control the electric field generator, and programmed to control the electric field generator first to apply a particle-collection electric field gradient along at least a portion of a main axis of the chamber, whereby in use the charged particles move under the influence of the electric field gradient to form at least one group of charged particles, the front edge of the group moving more slowly than the trailing edge of the group due to a smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is reduced; and then to apply a particle-separation electric field along the separation channel whereby electrophoresis is performed on the group of charged particles.

The electrophoresis device achieves all the benefits of the separation method described in the fourth aspect of the invention. Suitable examples of physical constructions for the electrophoresis device are disclosed in WO-A-2006/070176 or WO-A-2012/153108, although the controller will of course be programmed differently in the manner specified above. The controller may be further configured to control the electric field generator to perform the method in accordance with any of the preferred implementations discussed above.

The electric field generator could take any form which enables the application of electric field profiles of the sort described along the main axis of the chamber and separation channel. In preferred examples, the electric field generator comprises an array of electrodes disposed along at least the first portion of the main axis of the chamber, and the separation channel. In use, appropriate voltages will be applied to each electrode so as to give rise to the desired electric field gradients. Electrode arrays are particularly well suited to the generation of the electric field gradients in small scale (e.g. microfluidic) implementations of the device. The array of electrodes could have the same general form as disclosed in WO-A2006/070176 or WO-A-2012/153108, for example.

Preferably, the chamber is provided with at least one input port for injecting a sample into the chamber, the sample comprising at least the charged particles of at least two substances. Desirably, the separation channel is provided with at least one exit port for extraction of groups or bands of charged particles from the chamber. Additional electrodes may be associated with the input and/or exit ports to enable the application of further electric fields for injecting the particles into or drawing them out of the channel.

In preferred examples the device further comprises a detector adapted to detect groups or bands of charged particles in the separation channel, the detector preferably being adapted to image the groups or bands.

BRIEF DESCRIPTION OF DRAWINGS

Examples of methods for converging charged particles, reacting substances and/or separating substances, as well as corresponding devices therefor, will now be described with reference to the accompanying drawings, in which: Figures 1A and 1B schematically show a device for converging charged particles, before and after performing a convergence method, respectively; Figure 2 diagrammatically shows an exemplary chamber and electric field generating means which may form part of the device shown in Figures 1A and 1B; Figures 3A and 3B are plots of electric field, E, vs. distance along the main axis, x, of the chamber, in two respective steps of a first embodiment of a convergence method in accordance with aspects of the present invention; Figures 4A and 4B are plots of electric field, E, vs. distance along the main axis, x, of the chamber, in two respective steps of a second embodiment of a convergence method in accordance with aspects of the present invention; Figure 5 is a flowchart depicting steps performed in embodiments of a convergence method in accordance with aspects of the present invention; Figures 6A, 6B and 6C are plots of electric field, E, vs. distance along the main axis, x, of the chamber, at three respective instances in time during a first step of 5 a third embodiment of a convergence method in accordance with aspects of the present invention; Figures 7A, 7B and 70 are plots of electric field, E, vs. distance along the main axis, x, of the chamber, at three respective instances in time during a second step of the third embodiment of a convergence method in accordance with aspects of the present invention; Figures 8, 9, 10, 11 and 12 are plots showing results from a simulation modelling the behaving of exemplary charged particles which are subject to an embodiment of a convergence method in accordance with aspects of the present invention; Figure 13 is a flowchart depicting steps performed in another embodiment of a convergence method in accordance with aspects of the present invention; Figures 14A and 14B are plots of electric field, E, vs. distance along the main axis, x, of the chamber, Figure 14A showing exemplary first electric field gradients which may be applied during a first step of an electric field sequence, and Figure 14B showing exemplary second electric field gradients which may be applied during a second step of an electric field sequence, each plot showing two respective electric field gradients suitable for two respective stages of the method set out in Figure 13; Figures 15, 16 and 17 are plots showing results from a simulation modelling the behaving of exemplary charged particles which are subject to another 25 embodiment of a convergence method in accordance with aspects of the present invention; Figure 18 is a flowchart depicting steps performed in a further embodiment of a convergence method in accordance with aspects of the present invention; Figure 19 is a plot of electric field, E, vs. distance along the main axis, x, of the chamber, depicting an exemplary static electric field profile, an exemplary first electric field gradient and an exemplary second electric field gradient all suitable for use in the Figure 18 embodiment; Figures 20 and 21 each schematically depict exemplary layouts of device components suitable for use in embodiments of the invention; Figure 22 is a flowchart depicting steps performed in an embodiment of a separation method in accordance with aspects of the present invention; Figures 23 and 24 each schematically depict exemplary layouts of device components suitable for use in embodiments of the Figure 22 method; Figures 25A, 25B and 250 are plots of electric field, E, vs. distance along the main axis, x, of the chamber, during respective steps of an exemplary implementation of the method of Figure 22; Figures 26A, 26B, 27A and 278 are plots showing results from simulations modelling the behaviour of particles which are subject to exemplary separation methods (i) according to a comparative example (dashed lines), and (ii) according to an embodiment of the present invention (solid lines); Figures 28A, 283 and 280 are plots of electric field, E, vs. distance along the main axis, x, of the chamber, during respective steps of another embodiment of a separation method in accordance with aspects of the present invention; Figure 29 is a plot of electric field, E, vs. distance along the main axis, x, of the chamber, in one step of the first embodiment of a convergence method, showing the effect on particle of mixed charge sign; Figures 30A and 303 are plots of electric field, E, vs. distance along the main axis, x, of the chamber, in two respective steps of another embodiment of a convergence method in accordance with aspects of the present invention; Figures 31A and 318 are plots of electric field, E, vs. distance along the main axis, x, of the chamber, in two respective steps of a further embodiment of a convergence method in accordance with aspects of the present invention; Figures 32A and 32B are plots showing the results of a test for Covid-19 performed using a method in accordance with an embodiment of the invention utilising an oscillation stage, Figure 32A showing a negative result and Figure 32B showing a positive result, Figure 32B0) showing an enlarged detail of the plot of Figure 32B; Figure 33 is a plot showing the results of a (positive) test for Covid-19 performed using a method in accordance with an embodiment of the invention utilising a static collection stage; Figures 34A and 34B are plots showing the results of a test for Lassa virus performed using a method in accordance with an embodiment of the invention utilising an oscillation stage, Figure 34A showing a negative result and Figure 34B showing a positive result; and Figure 35 is a plot showing the results of a (positive) test for I nfluenza-B performed using a method in accordance with an embodiment of the invention utilising an oscillation stage.

DETAILED DESCRIPTION

The description below will focus on small-scale, e.g. microfluidic, implementations of the invention. As will be discussed in more detail below, the methods herein described are of particular use in the context of biological assays. In particular, the methods herein disclosed are particularly useful for the determination of binding between two substances (e.g. protein-protein interactions or protein-DNA interactions) and/or the characterisation of said substances. Accordingly, the present invention may be used as an alternative method to a number of commonly used biological assays in this area, for example, immunoprecipitation assays, pull-down assays, mobility shift assays, ELISA assays and Western Blots. The advantages of the present invention are numerous in this context, including but not limited to, a reduction in assay run-time and enhanced interaction between the two substances. In the instance where the substance is a protein, a further advantage is that the method allows for binding of the proteins in their non-denatured state to be assessed and characterised. This is in contrast to other commonly used methods in this field, such as Western Blot and SDS-PAGE that require denaturation of the protein. Classical denaturing electrophoresis does not offer the advantages provided by the present invention. Further, the speed and accuracy of the methods of the invention may allow for the detection of fleeting interactions between molecules, which may be useful for identifying receptor-ligand pairs in heterogenous mixtures of analytes. The methods may also be useful for detecting low affinity interactions between proteins. However, as noted above, the same principles can be applied on a much larger scale if desired. In such cases, implementations of the invention could be used in applications such as producing chemical or bio-chemical substances. For example, reaction products could be produced and then separated from the enzyme and substrate using the methods of the invention.

A further example of the potential utility of the present methods is the purification of analytes, such as proteins, using binding molecules, such as antibodies. Akin to a typical immunoprecipitafion (or pulldown) assay, a sample solution containing analyte may be incubated with antibodies specific to analytes of interest and subjected to the methods of the invention. This will concentrate the analyte at a certain position in the field. The separation phase may then be used to separate the analyte from the antibody and then extract it. The non-denaturing conditions of the methods will assist in producing native folded protein, which may be used as needed. Of course, analytes may be tagged using methods known to the person of skill in the art, and the binding molecules used to separate the analytes through interaction with the tag. Thus, the present invention may provide an improved alternative method of identifying protein-protein interactions over the methods set out for example in Arifuzzaman et al (2006) Large-scale identification of protein-protein interaction of Escherichia coli K-12, Genome Res. 16(5): 68691 (incorporated herein by reference).

The present invention may be used in conjunction with other product preparation methods and in synthetic biology methods to accelerate interactions between key molecules of interest at key times in the production methods. For example, the methods of the invention may be of use alongside cell free protein synthesis methods, such as those described in Khambhati et al (2019) Exploring the Potential of Cell-Free Protein Synthesis for Extending the Abilities of Biological Systems, Frontiers in Bioengineering and Biotechnology 7 (incorporated herein by reference), and/or in Gregorio et al (2019) A User's Guide to Cell-Free Protein Synthesis, 2(1): 24 (incorporated herein by reference). The methods of the present invention could play a critical role in accelerating synthetic biology workflows at several levels in the whole algorithm from DNA synthesis to protein expression and purification, as well as identifying and characterising protein-protein interactions, all at high speed and accuracy.

Figures 1A and 1B schematically illustrate a device 1 suitable for use in embodiments of the invention. The device 1 comprises a chamber 2, defining a volume in which charged particles 10 are placed in use. For instance, the chamber 2 may be formed as a cavity in the surface of a substrate (or as part of a larger cavity). The charged particles 10 may be in a fluid 8, such as a solution having a certain pH for controlling the apparent charge of the particles 10, or a sieving matrix. In some cases the fluid 8 could perform multiple functions, such as charge control and sieving. The fluid 8 could be replaced by a gel or other sieving matrix. In other cases, the fluid 8 may be absent, e.g. in gaseous implementations. It should be appreciated that there is no need for a sieving matrix to be present in order for particle convergence to be carried out in the manner about to be described, since it does not require the particles 10 to experience hydrodynamic forces (although they will of course do so if a fluid or other sieving matrix is present).

The chamber 2 has a main axis, x, along which an electric field can be established by an electric field generator 4. It should be noted that while in Figures 1A and 1B the main axis is shown as parallel to an elongate direction of the chamber, this is not essential and indeed the chamber 2 need not be elongate. In the present embodiment, the electric field generator 4 comprises an array of electrodes spaced along the main axis, x. The array comprises at least two electrodes, preferably more than two (and most preferably significantly more). It will be appreciated that the geometry of the chamber may also have an effect on the electric field profile -for example if the cross-sectional area of the chamber is not constant along the main axis, this may contribute to establishing the desired electric field shape. Examples of suitable structures will be given below.

The electrodes will be connected to an appropriate voltage source (not shown) via a controller 6. The controller is configured to control the voltage level applied to each electrode in such a way as to apply desired electric field profiles to the interior of the chamber 2, along the main axis, x. That is, the electric field generator 4 is able to control both the magnitude and the sign of the electric field at each point within the chamber 2 along the main axis, x (at least along a portion of the main axis, x). Examples of suitable electric field generators, including electrode array constructions and controllers, are disclosed in WO-A-2006/070176 and WO-A2012/153108, both herein incorporated by reference. Parts of the separation channels disclosed in those documents are also suitable for use as the chamber 2 in the present method.

The particles 10 to be converged will be injected into the chamber 2, e.g. by pipette or from an input port (not shown) equipped with one or more electrodes for applying an appropriate force to move the particles into the chamber In the present example, all of the charged particles 10 have the same apparent charge sign (e.g. negative) although they may be of different types and therefore may have different negative charge levels. The particles 10 form a group 12 which will initially be spread over a wide volume of the chamber along the main axis, x, as shown in Figure 1A. After performance of the convergence method, to be described below, the width of the group of particles 12 is reduced, as shown in Figure 1B. That is, the group 12 occupies a smaller distance along the main axis, x, than previously. The process thus spatially focusses the group of particles into a smaller volume. It should be noted that the focussed group of particles 12 may not comprise all the particles 10 originally injected into the chamber 2. There may be some outliers which are not captured by the process and which remain unfocussed. However the method is preferably configured such that a large majority of the particles 10 will be captured into a group and converged Methods for enhancing this are discussed below.

Figure 2 again shows the chamber 2 and identifies a first portion Pi of the main axis within which an electric field sequence will be applied during the converging process. This first portion Pi could extend over the full size of the chamber 2, but more typically will be a sub-portion thereof. The first portion Pi is that area along the main axis in which an electric field gradient is applied at some point during the electric field sequence. A first region R1 of the main axis is also defined, which is substantially fixed relative to the main axis. Particles located within the first region Ri of an appropriate charge sign will be converged by the process. In the first embodiment to be described with reference to Figures 3A and 3B, the extent of the first region R1 is equal to that of the first portion P1 but this need not be the case.

The converging method involves applying an electric field sequence along the first portion P1 of the main axis, x. A first example of a suitable electric field sequence, used in a first embodiment, will be described with reference to Figures 3A and 3B. In a first step S111 of the sequence, a first electric field gradient 21 is applied via the electric field generator 4, across the whole of first region Ri. The first electric field gradient 21 comprises positive values of electric field E at all points in the first region R1 and the magnitude of the electric field decreases along the main axis a first direction (here, in the +x direction). The effect of the first electric field gradient 21 on the charged particles will be illustrated by reference to just two exemplary particles 10a and 10b located at opposite edges of particle group 12. The group 12 of charged particles moves under the influence of the first electric field gradient 21 in the first direction along the main axis. However, the front edge of the group 12 (represented by particle 10b) moves more slowly than the trailing edge of the group (represented by particle 10a) due to the smaller electric field magnitude acting on the particles at the front edge. This is a result of the shape of the electric field profile, which decreases in value in the +x direction so particles further along the main axis will experience a lower force than those behind them (as denoted by the differently sized arrows representing the electric force on each particle in the Figure). As a result, the trailing edge "catches up" with the leading edge to an extent, and the width of the group of charged particles 12 is reduced.

In a second step S112 of the electric field sequence, a second electric field gradient 22 is applied across the whole of first region Ri On place of the first electric field gradient). This is shown in Figure 3B. The second electric field gradient 22 comprises negative values of electric field E at all points in the first region R1 and the magnitude of the electric field decreases along the main axis in a second direction, opposite to the first direction (here, in the -x direction). As a result, the particles 10a and 10b reverse and now move in the second direction. Once again, the leading edge of the group 12 (now represented by particle 10a) experiences a lower force than does the trailing edge of the group 12 (now represented by particle 10b). As such, particle 10b moves faster than particle 10a and once again the width of the group narrows.

Essentially, what happens is that when the applied electric field gradient pushes a group 'to the right' (Figure 3A), the left side of the group 12 sees a higher E field than the right side of the group. So the left side moves faster than the right, resulting in focussing. Then when the field gradient reverses (Figure 3B), it pushes the particles towards the left, and the same thing happens in reverse (still focussing). This mechanism does not require hydrodynamic forces or sieving. The steeper the slope of the electric field gradients 21, 22, the greater the focussing effect. The difference in electric force between the front (leading edge) and the rear (trailing edge) of the band is proportional to: dE Eleading -Etrailing (1) where E represents electric field. Therefore, the focusing rate is proportional to the gradient of the electric field.

The magnitude of the applied electric field will also have an impact, greater field magnitudes causing the particles to stay further from their equilibrium conditions, which also enhances the focussing effect. To understand the mechanism behind this, we need to consider the more complete set of forces acting on each particle: Floading = q*Eleading -f.Vloading (2) Ftrailing= q.Etrailing-f. Vtra iling (3) 1A/here q is the charge on the particle, f is the friction coefficient and v is the speed. Here, Heading is the force on a representative particle at the leading edge of the group, and Ftrailing is the force on a representative particle at the trailing edge. This model assumes electrophoretic conditions, where molecular mobility (drag) is significant.

Therefore the difference Fleading-Ftrailing is also proportional to f(Vleading-Vtra fling). However, the latter term is a function of the velocities of the edges and therefore the equation is non-linear. In fact for an electric field profile that moves with velocity k, equations (2) and (3) become zero when the particle is at the equilibrium position on the field profile (where friction equals electric force). In contrast, the force is maximum far from the equilibrium where the term q.E dominates over f.v. This causes a non-linear asymptotic relationship between the distance from the equilibrium position and the focusing force (which is effectively Fieading -Ftrailing)* This difference is approximately q(Efront-Eback) when the particles are far from the equilibrium point, and zero when at the equilibrium.

For a static field the equilibrium point is at E = 0. Therefore for a static field, a sufficiently non-zero average electric field magnitude experienced by a band is the condition to be far away from equilibrium and thus to accelerate focusing.

Similar arguments can be said about a non-zero field speed: the further away the particles are from E = 0, the greater the focusing force the band will experience. In this case it should be noted that if the average particle position is sufficiently far away from the equilibrium point for the particular field speed, and in particular on the higher side of the equilibrium point, then the speed of the field can be practically considered zero. For example, if the equilibrium field is 100 V/cm for a field that moves with 0.5 mm/s but the current average position of the band is at 500 V/cm, for as long as the band remains this far from equilibrium, the speed of the field is irrelevant as the focusing rate is no different for field speed 0.5 mm/s and 0 mm/s. It will be appreciated that while, in the example given above, the first electric field gradient 21 is positive and the second electric field gradient 22 is negative, the opposite could be true. Also, the slope of the two gradients 21, 22 could be reversed (i.e. the first direction could be the -x direction and the second direction the +x direction) in which case the process would be effective on particles of the opposite charge sign -this will be discussed further in relation to Figures 29 to 31.

The steps S111 and 5112 together form one cycle of the electric field sequence.

In some embodiments, this cycle could be performed a single time. However, more preferably, the sequence will be repeated a plurality of times -potentially lOs or even 100s of times. Repeating the sequence causes the group of particles 12 to oscillate along the main axis of the chamber about an average position, one oscillation for each repeat cycle performed. The more repeat cycles are performed (and hence the greater the number of oscillations), the greater the degree of convergence that will ultimately be achieved. As discussed below, the number of repeats performed in practice will depend on the desired outcome which may vary by application.

In the example shown, the slope of the first electric field gradient 21 (i.e. the rate of change of electric field E over a distance x) is the same as that of the second electric field gradient 22. This is preferred in order to achieve a more symmetric outcome (the degree of narrowing and also the amount of movement of the group 12 being the same in both steps), but this is not essential. For the same reason, the maximum and minimum magnitudes of the field gradients 21,22 are preferably substantially the same (as shown). It should also be noted that while the example field gradients 21, 22 are rectilinear, this is also not essential and in other cases the field gradients could be curved (e.g. parabolic or exponential). Preferably the two field gradients 21, 22 will be of the same shape as one another. It should also be noted that whether an electric field profile exists outside the first region R1 (and what characteristics it might have if so) in step S111 and/or S112 is unimportant and will not affect the working of the described method. Thus while Figures 3A and 3B show zero electric field outside the first region Ri, this might or might not be the case in practice.

A second embodiment in which electric fields are also applied outside the first region R1 will now be described with reference to Figures 4A and 4B, which show the electric field profiles applied along the main axis x of a chamber 2 in first (S111) and second (S112) steps of the electric field sequence, respectively. In this case, a composite electric field gradient 29 is used, which comprises both a first part constituting a first electric field gradient 21 and a second part constituting a second electric field gradient 22. As in the previous embodiment, the first electric field gradient 21 is all-positive and decreases in the +x direction while the second electric field gradient 22 is all-negative and decreases in the -x direction. In this example, the composite electric field gradient 29 is continuous and the two parts meet one another at a zero crossing point 28, forming together a linear field gradient which crosses zero. However in other cases the composite electric field gradient 29 could be discontinuous, e.g. having an area of zero field magnitude between the two parts 21, 22.

The same composite electric field gradient 29 is applied to the chamber in both first and second steps of the sequence, but at different positions along the main axis, x. Thus, in a first step S111 of the sequence, as shown in Figure 4A, the composite electric field gradient 29 is applied such that its first part, corresponding to first electric field gradient 21, coincides with the first region Ri. In a second step 5112 of the sequence, the composite electric field gradient 29 is applied at a position shifted along the main axis x such that its second part, corresponding to second electric field gradient 22, coincides with the first region Ri. Here, the position of the particle group is represented by lines 12 and it will be seen that this narrows as the method proceeds in the same manner as previously described.

Figure 5 is a flowchart showing steps in an exemplary convergence method employing the principles already described. Steps shown in dashed lines are optional. In first step S111, a first electric field gradient 21 is applied in the first region R1, which may for example take the form shown in Figure 3A or 4A. The field is then switched to the second electric field gradient 22, such as that shown in Figure 3B or 4B, in step 5112. This switch may be discrete, with no in-between transition from one gradient to the other, and may preferably be instantaneous (i.e. substantially zero time elapses between steps S111 and 5112). In other words, the applied field in the first region R1 "jumps" from the first electric field gradient to the second electric field gradient and vice versa. However, in other implementations there may be one or more other steps performed between steps S111 and 5112, such as a transition step S111' during which the first electric field gradient is changed to the second electric field gradient. For example, if the two electric field gradients are implemented as parts of a composite electric field gradient 29 (as in Figures 4A and 4B), this could involve translating the composite electric field gradient 29 along the main axis, x, during the transition step S111'. There may likewise be a transition step S112' after step S112, in which the second electric field gradient is changed to the first electric field gradient, ready for the sequence to be repeated. Again this could involve translating the composite electric field gradient 29 back to its original position. As such, in an embodiment of the method, the process could involve repeatedly moving a composite electric field gradient 29 back and forth along the main axis, x.

If transition steps S111' and/or 5112' are employed, each is preferably shorter in duration than either of steps S111 and S112. The transition steps are ideally as short as possible, so that even particles of high mobility do not have sufficient time to reach locations on the electric field profile with low or zero field magnitude (which might exist in the first region during the transition), which could detract from the focussing effect. By ensuring the particles experience a relative high electric field magnitude throughout as much of the process as possible enhances the convergence achieved. As will be seen by the simulations below, the presently disclosed method has the result that the particles "see" a non-zero average electric field magnitude (averaged over the duration of the electric field sequence). This is true whether or not transition steps S111' and/or 5112' are involved. In preferred examples, the duration of each of transition steps S111' and S112' (if provided) at least 10 times shorter, more preferably at least 50 times shorter, most preferably at least 100 times shorter than the duration of steps S111 and S112.

Steps S111 and S112 therefore preferably form together a major proportion of the electric field sequence. In some preferred examples, steps S111 and S112 together occupy at least 80%, preferably at least 90%, more preferably at least 99%, of the duration of each cycle. The overall duration of the sequence (i.e. the cycle repeat time) has a significant impact on the results achieved and will be chosen in dependence on the application and the desired outcome. Preferably the duration of each of steps 5111 and 5112 is substantially the same. The longer the duration of the sequence, the further the group of particles 12 will have time to travel in each step before reversing. As such, a longer cycle time leads to greater movement of the particles and a wider oscillation of the group 12 back and forth along the main axis, x. Depending on the application, this may or may not be desirable as discussed below.

In general terms, it has been found that electric field sequences having a total duration (cycle time) of between 0.1 and 100 seconds, preferably between 0.1 and 10 seconds, more preferably between 0.1 and 5 seconds, most preferably between 1 and 3 seconds, produce good results. For instance, in an example (detailed further below), the duration of each of steps S111 and S112 was 1.15 seconds and the transition between them was instantaneous, making a cycle time of 2.3 seconds corresponding to one oscillation of the group 12.

Repeating the electric field sequence causes the particles to oscillate along the main axis about some average position in the chamber. If there are multiple different particle types, they will form into bands which all oscillate about the same average position but with different amplitudes of oscillation depending on their molecular mobility or mass. The resulting width of oscillation exhibited by each group (or band) of particles will depend on the mobility of the particles as well as on the characteristics of the applied electric field sequence. Each band will focus as it oscillates.

As noted above, desirable oscillation widths will vary according to the application but in some general cases the first and second electric field gradients and the duration of each cycle may being configured such that the width of the oscillation being in the range 0.1 to 10 mm, preferably 0.5 to 5 mm, more preferably 1 to 3 mm. The width of the oscillation may preferably be caused to narrow as the method progresses, e.g. by reducing the cycle time as discussed further below.

The electric field sequence (consisting, in this example, of steps S111 and 5112, and optionally steps S111' and S112') will be performed one or more times until in step S113 it is determined that sufficient convergence has been achieved. This may mean that a certain number of repeats have been performed (e.g. which has been determined by modelling to provide sufficient convergence) or could involve monitoring the width of the group 12 using a detector (e.g. which may image the contents of the chamber 2) and identifying that the width has reached a predetermined threshold. At that stage the electric field sequence may be stopped and the convergence process ends (S114) although there may be onward steps, such as performing a separation on the contents of the chamber and/or extracting one or more substances as discussed below. In some embodiments, the achievement of a desired degree of convergence may not trigger the end of the repeating of the electric field sequence, since it may be desirable to continue oscillations for a period of time (or indefinitely) to maintain the converged nature of the particle group and avoid it spreading as a result of diffusion.

As alluded to above, it is desirable to keep the average electric field magnitude experienced by the particles high during the electric field sequence, so as to maintain higher electric forces on each particle and a greater focussing effect. An embodiment in which this is enhanced by a further refinement of the electric field sequence will now be described with reference to Figures 6 and 7. Here, the shapes of the first and second electric field gradients 21, 22 applied during the first and second steps S111 and 5112 respectively are the same as described in connection with the first embodiment (Figures 3A and 3B). In the first embodiment, as the group of particles 10 moves in the +x direction during step S111, it will inevitably move towards an area of lower electric field magnitude (due to the slope of the gradient 21), thereby lowering the overall average field seen by the particles. Likewise, the same will occur in step 5112. The present embodiment redresses this by also translating the first and second electric field gradients 21, 22 during each of steps S111 and 5112 respectively.

Figures 6A, 6B and 6C show this translation of the first electric field gradient 21 during step S111. At a first instance in time, t1, the electric field gradient 21 is located at a first position as shown in Figure 6A. At the position of the group of particles 12, the electric field has a magnitude of around (+)Ei (the exact value being greater at the trailing edge of the group than at the leading edge). At a subsequent time point t2 (still within step S111), shown in Figure 6B, the particles 12 have moved in the +x direction but so has the electric field gradient 21 (under the control of controller 6). As such, the particles still experience an electric field magnitude close to El. The same is true at a still later time point t3, as shown in Figure 60. In this way, the average electric field magnitude seen by the particles remains higher throughout step 5111 than it would do if the electric field gradient 21 were static during the step.

Figures 7A, 7B and 7C show the same principle applied in the second step S112. Now, the second electric field gradient 22 is translated in the -x direction as the step progresses, maintaining the magnitude of the electric field seen by the particles 10 around a value (-)E2 which is greater than would be the case if the

field gradient were static.

It will be appreciated that while the concept has been described in such a way that the electric field experienced by the particles 10 remains substantially constant during each respective step (the field gradients 21, 22 being translated at approximately the same speed as that at which the particles are moving), this will not tend to be the case in practice. However, there will still be some increase in the average field seen relative to the static scenario. It will also be noted that while this improvement has been described as a variant of the first embodiment, it can equally be applied to the second embodiment and all embodiments described below.

The results of a computer simulation performed to demonstrate the effects achieved by the presently disclosed method will now be discussed with reference to Figures 8 to 12. Here, the behaviour of a mixed set of six particle types, each of different mobility but the same charge sign (i.e. all positive or all negative) was modelled in the presence of an electric field sequence according to an embodiment of the invention As will be seen, each particle type separated into a distinct group, or band, of like particles having the same mobility during the process. The mobilities of each group are set out in table 1 below:

63 Table 1

Group Mobility 1 2.48 x10^-8 (m/s)/(V/m) 2 2.43 x10^-8 (m/s)/(V/m) 3 0.50 x10^-8 (m/s)/(V/m) 4 0.51 x10^-8 (m/s)/(V/m) 3.80 x10^-8 (m/s)/(V/m) 6 4.10 x10^-8 (m/s)/(V/m) The first and second electric field gradients 21, 22 applied sequentially were formed of a composite electric filed gradient of similar shape to that shown in Figures 4A and 4B. No transition steps were employed so the switch between gradients was substantially instantaneous. In this case, the first cycle formed an optional "collection" stage (explained below with reference to Figures 13 and 14) in which the applied field gradients were relatively shallow. The composite field gradient extended from +800 V/cm to -800 V/cm over 48 nodes (121.92 mm), where each node corresponds to one electrode of the array. Each step had a duration of 1.25 seconds and during this the field gradient was moved in the relevant direction at a low speed of 0.015 cm/s (as described with reference to Figures 6 and 7 above), which is optional. After this "collection" cycle was complete, the field parameters were changed to perform a "convergence" phase.

Here the applied field gradients were steeper, being formed by a composite field gradient extending from +800 V/cm to -800 V/cm over 7 nodes (17.78 mm). The first electric field gradient (step S111) was applied for 1.25 seconds and then the second electric field gradient (step S112) was applied for 1.25 seconds in each repeat, making a cycle time of 2.5 seconds. During each step the respective electric field gradient was translated along the main axis at 0.03 cm/s in the manner described with reference to Figures 6 and 7 above (which is optional). The in the convergence phase, the electric field sequence was repeated 30 times in total to achieve the desired degree of focussing. This was followed by a separation step, starting at about t=77.5s. For the time being, only the convergence phase will be discussed and the separation phase will be returned to below.

Figures 8, 9 and 10 are plots showing three different representations of the trajectories of the six bands of particles during the run. It is noted that some of the six paths overlap one another, at least in places, hence the reason fewer than six traces are visible in some cases. Figure 8 is a space vs time representation showing on the vertical axis the position of the band along the main axis (i.e. the x-axis position), and time on the horizontal axis. The oscillating motion of the bands about an average position approximately 2 cm along the main axis can be clearly seen. The six trajectories initially have different average positions, but eventually they sync with the switching field gradient and oscillate around a common centre. It may also be seen that the overlapping of the various bands increases as the oscillation progresses, showing that the overall width of the group of particles (i.e. the set of bands) also decreased over time.

Figure 9 shows time vs the electric field position magnitude seen by each group (i.e. the field that each group experiences at each point in time -this is not the same as the applied field as each group will take time to "catch up" with the applied field -note also that the sign of the electric field is not factored in). It can be seen that the average field magnitude is around 245 V/cm. Significantly, this value is not zero which indicates the strong focussing effect of the electric field sequence.

Figure 10 shows position along the main axis, x, on the horizontal axis vs electric field on the vertical. Note since the horizontal axis is position x and not time, the lines can go "backwards" as well as forward. The plot shows that all six trajectories follow larger closed loops initially, gradually converging to tighter loops with average electric field values of around 245 V/cm (magnitude). The "drift" shown by the trajectories may be caused by the initial position of the injected "plug" of particles not being aligned with the centre of the first region across which the converging electric field sequence is applied -as the method proceeds, the particles move towards that centre. Figure 11 shows a zoomed in version of part of one of the trajectories from Figure 10. Here one can clearly see the closed loop trajectory of one group of particles. The first part of the trace (arrow OD shows the particles moving in the +x direction during step S111 and experiencing a decrease in the field strength seen as they do so from around +260 V/cm to around +235 V/cm (due to the slope of the gradient, as explained above). Path (ii) shows the field switching to step S112, at which point the particles stop moving and suddenly experience a high field in the opposite direction. Path (iii) shows the particles moving in the -x direction during step S112 and again the field magnitude decreases from about -260 V/cm to around -235 V/cm. The field direction then switches again as shown by line (iv). For the vast majority of the cycle time, the particles experience either part (i) or part (iii) of the loop shown. Parts (ii) and (iv) have a zero or near-zero duration. As such the average field magnitude "seen" by the particles over the full cycle is largely determined by that experienced during parts (i) and (iii), hence approximately 245 V/cm.

Figure 12 is a plot illustrating the relationship between cycle time (i.e. total duration of one instance of the electric field sequence) and the oscillation width, as well as its dependence on particle mobility. The exemplary particles used in the simulation here have respective mobilities as follows: High Mobility ("Fast" trace) = 2.5 x10-3 (m/s)/(V/m), Medium Mobility ("Medium" trace) = 1.25 x10-4 (m/s)/(V/m) and Low Mobility ("Slow" trace) = 2.5 x10-5 (m/s)/(V/m). It will be seen that to achieve an oscillation width between 0.5mm and 5mm (useful in many applications), the oscillation period required varies between 0.2s to over 100s. The longer the cycle time, the greater the width of the oscillation induced by the electric field sequence. Particles of higher mobility have a greater oscillation width than those with lower mobility. This is because the higher mobility particles are able to move a greater distance during each step of the cycle. Typically, when oscillating it is often desirable to minimise the oscillation width, as an oscillation of small width reduces the convergence time and also reduces the size of the device.

The particle oscillation may tend to narrow over time for the same field gradients. However as it narrows, the particles increasingly stay in regions of lower electric field magnitude, which is not desirable. To reduce this effect, in some preferred embodiments, the oscillation width may be narrowed as the method proceeds, e.g. by reducing the cycle time. This helps to reduce or eliminate this 'dropping' of the particles to lower field values. In essence, this has a similar effect to the 'translation' of the first and second field gradients 21, 22 performed in preferred embodiments as described in relation to Figures 6 and 7 above. Both of these approaches could be applied in combination if desired.

As mentioned at the outset, the initial sample (or "plug") of charged particles can be spread over a wide volume of the chamber 2 -it can even be centimetres in length. In such a case, if a method of the sort described above is applied over a first region R1 of only a few millimetres in width, the injected plug could be "cropped" leaving a significant proportion of the charged particles outside of the applied electric field sequence, which will not form part of the group of particles that converges. To address this, embodiments of the invention can include a particle collection stage before the convergence stage begins.

A first example of such an embodiment will be described with reference to Figures 13 and 14. In this case, the particle collection stage and the subsequent particle convergence stage each involve the application of an electric field sequence of the sort already described, but with different characteristics. The two stages could be discrete or one could merge into the other by gradually changing the parameters of the sequence. Figure 13 is a flowchart showing an example technique. Here, steps S101, S102 and S103 form the particle collection stage while steps S111, S112 and 5113 form the convergence stage. In the particle collection stage (Si), first and second electric field gradients are applied across a first region Ri(S1) which is relatively wide. This is shown in Figures 14A and 14B.

The first electric field gradient 21' applied in step S101 is relatively shallow in slope as compared with that which will be applied during the later convergence stage (i.e. the spatial rate of change of E with distance x is lower). Figure 14A shows this field gradient 21' and a steeper first electric field gradient 21" that may be applied during later step S111 in the convergence phase (S2), across a narrower first region Ri (52) for comparison. Likewise, the second electric field gradient 22' applied in step S102 is relatively shallow and wide as compared with that (22") applied in later step S112. Thus, steps 5101 and 5102 are configured to collect particles over a wide area with relatively weak focussing, bringing them together into what will be the first region Ri(S2) for the particle convergence phase. A set of steeper field gradients is then applied in the manner already described to achieve strong focussing. The duration of steps 5101 and 5102 may be longer than those of steps S111 and 5112 to allow more time for the particles across the wide area to move together.

As Figure 13 shows, the particle collection steps 5101, 5102 could be performed a single time or could be repeated a number of times until a sufficient proportion of the injected particles have been collected together (step 5103). Once the method moves on to the convergence stage, steps S111 and S112 could also be repeated a plurality of times in the aforementioned manner. However, it is also possible to move continuously from a stage optimised for particle collection On which the gradients are relatively wide and shallow) towards a stage optimised for particle collection (using relatively narrow and steep gradients 21, 22). Indeed, this gradual change could continue throughout the whole convergence process if desired, e.g. with the slope and/or lateral extent of the gradients 21, 22 being changed between every N repeats of the sequence, where N is an integer greater than or equal to 1.

In another variant, the method of Figure 13 could be performed using only one of steps S101 and S102 (and not the other) before proceeding to step S111. That is, a single application of a relatively wide and shallow electric field gradient (e.g. 21' or 22') could be performed to achieve some particle collection before proceeding to converge the group of particles.

Figures 15, 16 and 17 are plots from another computer simulation of a method employing these principles and extending them further. In this embodiment, the particle collection stage S1 is followed by two particle convergence stages Sz and S3, which use increasingly narrow and steep gradients 21, 22 in each stage.

Convergence stage S3 is followed by an optional separation phase S4 which will be discussed in more detail with reference to later embodiments.

In this case, the simulation used two particle types, of mobilities 0.5x10-8 (m/s)/(V/m) and 0.51x10-8 (m/s)/(V/m) respectively, which are of similar magnitudes to typical bio-particles such as antigen molecules. For clarity, only a single trace (i.e. one of the bands) is shown in each plot since it was found that these were near identical for each band.

Figure 15 shows the electric field seen by the band as a function of time and as noted above this will not be identical to the applied field values since the particles take a finite time to move within the chamber and thus experience the field applied at any particular location. Figure 16 shows the position of the band along the main axis, x, vs. time and it should be noted that here the time axis extends further than that in Figure 15. Figure 17 shows the electric field seen by the band as a function of distance along the main axis, x.

Table 2 below shows the parameters of the electric field sequence applied in each of stages 51 to S3, as well as that of the electric field profile used in the separation stage 54. It should be noted that, in practice, the first and second electric field gradients applied in this scenario were established as part of a composite field gradient, using the same approach as described above in relation to Figures 4A and 4B. Hence in the table, the parameters given for the first electric field gradient correspond to the positive section of the composite gradient, and those for the second electric field gradient to the negative section. The parameter "field oscillation width" (or "jump") refers to the distance along the x-axis by which the composite field is moved back and forth (or jumps) to achieve the desired oscillation between the first and second electric field gradients in the first region.

Table 2

Stage Si Stage Sz Stage Sz Stage Sia No. of cycles 4 15 15 n/a First electric field gradient min/max and width (Step S111) 0 to 800 V/cm over 18.75 mm 0 to 800 V/cm over 8.75 mm 0 to 800 V/cm over 8.75 mm [translating field grad. of -800 to +800 V/cm over 62.5 mm] Second electric field gradient min/max and width (Step S112) -800 to 0 V/cm over 18.75 mm -800 to 0 V/cm over 8.75 mm -800 to 0 V/cm over 8.75 mm Step duration 8.33 s 1.25 s 0.125 s [300s] Field oscillation width (jump) 15 mm 5 mm 2.5 mm n/a Hence, during the collection stage Si, the first and second electric field gradients 21, 22 are each applied over a first region of 18.75 mm width and the sequence is repeated with these parameters four times, completing at time t, (approximately t = 66.64 s). During this stage Si, as seen best in Figures 16 and 17, particles are collected from across the broad first region and caused to oscillate with a reasonably wide amplitude. Between time ta and time tb, in stage S2, the slope of the first and second electric field gradients 21, 22 is greater (relative to that in stage Si), which here is achieved by narrowing the first region to 8.75 mm in width and maintaining the same maximum and minimum E values as in the previous stage. It will be seen that both the period of the oscillation and the amplitude of the oscillation is significantly decreased, relative to stage 51. The electric field sequence is repeated with these parameters 15 times, during which the band will be significantly narrowed. Once complete, at time tb, the third convergence stage S3 begins. During this stage S3, the same first and second field gradients 21, 22 as in step S2 are alternatively applied, but at a faster rate of switching (i.e. a shorter step duration). As such, the particles are constrained to a region of the gradient with a higher field magnitude and they experience a stronger focusing effect as seen in Figure 15. The sequence is repeated at these parameters for another 15 cycles (oscillations).

At the end of stage S3, the desired degree of convergence has been achieved and, in this example, the bands of particles are then subject to an optional separation stage S4 in which an electric field gradient of shallower slope is applied and translated along the main axis. In the present case, the separation electric field gradient is translated with a speed of 0.015 cm/s over a distance of about 7.5 cm. The separation process will be discussed further below in relation to later embodiments.

Figures 18 and 18 illustrate another embodiment of a convergence method which includes a particle collection step prior to commencing convergence. As indicated in the flowchart of Figure 18, in an initial step 5101', a static electric field profile 31 is applied across a second portion P2 of the main axis, which includes the first portion P2 as depicted in Figure 19. The static electric field profile includes a zero-crossing point 38, at which the magnitude of the electric field is zero. Immediately to one side of the point 38, the electric field profile 31 has a positive value and immediately to the other side of the point 38, it has a negative value. The slope of the gradient 31 may be less than that of gradients 21 and 22 to be applied in the later particle convergence steps S111, S112. The particle-collection electric field profile 31 has a weak focussing effect on the charged particles in second portion P2, causing them to migrate towards the zero-crossing point. This brings the particles into the first portion P1 of the main axis, x, (and preferably into the first region Ri thereof) such that they will be subject to the convergence process.

An electric field sequence as already described, comprising at least steps S111 and S112, repeated as many times as necessary, will then be performed in portion P1.

Some examples of suitable device layouts in which methods such as those described above may be performed will now be described with reference to Figures 20 and 21. In both examples, the chamber 2 takes the form of an open or closed cavity in the surface of a substrate 19, formed for instance of polymer, glass or ceramic. However many other implementations are possible, including the use of a container such as a tube (e.g. a capillary tube) or an assay well. In other cases the chamber could simply take the form of a volume of gel (or other sieving matrix), with no walls as such, provided it is sufficiently self-supporting. The electrodes 4 of the electric field generator will typically be arranged at the periphery of the chamber 2, spaced from one another along the direction of the main axis. Further details as to how the electrode array 4 can be implemented are disclosed in WO-A-2006/070176 and WO-A-2012/153108. In the Figure 20 example, the chamber 2 (which may or may not be elongate, but preferably has a substantially uniform cross-sectional area along the main axis) is provided with at least one input port 5a, 5b (two are shown here), and an extraction port 7, each of which is optional. The input port(s) are configured for receipt of charged particles, e.g. carried in a solution, which can then be injected from the port(s) to the chamber 2 by application of an appropriate electric field between the input port(s) and the chamber 2 (one or more electrodes for applying this field may be provided but are not shown in the Figure). If the device is to be used for performing a reaction between two or more substances (i.e. as a reactor), the respective substances may be input On the form of charged particles) to respectively different ones of the input ports such that mixing takes place on injection into the chamber 2. Alternatively, a mixed sample of the substances may be prepared and placed into one input port (or directly into chamber 2). One or more extraction ports 7 may also be provided for extracting substance(s) from the chamber 2. Again, the extraction port will typically be equipped with electrodes (not shown) across which an electric field can be generated which will draw out the desired charged particles in use.

Thus the convergence methods described above can be advantageously used to accelerate a reaction between two or more substances, e.g. in a device of the sort shown in Figure 20. The term "reaction" (and related words such as "react") is used throughout this disclosure to encompass both chemical reactions and biological interactions such as binding events. Charged particles of the two or more substances are introduced into the chamber and an electric field sequence applied along at least a first portion of the main axis, x, in the manner described in relation to any of the above embodiments. The group of charged particles becomes narrower (i.e. focuses) as a result of the applied electric field sequence and this convergence of the different substances increases the rate of the reaction between them, due to coincidence of the different particles on the same position over some amount of time, and the increased concentration of the substances (since the volume of space over which the same amount of material is present is reduced). Since the particles of different type will have different mobilities, in any medium other than a vacuum, they will also separate into bands of like particles, which as described with reference to Figures 8 to 12 above, move back and forth along the main axis of the chamber 2, preferably oscillating if the electric field sequence is repeated a plurality of times. The bands overlap one another during at least part of this motion, and preferably cross over one another at least once, preferably multiple times, as the method proceeds, further improving the rate of reaction since the likelihood of collisions occurring between the particles is increased.

When using the disclosed technique in the context of accelerating a reaction, the primary aims are to converge the group of particles as much as possible (to increase concentration) and to achieve a large amount of physical overlap between the respective bands of particles during the cycle. In order to promote these outcomes, first -in order to achieve strong focussing, it is useful to apply an electric field sequence in which the first and second electric field gradients 21, 22 are applied over a reasonably narrow first region Ri and the slope of the gradients is relatively steep, e.g. changing from 0 to at least +/-800 V/cm over a distance of 1cm or less. The example field gradients shown in Figures 4A and 4B are suitable for this. Second -to ensure the bands remain in close proximity during the method, and hence overlap significantly, the oscillation width of the bands is desirably kept small, which may be achieved through control of the cycle time. For instance, a cycle time of about 2 seconds would cause oscillation widths of between 1 and 2 mm for the particle mobilities mentioned in table 1 above.

When selecting a suitable cycle time, a balance needs to be struck between achieving a small oscillation width (requiring a small cycle time) and ensuring that as much of the reactant substances provided in the chamber are captured as possible. The latter can optionally be improved by performing a collection step before convergence begins (as has been discussed) but alternative or in addition would be enhanced by a longer cycle time since this gives more time for widely-spread particles to move together during each step. Hence it may be advantageous either to select a "generous" duration/cycle time (translating to a wide "bucket") to reach the molecules from the edges of the chamber and then they would slide down the gradient to a narrower oscillation without reducing the step duration, or to follow convergence strategies as described elsewhere to keep narrowing the step and increasing the frequency to find an optimum concentration-reaction sequence for the particular experiment. There may also be hardware limitations to consider, because the pulse needs to jump from one situation to another during a full period. If the frequency is too high, electronic distortions may be imposed on the applied field shape. So there may also be a lower limit on the frequency caused by the hardware.

The degree of acceleration, and ultimately the yield of reaction product, will also depend on the number of times the electric field sequence is repeated. The more cycles are performed, the more convergence will be achieved and hence the further the reaction will proceed in the same time frame (until one or more of the reactant substances is fully reacted). Thus, in step S113 of the methods of any of Figures 5, 13 or 18 above (and Figure 22 below), whether or not sufficient convergence has been achieved may be a question of whether the reaction has completed, or at least reached a desired stage. For instance, the electric field sequence of at least steps S111 and S112 could be repeated until at least a predetermined proportion of the primary reactant substance has reacted, e.g. 50%, 60%, 70%, 80%, 90%, 99% or even 100%. The number of cycles required may typically be at least 30, preferably at least 50 or even 100 or more. In order to establish the correspondence between number of oscillations and reaction progress, one or more runs of the process may be performed beforehand (e.g. as part of a calibration process on the same or a corresponding device). For instance, a test run may be performed in which N oscillations are performed, after which the resulting contents of the chamber are analysed, e.g. by performing a separation and using a detector to look for the presence or absence of a particular component, preferably the primary reactant. Multiple runs with different values of N may be performed in order to establish the minimum value of N at which there is substantially no primary reactant detected (meaning that 100% of the primary reactant has reacted). Subsequently, the method can be performed on the real samples with N (or more) oscillations if it is desired to react 100% of the primary component, or with a corresponding fraction of N oscillations should some lesser degree of reaction progress be desired (e.g. 0.5N if only 50% of the reactant is to be reacted).

The reaction method could end with the converging process -i.e. once the desired number of repeats of the electric field sequence have been completed, or once a certain amount of one of the reactants has been used up. One or more of the substances contained within the chamber 2 could then be extracted, e.g. via an extraction port 7. However, if more than substances is present in the chamber at this stage, it will often be difficult to selectively extract only one of them, especially given that they have been actively converged by the applied field, and so will be closely co-located. For the same reasons may not be possible to identify exactly what substances exist in the chamber 2 after a reaction has taken place, e.g. to confirm the presence or absence of a particular reacted complex.

As such, it may be desirable to perform a separation process on the contents of the chamber 2 after the reaction has completed, or reached a certain stage. In a separation process, the different particle types are sorted into respective bands which become spaced from one another and can thus be readily distinguished and identified. For instance, the separation may be performed by electrophoresis, in which an electric field is applied so as to cause the particles to move under its influence through a sieving matrix such as a fluid or gel. This gives rise to a hydrodynamic force acting on each particle in the opposite direction. The magnitude of the electric force experienced by a particle will depend on the electric field at the particle's location and the charge on the particle. The hydrodynamic force will depend on the nature of the sieving matrix and on factors such as the particle's size and shape. Under the combined forces, each particle migrates to a position at which the opposing forces balance one another out, i.e. its equilibrium position. This position will be different for different particle types and hence the mixture of particles becomes sorted by type. Many different examples of suitable electrophoresis methods exist -for instance, in a simple implementation separation could be performed by placing the contents of the chamber 2 into a separation channel containing a sieving matrix and applying a uniform electric field. However, better results can be achieved by applying more complex fields. For example, the electrophoresis techniques disclosed in WO-A-2006/070176 or WO-A-2012/153108 are particularly advantageous. These involve applying an electric field profile (e.g. a gradient) which is then moved along the separation channel. The various bands find equilibrium positions spaced along the gradient and ultimately move with the field gradient as it translates along the channel, becoming better resolved as they do so. The bands can then be individually detected, identified and/or extracted as necessary.

VVhile the contents of the chamber 2 could be removed and transferred to a separate device for performing the separation (e.g. by pipette), in preferred embodiments a separation channel is incorporated into the same device which is used to perform the reaction. A first example of a suitable layout for this is shown in Figure 21. Here, a chamber 2 On which the convergence process is performed to accelerate a reaction as described above) and a distinct, elongate, separation channel 15 are both formed as separate cavities in a substrate 19. The chamber 2 is fluidically linked to the separation channel 15 by a conduit 9, which may be provided with suitable electrodes (not shown) for applying an electric field which draws particles out of the chamber 2 and into the separation channel 15. Since the chamber 2 and separation channel 15 are distinct from one another, in this embodiment the main axis, x, of the chamber 2 and the elongate axis of the separation channel 15 could be at any relative angle to one another (including parallel or perpendicular).

In use, as shown by the flowchart of Figure 22, charged particles of at least two substances will be injected into the chamber 2 (e.g. using input ports 5a and/or 5b, which are optional) and then, after one or more optional steps for collecting the particles together (denoted by step S101"), a convergence method will be performed on the particles within chamber 2 by applying an electric field sequence of the sort described in any of the embodiments above, via electric field generator 4 (steps S111 to S113). Some or all of the contents of the chamber 2 will be transferred through conduit 9 into separation channel 15, forming an injected "plug" of mixed particles therein. Typically this may include particles of at least one reaction product and, in some case, particles of one or more of the original reactant substances. A separation process is then performed (step S120) by applying an appropriate electric field along the elongate axis x' of separation channel 15, e.g. using further electrodes of the electric field generator 4. For instance, this electric field may have an electric field gradient of the sorts disclosed in WO-A-2006/070176 or WO-A-2012/153108, and may be moved along the elongate axis x' as described therein. Ultimately the respective bands of like particles will separate from one another and become spaced along the elongate axis, x'. The bands may then be detected and identified to confirm the presence or absence of a substance (and hence the result of a test, for example). Alternatively or in addition, one or more of the bands can be extracted by using the electric field profile to move the selected band(s) to an extraction port (not shown), and there extracting them from the separation channel to output a yield of the material(s) in question (step S125).

The flowchart of Figure 22 is also applicable to other implementations in which a convergence phase occurs prior to a separation, whether or not a reaction is involved. If the convergence is performed along the same axis as the subsequent separation, this "pre-focusses" the particles into a narrow volume, with the result that each band becomes distinct from the next at an earlier point in the separation than would otherwise be the case-both in terms of time taken, and distance along the separation channel. This significantly reduces the running time of the separation, allowing results to be attained more quickly, and also enables the overall size of the device containing the separation channel to be reduced.

An example of a suitable device layout which enables these benefits to be achieved is shown in Figure 23. Here, the chamber 2 (in which convergence takes place) and the separation channel 15 (in which separation takes place) are contiguous and each form part of one and the same cavity in substrate 19, and are therefore filled with the same sieving matrix. Other components shown in Figure 23 are the same as those previously described with like numbers. Again, as mentioned previously, many other physical formats are possible including providing a tube or other container which contains the two volumes. Importantly, the elongate axis of the separation channel 15 is aligned with the main axis, x, of the chamber 2 (i.e. they are parallel and there is no substantial offset). In this embodiment there is no physical distinction between the chamber 2 and the separation channel 15 and indeed the point where one ends and the other begins is determined by the nature of the electric field which is applied to each, and indeed the lateral extents of these can be changed. It will also be noted that the separation process may begin in the chamber 2, before moving into the separation channel 15.

This is best shown in Figures 25A, 25B and 250, which are each plots showing the electric field applied along the main axis, x, at different points during the process. Figure 25A depicts the situation during the converging phase. As described above, during this phase an electric field sequence will be applied in which a first electric field gradient 21 is applied (step S111 of Figure 22), and then a second electric field gradient 22 is applied, across a first region Ri within chamber 2. This results in the charged particles (which will include a mixture of types, whether the outcome of a reaction or corresponding to the mix first injected into the chamber 2) forming bands of like particles which converge and become increasingly narrow. If the sequence is repeated multiple times, the bands will oscillate within the chamber 2. Once a sufficient degree of convergence has been reached, the electric field sequence will end (step S113) and a separation process will begin (step 5120). As depicted in Figures 25B and 250, this may involve the application of an electric field gradient 41, which is translated along the separation channel in the direction of the main axis. Figure 25B shows the applied field profile at a first time instance t1 during step S120, while Figure 250 shows the applied field profile at a later time t2 still during step S120. As described in more detail in WO-A-2006/070176, the respective bands of like particles each move to different equilibrium positions along the electric field gradient 41 and ultimately move with the translating field. The movement of the field and shape of the gradient can be controlled to place the respective bands at desired positions along the axis. The bands can be detected, e.g. by imaging the channel 15 and/or one or more may be extracted at an (optional) port 7.

It should be noted that while the separation channel 15 has been depicted here as rectilinear, this is not essential and a curved (e.g. circular) separation channel could be used instead, arranged for example as shown in Figure 24. Here the chamber 2 is positioned on a tangent to the circular separation channel. This "closed loop" channel allows the separation to continue as the particles circuit the device. Suitable configurations of the separation channel and electric field applied are disclosed in WO-A-2006/070176.

Results of computer simulations showing the behaviour of particles subject to a separation process of the sort described above will now be described and contrasted with a comparative example based on the principles provided in WO-A-2006/070176. Figure 26A shows the distance of the particles from their equilibrium position, as a function of time. Figure 26B shows the distance between the bands (i.e. resolution) as a function of time. Figure 27A shows the band width as a function of migration distance along the main axis, x. Figure 27B shows the band width as a function of time. In each of the plots, the dashed lines show the outcome of the simulation according to the comparative example, while the solid lines show the outcome of the simulation using the presently disclosed method.

In both cases the layout of chamber 2 and separation channel 15 was the same as that shown in Figure 23 above, and a mixture of two particle types was injected into the chamber 2 at time zero, with particle mobilities of 0.5x10-8(m/s)/(V/m) and 0.51x10-8 (m/s)/(V/m) respectively (which are similar to typical mobilities of bio-particles). It should be noted that only one trace is shown for each simulation in each Figure -this is because in Figures 26A, 27A and 27B the behaviour of each band is near-identical within each respective simulation. Figure 26B is a measure of the distance between the two bands, expressed in terms of multiples of the width of one of the bands (measured as the standard deviation a of its particle distribution), and hence effectively takes data from both bands.

In the comparative example, a field gradient similar to that shown in Figure 25B was immediately applied and translated along the main axis. The simulation used a field gradient extending from +800 V/cm to -800 V/cm over 62.5 mm along the x axis, moving at 0.015 cm/s over a distance of about 7.5 cm. As shown in Figure 26A (dashed line) this caused the band to migrate towards its equilibrium position, reaching it at around t = 50 s and then staying at equilibrium for the rest of the separation (during which the equilibrium position will move along the x axis with the translation of the field gradient). However, at the point of reaching equilibrium, the first band is still merged with the second band as can be appreciated from Figure 268, where the distance between the two bands at around t = 50 s is much less than the width of either band, a. Figures 27A and 278 (dashed lines) also show that the band width is not much reduced from its original size at this stage. The bands are not considered resolved from one another until the distance between them is at least 2a (i.e. twice the width of one of the bands), which occurs at t = t _* preV around 240 seconds after the start of the process.

In the simulation performed according to an embodiment of the presently disclosed technique, upon injection of the two particle types into the chamber, the particles are first subject to a convergence process involving the application of an electric field sequence as described above, with multiple repeats (Figure 25A).

The first field gradient 21 extended from +800 V/cm to 0 V/cm over 8.75 mm, and the second field gradient 22 extended from 0 V/cm to -800 over 8.75 mm. In practice this was achieved using a composite gradient which was moved back and forth along the main axis by 5mm (i.e. the field oscillation width was 5mm). The step duration was 1.25s, making a cycle time of 2.5 s. This led to a typical particle oscillation width of around 3 mm. The sequence was repeated 30 times. The resulting oscillation of the particles is visible in Figure 26A (solid line), continuing up to about t = 75 seconds. During this phase, the particles oscillate back and forth along the main axis and each band narrows as previously described, which can be seen in Figures 27A and 27B (solid lines).

The electric field sequence is then stopped and a separation step initiated, by applying an electric field gradient which is then translated along the axis (Figures 25B, 250). The parameters during this step were the same as applied in the comparative example, i.e. a field gradient extending from +800 V/cm to -800 V/cm over 62.5 mm along the x axis, moving at 0.015 cm/s over a distance of about 7.5 cm. The outcome is visible in Figure 26A (solid line) from about t = 75 seconds: the band migrates towards its respective equilibrium position. However, as shown in Figure 26B (solid line), the two bands now become resolved (i.e. spaced from one another by at least 2a) at t = about 110 seconds. Thus, even at this early time in the process the two bands are visibly distinct from one another and remain so as they continue moving towards their equilibrium positions. As such, the bands can be distinguished from this point forward and the separation "competes" at this point, significantly faster than in the comparative example. Generally, tests have shown that, using the presently disclosed techniques, separation can be achieved in less than half the time of conventional methods, and in some cases in as little as a third or a quarter of the conventional duration.

This is due to the separation phase having been performed on already-focussed bands, rather than on the original (wide) injection plug. This not only significantly shortens the run time, but also allows for much shorter chips (especially in the case of linear implementations), since the separation requires less distance along the main axis to "complete".

It should be appreciated that the partial objective of the initial convergence phase is the opposite of a separation step, i.e. to bring all bands/groups approximately to one point along the main axis, x, and to focus them there, before going to the second stage of separation. In the two step separation approach, the first focussing step gives a head start to the separation step because the separation starts with already focused bands. Thus the separation happens much quicker and before the bands reach their respective equilibrium points.

To illustrate some of the benefits, test data obtained from implementing embodiments of the invention in the form of assays for detecting Covid-19, Lassa fever and influenza, respectively, will be provided below (and contrasted with conventional techniques) with reference to Figures 32 onward.

The simulation already discussed above with respect to Figures 8 to 12 also demonstrates the benefits of performing a convergence process (as disclosed herein) prior to a separation. Here, as mentioned above, after oscillating the bands about a position about 2 cm along the main axis, x, a separation gradient is applied from approx. t = 75 seconds. It can be seen from Figure 8 that clear separation has been achieved by the time the bands reach about 2.5 cm along the main axis, x, which is at around t = 80 seconds. At that point the six trajectories are not parallel, which means that they are still in a transient state, and have not reached their equilibrium (at least not all of them). This allows for a separation runway of about 0.5 cm compared to -4.5 cm in conventional methods -as such it is possible to provide a separation channel that is 9 times shorter than was previously needed.

It should be appreciated that the preferred parameters of the electric field sequence described in the embodiments herein will depend on the desired outcome. Where the convergence process is being used solely to accelerate a reaction, the main aim is to achieve maximum concentration of the substances and also to coincide the particles as closely together as possible, in order to obtain a high reaction yield. Thus, the sequence may be repeated a large number of times in order achieve this -e.g. 100 oscillations or more. In contrast, when the convergence process is being used to focus the group of particles prior to a separation, the main aim is to minimise convergence time in order to proceed to separation as quickly as possible whilst still achieving enough focussing to decrease the separation time also. For example, the application of between 20 and 40 (e.g. 30) oscillation periods has been found to achieve a good reduction in overall running time (and separation distance needed). In cases where the convergence process is being used both to accelerate a reaction and as a pre-focussing step for a separation (in which the reaction product(s) will be separated), a balance between the two factors will be required and some intermediate number of oscillations (e.g. between 50 and 80) may prove to be optimal.

In alternative embodiments, similar principles can be applied to achieve a lesser degree of band focussing prior to beginning separation. These still result in a shorter running time as compared with conventional techniques. For example, rather than apply both first and second electric field gradients 21, 22 (sequentially) in the manner described above, a degree of convergence will be achieved if just one such gradient is applied prior to beginning separation. That is, either the first electric field gradient 21 or the second electric field gradient 22 (not both) might be applied a single time, followed directly by the separation. In this case there will be no oscillation of the bands. The field gradient 21 or 22 used for focussing is preferably relatively steep relative to that (41) used during the subsequent separation step. The change from one gradient to the other may be discrete or continuous. For example, if the first field gradient 21 is used for focussing, this could then start to move along the channel to begin separation and at the same time gradually reduce in slope in order to adopt a configuration which optimises separation as the gradient moves along the rest of the channel. The gradual change in slope could, in some cases, continue for the whole of the separation.

Another alternative is depicted in Figures 28A, 28B and 28C. In this case, the initial particle convergence is performed by applying a static field 31 with a zero crossing point, as shown in Figure 28A. This is similar to the particle collection step S101' mentioned above. The particles migrate towards the zero-crossing point, which is located within the first region in the chamber 2, thereby focusing the group of particles to an extent. The separation step 5120 can then be begun, e.g. by applying a gradient 41 which is translated along the separation channel as previously described (Figures 28B, 280). There is no oscillation of the bands in this embodiment. The transition between the convergence phase S101' and the separation phase 5120 can be discrete or continuous as in the previous example.

Again, the separation will complete more quickly than in conventional techniques due to the increased convergence of the bands at the onset of the separation.

So far, the embodiments presented have been described in the context of scenarios in which all the particles of interest have the same charge sign -i.e. they are all positively charged, or they are all negatively charged (although the magnitudes of their charges will typically differ unless all the particles in the sample are of the same type). This is often preferred, especially where the convergence is being performed in order to promote a reaction between particles of different substance (since, as will be seen, this enables the bands to be more closely intermingled). It is generally possible to arrange for all the particles to have the same charge sign, e.g. through control of the pH of a solution in which the particles are carried (in a fluid-based implementation). However, techniques in accordance with the invention can be applied to other situations in which the initially injected particles are of mixed charge sign (i.e. some positive and some negative).

For context, Figure 29 illustrates the case where a mixed set of particles 10a, 10b, 10c and 10d is subject to a first field gradient 21 of the sort described in relation to the first embodiment above (Figure 3A), in a first step S111 of an electric field sequence. The negative particles 10a, 10b move in the manner previously described, in the +x direction, and converge, forming a first group 12a. However, any positive particles in the same first region Ri, such as particles 10c and 10d shown here, will move in the opposite (-x) direction and hence will not form part of group 12. Moreover, particles will experience an increasing field magnitude as they move in the -x direction (as denoted by the arrows in Figure 29), with the result that particles 10c and 10d become more diverged from one another during the step. Typically, they will be accelerated out of region R1 and no group is formed. This may not be problematic if the particles in question are not of interest.

However, in other cases it will be desirable to focus both the positive particles and the negative particles. Embodiments in which this is possible will be described with reference to Figures 30 and 31. Essentially, these involve simultaneously performing an electric field sequence of the sort already described, in two laterally offset regions along the main axis, with the directions of the electric field gradients reversed relative to one another in the respective regions. In this way, one region can be used to focus positive particles while simultaneously the other is focusing negative particles.

Figure 30A shows an exemplary electric field profile as may be applied along the axis x in a first step (S111) of such a sequence. In a first region R1, a first electric field gradient 21 is applied as already described above. Simultaneously in a second region R2, a third electric field gradient 23 is applied. All points on the third electric field gradient 23 are of the same sign as those on the first electric field gradient 21 (e.g. positive, in this example), but the magnitude of the field decreases in the opposite direction along the main axis, x, as compared with the first field gradient 21. That is, the direction of the slope is opposite. The first and third electric field gradients 21, 23 may together form a composite electric field profile 29'. This has the result that particles of a first charge sign (e.g. negative) will be converged to form a first group 12a in the first region Ri, while particles of the other charge sign (e.g. positive) will be converged to form a second group in the second region Rz. In the next step of the sequence S112, while the second electric field gradient 22 (as described previously) is applied in the first region R1, a fourth electric field gradient 24 is applied in the second region Rz. Again, this will have the same sign as the second field gradient but its slope will be in the opposite direction, causing convergence of the second group 12b to continue. The second and fourth electric field gradients 22, 24 together form another

composite field profile 29".

The steps S111, 5112 may be performed just once or, more preferably, will be repeated a plurality of times in order to achieve a greater degree of convergence. In this case, the positive particles will oscillate primarily in one region while the negative particles will oscillate primarily in another. It is for this reason that particles of substances which are to react with one another are preferably configured to have the same charge sign. That said, it is also possible to arrange for the oscillation width of each group of particles to be so large as to extend into the other region: if the particles are moving quickly enough with the changing fields, the groups may not be significantly affected by crossing into the other region, and hence a reaction between the two groups can still occur. Once sufficient convergence has been achieved, the contents of the chamber can be extracted and/or subject to a separation process in any of the manners previous discussed.

An enhancement to the Figure 30 embodiment is shown in Figure 31. During application of the fields shown in Figure 30A, it is likely that some of the particles will reach the boundary between regions Ri and Rz. Should this occur, some of the negative particles of group 12a would cross into the second region R2 where, if they remain there for a sufficient period of time, they would be accelerated in the +x direction and lost from the convergence process under the influence of the third electric gradient 23. The same could happen to any positive particles which crossed into the first region R1 from the second region R2 (although in the -x direction). To reduce the likelihood of this happening, the first and second regions R1, R2 are preferably spaced from one another along the main axis x by a non-zero distance, forming a third region R3. An example of this is shown in Figures 31A and 31B.

It is not essential to apply any electric field in the third region R3. A space where the field magnitude is zero will act as a buffer zone, to an extent, reducing the likelihood of particles reaching the other region. However, if the field is zero in the third region, there is a risk of particles becoming stationary there and not re-joining their respective groups. To avoid this, it is desirable to apply a non-zero electric field in the third region of the same sign as the field gradients on either side. Hence, as shown in Figure 31A, during step S111, a positive non-zero electric field 25 is applied in the third region and during step S112, a negative non-zero electric field 26 is applied there. Both fields in this example are uniform (i.e. have a substantially flat profile) across the third region R3 -this is desirable in order that there is no converging or diverging effect here on particles of either charge sign. However, this is not essential since field profiles with a shallow slope will have little focussing/defocussing effect and therefore could also be used. Any profile in the third region should, however, be less steep than those employed in the first and second regions.

The focussed bands of positive and negative particles formed in the first and second regions R1, R2 will, at various times in the sequence, arrive at the third region R3 where they can interact with one another. As such, reactions between particles of opposite charge signs can be carried out with such an arrangement, although the reaction efficiency is likely to be lower than that achieved between particles of like charge sign.

As mentioned above, aspects of the present invention provide methods in which at least two substances are bought into close proximity with one another using the converging technique herein described and, optionally, subsequently subjected to a separation step in order to isolate the reaction product and/or identify left over reactants. As described above, such a method has broad applicability in the biomedical field, in which enhancing the speed and/or efficiency of a co-location, binding interaction or reaction between two substances can be crucial.

Accordingly, the present invention may be used in any biological assay, protocol or application in which an enhanced rate of reaction is desirable or where an enhanced co-location or specific binding of multiple molecules is desired. For example, the present invention may be used to determine protein-protein interactions, to determine protein-DNA interactions, in pull-down assays, in immunoprecipitation assays, in high-throughput screening assays, in synthetic biology and/or in drug production.

Understanding the way in which proteins react with one another, and proteins react with nucleic acids (e.g. DNA and/or RNA) or other biological molecules (such as polysaccharides and/or lipids), is of paramount importance in the drug development process. Therapeutic drugs are often designed to target specific disease pathways by targeting specific key proteins or genes within said pathway.

For example, the therapeutic drug may therefore act to enhance the action of the target protein (agonist) or, alternatively, act to disrupt the action of the target protein (antagonist). Accordingly, many pre-clinical studies involve in-vitro studies, which aim to assess the interaction between a test therapeutic drug and the target protein. Additionally, pre-clinical studies may also be used to identify binding partners of a protein of interest or to generate structure details of protein complexes (Zhou et al., ChemMedChem, 11:738-756, 2016).

These protein-protein interactions and protein-DNA interactions can be identified and visualised via a variety of methods, for example, pull-down assays, immunoprecipitation assays, fluorescence polarisation, surface plasmon resonance, nuclear magnetic resonance, circular dichroism, static and dynamic light scattering, analytical ultra-centrifugation, isothermal titration calorimetry or microscale thermothesis. However, these methods suffer from a number of disadvantages, including long assay times, high sample consumption and low through-put (Zhou et al., ChemMedChem, 11:738-756, 2016).

The present invention provides an improved method in which the aforementioned disadvantages are overcome. For example, in the context of determining protein-protein interactions, the method herein disclosed can be used to converge the protein of interest with a test drug to assess the level of binding. Alternatively, the method herein disclosed can be used to detect the presence of one protein via the use of another protein, i.e. the method herein disclosed can be used in the detection of an analyte. As a result of the disclosed technique, a rapid convergence between the protein of interest and the test drug/binding protein will be achieved. This increases the concentration of each of the respective substances at a specific point. This enhanced co-location may act to accelerate any reaction between the two or by the two. Alternatively, the rapid co-location may enhance binding, for example where detection of an analyte is required.

Optionally, the method herein disclosed may further involve an electrophoresis step in which the end product can be isolated and/or visualised, e.g., the protein-drug complex or protein-protein complex, from the starting individual materials. Additionally, the optional electrophoresis step may be used to identify unbound proteins (i.e. those not present in a complex) and subsequently used as a measure or indication of the level of binding that has occurred between the protein of interest and the test drug/binding protein.

An additional method by which physical interactions between proteins can be identified and/or detected is the in-vitro technique known as a "pull-down" assay. Typically, this method involves the use of affinity purification with various wash and elution steps, wherein a "bait" protein is used to isolate other proteins in a sample.

Following the "pull-down" stage, protein fractions can be resolved using sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently visualised by common methods, such as gel-staining or western-blotting. However, a "pull-down" experiment followed by the aforementioned electrophoresis step is extremely time consuming and laborious for the user. This is demonstrated in Louche et al., 2017 (Methods in Molecular VBiology, 1615: 247255) whereby a "pull-down" protocol is provided. Said protocol details a method in which there is over 2.5 hrs of pre-and post-incubation steps and 2.5 hrs of incubation steps, totalling over 5 hrs of experimental steps (not including the subsequent electrophoresis steps).

The invention herein disclosed overcomes these disadvantages of the pull down method. The techniques herein disclosed remove the need for almost all sample preparation steps and furthermore, reduce the incubation times to a fraction of that outlined above (e.g. approximately 1 minute or less). As such, the methods herein disclosed could be used to significantly reduce the time needed to arrive at the analysis of the protein-protein interaction, as well as reducing the number of reagents needed to successfully complete the assay. It is expected that the only preparation of the sample will be to contact the sample with an antibody solution, which can then be used directly in the apparatus of the invention, i.e. introduced into a receptor of the chip and then electrokinetically injected into the channel.

Prolonged prior incubation of the antibody with the sample is not expected to be necessary. It is also expected that all steps may be performed at room temperature. The method of the invention therefore dispenses with the multiple steps involved in a typical immunoassay, such as for example with an ELISA, immobilising antigen on a surface, blocking, washing, primary antibody incubation, washing, secondary antibody incubation, washing and then visualisation, or alternatively precipitation steps with column methods. Further advantages of the present invention are that the antibody binding to the analyte is performed in solution and not on a surface, which provides more natural configurations and more accurate results, particularly with low affinity protein binding analysis.

Accordingly, the present invention also provides the method herein disclosed as an alternative method to a "pull-down" assay. Accordingly, the method herein disclosed can be used to confirm the existence of a protein-protein interaction in a particular sample predicted by other techniques, or used as an initial screening assay for identifying previously unknown protein-protein interactions. Furthermore, the method herein disclosed can be used to detect the activation status of specific proteins. For example, proteins that are activated in response to tyrosine phosphorylation can be co-incubated with an SH2 domain. Said domains target the phosphorylated tyrosine on a given protein. Therefore, the method herein disclosed may be used to facilitate interaction with the phosphorylated proteins, thereby identifying proteins that have been phosphorylated The method herein disclosed may further comprise an electrophoresis step in which the reaction product can be identified and isolated.

An additional technique commonly used in the biomedical field is that of immunoprecipitation. This technique is similar to the "pull-down" method described above, with the exception that immunoprecipitation is based solely off an antibody-antigen interaction. This technique involves the precipitation of a protein antigen out of solution using an antibody that specifically binds to that particular protein. Various forms of immunoprecipitation exist, including individual protein immunoprecipitation, protein complex immunoprecipitation, chromatin immunoprecipitation and RNP immunoprecipitation. lmmunoprecipitation assays can be followed by various methods to further analyse the interaction of interest. For example, Perez et al., 2007 (Methods in Molecular Medicine, 131:123-139) discloses the use of chromatin immunoprecipitation followed by an electrophoretic mobility shift assay to study DNA-binding protein interactions in-vitro. However, similar to the drawbacks of the "pull-down" assay described above, the disclosed protocol involves numerous laborious steps for the user and significant incubation times. In contrast, the present invention offers a time-efficient and significantly more streamlined protocol, enabling the user to arrive at the desired results in a fraction of the time. Accordingly, the present invention also discloses the method herein disclosed as an alternative method to an immunoprecipitation assay.

The drug development process takes many years and involves many stages. One vital stage of this process is the high-throughput screening of drug candidates to shortlist the most promising drug candidates for further development and research. As already described, the invention herein disclosed significantly reduces the time it takes to assess the level of co-location or specific binding of target molecules, and significantly reduces the amount of reagents needed compared to the common biological assays described above, characteristics particularly advantageous in the context of high-throughput screening. As such, the present invention also discloses the method herein disclosed for use in a high-throughput screening assay. That is, the method of the present invention, in which at least two substances are bought into close proximity with one another using the converging technique herein described, may be used to rapidly assess the ability of drug candidates to bind to a drug target molecule (e.g. proteins) or to assess the activation/inactivation of biological activity associated with said drug target molecule. Additionally, the method herein disclosed results in an enhanced rate of the two substances being in contact with one another, allowing for a more time efficient process. Optionally, the method may further comprise an electrophoresis step in order to identify/isolate the reaction product.

The skilled person will readily recognise that all of the above-described bio-related assays have a part to play in the drug discovery process. However, the method herein described significantly improves the ability to run biological assays at an increased rate, resulting in a more stream-lined drug discovery process and the more efficient identification of successful drug candidates.

Exemplary test data obtained by performing Covid-19, Lassa Fever and Influenza assays using methods in accordance with embodiments of the invention will now be described and contrasted with conventional methods, with reference to Figures 10 32 onwards.

Figures 32A and 32B are plots showing the results of two runs of a test for Covid19, performed using a method in accordance with the present invention. In the first run (Figure 32A), the sample was negative (i.e. this is a control run), and in the second run (Figure 32B), the sample was positive, having been "spiked" with a known concentration of the SARS-CoV-2 antigen.. In both cases, the sample was prepared as follows: CoVid-19-negative swab samples from the machine user, were collected following the standard lateral flow device (LFD) protocol and diluted in 100 ul viral transport media (VIM from Neuromics). The swab samples were then mixed with Rabbit-anti-SARS-CoV-2 nucleocapsid protein polyclonal antibodies, concentration 100 ng/ul, and 20% Ficoll 400. The antibodies were previously tagged with Alexa Fluor 488. In the case of the "spiked" sample (Figure 32B), purified Nucleocapsid Antigen [Recombinant SARS-CoV-2 Nucleo-capsid Protein, RayBiotech] was added to the mix so as to achieve 1:1 billion dilution of the stock antigen sample. This corresponds to 100 molecules in the final solution.

After mixing with a pipette, the whole mixture was then transferred to an injection well of a device of the sort described above, and diluted one last time with Trisglycine buffer (pH 8.23 at 25 degrees C) prior to injection. No denaturing agent was added and therefore the proteins are run in their native folded form. The sample was injected with electrokinetic injection to the separation channel.

Next, the particles in the sample were converged and mixed by applying an electric field sequence in accordance with the techniques described above, causing the particles to oscillate and promoting a binding reaction between the antibody and any target antigen present in the sample. The electric field parameters are given in Table 3 below, and involved a static collection stage Si (without oscillation) followed by a convergence phase 32 during which oscillation took place, including translation of the field gradients during each step of the sequence. The first and second electric field gradients were provided by a composite field gradient (as in the Figure 4 embodiment), which was controlled to jump back and forth along the main axis in the manner previously explained. At the end of the oscillation phase the chamber will contain unreacted antibody ("Ab") and/or the bound complex ("Ab+Ag"). The complex will of course only be formed if the target antigen SARS-CoV-2 was present in the sample. An amount of the unreacted antibody ("Ab") will typically remain after reaction whether or not the target antigen SARS-00V-2 was found in the sample, since the antibody will typically be added in sufficient quantity that some will remain unbound. A separation stage 33 was then performed, causing the substance(s) to separate into bands and move along the separation channel as explained above.

A detector, such as a fluorescence detector, is located at a predetermined position along the separation channel. As each band passes the detector, the detector outputs a signal, e.g. based on detected fluorescence if each band carries a fluorophore. (Alternative detection methods can be used such as impedance detection or absorption imaging). Since the bands are separated along the channel they will each reach the detector at a different time. Figures 32A and 32B show the signal output by the detector, vs. time in terms of number of scans (each scan corresponds to 100 ms). The leading band will create a signal earlier, and the trailing band will create a signal later. Typically (but not necessarily) the smaller molecule (e.g. the antibody Ab in an immunoassay) will be leading, and the larger molecule (e.g. the Ab+Ag complex) will be trailing. In Figure 32A we see a single peak at about 6250 scans (approx. t = 10 minutes), which is the Ab only peak (since this was a negative control run). In Figure 32B, in addition to the Ab only peak, we see a second peak, labelled with the arrow "C", at about 7750 scans (approx. t= 13 mins) which is the complex (Ab+Ag) peak. The presence of the second peak C, corresponding to the complex, represents a positive Covid-19 test. It is notable that the technique has successfully detected just 100 molecules of SARS-COV-2 nucleocapsid antigens, which is a concentration much lower than can be detected using conventional methods. This demonstrates the increased sensitivity of the presently disclosed technique.

Table 3

Stage Si Stage Sz Stage Sz Function Collection Convergence / reaction Separation / detection Maximum electric field (E.) +800 V/cm +800 V/cm +800 V/cm Minimum electric field -800 V/cm -800 V/cm -800 V/cm (Emin)

Field Width 120 mm 37.5 mm 62.5 mm

Field Speed 0 (static) 1.5 mm/s 1.5 mm/s

Step duration (= n/a 0.83 s n/a half cycle time) Number of oscillations n/a 100 n/a Field oscillation width (jump) n/a 25 mm (10 electrodes) n/a Stage duration 175s 166s 1000 s It will be seen from Figures 32A and 32B that the run took a total of around 13 minutes to give a positive result (measured from the beginning of stage Si). This can be made shorter by moving the detector closer to the injection point on the device, so that the bands reach it sooner Nonetheless, such durations are still many times faster than conventional PCR test techniques which take about 100 minutes (35 x 3 minute cycle) to attain their highest sensitivity -which still could not detect such low concentrations of antigen as that demonstrated here.

Moreover, since the presently disclosed technique has such increased sensitivity, little or no sample incubation time (e.g. a minute or less) is needed prior to injection of the sample into the device. This reduces the overall time required to perform the assay very substantially.

In another experiment, a Covid-19 assay was performed without any oscillation step. The results are shown in Figure 33. In this run, the (spiked) sample was prepared in the same manner as just described above but, prior to injection, the sample was incubated in stock concentration with 3:1 excess antibody for 4 hours at room temperature. As such, the injected sample contained about 1000 antigen molecules (i.e. 10 times that provided in the Figure 32B test -although this is still an extremely high dilution). This was necessary in order to try to obtain a positive detection result, due to the lower sensitivity of the assay technique used in Figure 33.

Upon injection of the sample into the chamber, the device was controlled to perform a collection stage S1, followed by a static focussing stage S2 (no oscillation) and then a separation stage S3. The applied parameters for each stage are shown in Table 4 below.

Table 4

Stage Si Stage Sz Stage Sz Function Collection Static focussing Separation / detection Maximum electric field (Lax) +800 V/cm +800 V/cm +800 V/cm Minimum electric field (Ernin) -800 V/cm -800 V/cm -800 V/cm

Field Width 120 mm 37.5 mm 62.5 mm

Field Speed 0 (static) 0 (static) 1.5 mm/s

Step duration (= half cycle time) n/a n/a n/a Number of oscillations n/a n/a n/a

Field oscillation width (jump) n/a n/a n/a

Stage duration 175s 250s 1000 s It will be seen from Figure 33 that the detection signal contains much more noise than in the case of Figures 32A and 32B, due to the wider bands of particles (less focussing having been achieved). The antibody peak is still clearly visible at about 7000 scans, while the complex peak (labelled C) is only just noticeable despite the sample containing a much higher concentration of antigen than that used in the Figure 32 tests.

Next, a method in accordance with embodiments of the invention was applied to a Lassa Fever immunoassay using purified antigens. The results are shown in Figures 34A and 34B. To prepare the sample, first the antibody was labelled by mixing 100 pl of Mouse anti Lassa Fever virus GPI polyclonal antibodies (concentration 1000 ng/pl) with 10 pl of modifier reagent from abcam Alexa fluor 488 Conjugation kit. The whole mixture was then transferred to lyophilized material, resuspended and left to incubate in a dark room for 15 minutes. Once incubation was finished, 10 ul of quencher reagent was added, mixed and incubated for 5 more minutes. The samples were then aliquoted prior to storage.

The labelled antibody sample was then diluted to a factor of 1:10 (100 ng/pl) and mixed with 7 pl of distilled H20 and 2 pl of 20% Ficoll 400. In a control run, this mixture was spun for 1 minute and transferred to the injection well of the chip. The sample was diluted one last time with 100 pl of Tris-Glycine buffer (pH 8.23 at 25°C) and assayed.

To provide a "spiked" (positive) sample, lpl of Recombinant Lassa Virus GP1, C-terminal mouse Fc-tag (concentration 1000 ng/pl) was diluted in 19 pl of distilled H20 to achieve a dilution factor of 1:20(50 ng/pl). Then, 1 pl of the diluted sample was mixed with 1 pl of diluted antibodies and 1 pl of 20% Ficoll 400. The mixture was spun for 1 minute and transferred to the injection well of the chip. The sample was diluted one last time with 100 pl of Tris-Glycine buffer (pH 8.23 at 25°C) and assayed.

The assay performed on both the control sample and the spiked sample was a three-stage process involving a collection stage S1, an oscillating convergence/reaction stage S2 and then a separation stage S. The parameters applied were the same as set out in Table 3 above. In the control run, as shown in Figure 34A, a single peak is obtained corresponding to the antibody (i.e. a negative result). As shown in Figure 346, both the antibody-only peak (at approx.

6400 scans) and the complex peak C (at about 7000 scans) can be clearly distinguished, meaning a positive test result. The strength of the complex peak C is much more significant than would be the case using conventional techniques, where typically in order to get a comprehensive incubation of such samples, one needs to keep the sample at the stock concentration in room temperature for over one hour, preferably overnight.

Lastly the effectiveness of the technique was also demonstrated in an influenza B assay. The results are shown in Figure 35. To prepare the (spiked) sample, in a test tube, 6pL of dH20, 2pL of 25% Ficoll 400, 1 pL of Influenza B antigen (IVB Ag) [70 ng/pL] and 1 pL of labelled antibody (Lab) [100 ng/pL] were mixed before pipetting 2 pL of the solution into the injection well of a device. The sample was then diluted to roughly 100 pL of Tris-glycine buffer and assayed. The assay performed was a three-stage process involving a collection stage Si, an oscillating convergence/reaction stage S2 and then a separation stage S. The parameters applied were the same as set out in Table 3 above. As shown in Figure 35, both the antibody-only peak (at approx. 6400 scans) and the complex peak C (at about 6750 scans) can be clearly distinguished, meaning a positive test result. It will be noted that here the complex peak C has a much greater intensity than the antibody-only peak. This is due to a greater proportion of the antibody probe having reacted with the antigen as compared with the situation in the Covid-19 and the Lassa fever assays. This may be due, for instance, to a relatively smaller amount of antibody having been used in the influenza assay.

Claims

CLAIMS1. A method of converging charged particles, comprising: a) injecting charged particles of at least one substance into a chamber having a main axis; b) applying, along at least a first portion of the main axis of the chamber, an electric field sequence, comprising the steps: b1) applying a first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points within a first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a first direction of the main axis; and then b2) applying a second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points within the first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a second direction of the main axis which is opposite to the first direction; whereby, during step (b1), a group of the charged particles moves under the influence of the first electric field gradient in the first direction along the main axis, and in step (b2) the group of charged particles moves under the influence of the second electric field gradient in the second, opposite, direction along the main axis of the chamber, and during each of steps (b1) and (b2) the front edge of the group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is successively reduced.
2. A method according to claim 1, wherein the electric field sequence switches between step (b1) and step (b2) with no intermediate transition steps and preferably substantially zero time elapsing between steps (b1) and (b2).
3. A method according to claim 1, wherein the electric field sequence further comprises a transition step (b1') between step (b1) and step (b2) during which the first electric field gradient is changed to the second electric field gradient, the duration of the transition step (b1') being shorter than the duration of either step (b1) or step (b2), preferably at least 10 times shorter, more preferably at least 50 times shorter, most preferably at least 100 times shorter.
4. A method according to claim 1 or claim 3, wherein the electric field sequence further comprises a transition step (b2') after step (b2) during which the second electric field gradient is changed to the first electric field gradient, the duration of the transition step (b2') being shorter than the duration of either step (b1) or step (b2), preferably at least 10 times shorter, more preferably at least 50 times shorter, most preferably at least 100 times shorter.
5. A method according to any of claims Ito 4, wherein during at least part of step (b1), the first electric field gradient is translated within the first region along the main axis of the chamber in the first direction, and/or during at least part of step (b2) the second electric field gradient is translated within the first region along the main axis of the chamber in the second direction.
6. A method according to any of claims 1 to 5, wherein the electric field sequence is repeated a plurality of times, preferably at least 3 times, more preferably at least 10 times, still preferably at least 30 times, most preferably at least 50 times.
7. A method according to any of the preceding claims, wherein in step (b) the first and second electric field gradients are applied by applying a composite electric field profile along the first portion of the main axis of the chamber, the composite electric field profile including: a first part corresponding to the first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points and the magnitude of the electric field decreases in the first direction of the main axis; and a second part corresponding to the second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points and the magnitude of the electric field decreases in the second direction of the main axis which is opposite to the first direction; and applying the electric field sequence comprises changing the position of the first portion relative to the first axis by applying the composite electric field profile at different positions along the main axis of the chamber in steps (b1) and (b2) respectively, such that during step (b1) the first part of the composite electric field profile is applied along the first region of the main axis of the chamber, and during step (b2) the second part of the composite electric field profile is applied along the first region of the main axis of the chamber.
8. A method according to claim 7, wherein the electric field sequence further comprises a transition step (b1') between step (b1) and step (b2), during which the first electric field gradient is changed to the second electric field gradient in the first region, the duration of the transition step (b1') being shorter than the duration of either step (b1) or step (b2), and during transition step (b1') the first electric field gradient is changed to the second electric field gradient in the first region by moving the composite electric field profile along the main axis.
9. A method according to claim 7 or claim 8, wherein the electric field sequence further comprises a transition step (b2') after step (b2) during which the second electric field gradient is changed to the first electric field gradient in the first region, the duration of the transition step (b2') being shorter than the duration of either step (b1) or step (b2), and during transition step (b2') the second electric field gradient is changed to the first electric field gradient in the first region by moving the composite electric field profile along the main axis.
10. A method according to claim 8 or 9, wherein the electric field sequence is repeated a plurality of times by sequentially moving the composite electric field profile along the main axis in alternating directions.
11. A method according to any of the preceding claims, wherein either all of the charged particles injected in step (a) are positively charged or all of the charged particles injected in step (a) are negatively charged.
12. A method according to any of claims 1 to 6, wherein: in step b1) simultaneously with applying the first electric field gradient in the first region, a third electric field gradient is applied in a second region of the main axis laterally offset from the first region, in which the sign of the electric field is the same as that of the first electric field gradient at substantially all points within the second region and the magnitude of the electric field increases along the main axis in the second region in the first direction; and in step b2) simultaneously with applying the second electric field gradient in the first region, a fourth electric field gradient is applied in the second region, in which the sign of the electric field is the same as that of the second electric field gradient at substantially all points within the second region and the magnitude of the electric field increases along the main axis in the second region in the second direction; whereby, during step (b1), in the first region a first group of the charged particles having a first charge sign moves under the influence of the first electric field gradient in the first direction along the main axis while in the second region a second group of the charged particles having the opposite charge sign moves under the influence of the third electric field gradient in the second, opposite, direction along the main axis, and in step (b2) in the first region the first group of charged particles moves under the influence of the second electric field gradient in the second direction along the main axis, while in the second region the second group of charged particles moves under the influence of the fourth electric field gradient in the first direction along the main axis, and during each of steps (b1) and (b2) the front edge of each group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of each group of charged particles is successively reduced.
13. A method according to claim 12, wherein the first and second regions of the main axis are spaced from one another by a third region of non-zero width, and preferably: in step b1), simultaneously with applying the first and third electric field gradients, a non-zero electric field of the same sign as that of the first and third electric field gradients is applied in the third region, the non-zero electric field preferably having a spatial rate of change of electric field with distance along the main axis which is less than those of the first and third electric field gradients, most preferably being substantially constant across the third region; and in step b2), simultaneously with applying the second and fourth electric field gradients, a non-zero electric field of the same sign as that of the second and fourth electric field gradients is applied in the third region, the non-zero electric field preferably having a spatial rate of change of electric field with distance along the main axis which is less than those of the second and fourth electric field gradients, most preferably being substantially constant across the third region.
14. A method according to any of the preceding claims, wherein the first and second electric field gradients have substantially the same spatial rate of change of electric field with distance along the main axis as one another.
15. A method according to any of the preceding claims, wherein the first and second electric field gradients have substantially the same maximum and minimum electric field magnitudes as one another.
16. A method according to any of the preceding claims, wherein the first and second electric field gradients have substantially the same profile shape as one another.
17. A method according to any of the preceding claims, wherein the first and/or second electric field gradient is monotonic at least along the first region of the main axis.
18. A method according to any of the preceding claims, wherein the first and/or second electric field gradient is substantially rectilinear or curved, such as parabolic or exponential.
19. A method according to any of the preceding claims, wherein steps (b1) and (b2) have substantially the same duration as one another.
20. A method according to any of the preceding claims, wherein the electric field sequence is repeated a plurality of times, steps (b1) and (b2) together occupying at least 80%, preferably at least 90%, more preferably at least 99%, of the duration of each cycle.
21. A method according to any of the preceding claims, wherein the electric field sequence consists of steps (b1) and (b2), plus optional transition steps (b1') and/or (b2'), the electric field sequence preferably being repeated a plurality of times.
22. A method according to any of the preceding claims, wherein the electric field sequence has a total duration of between 0.1 and 100 seconds, preferably between 0.1 and 10 seconds, more preferably between 0.1 and 5 seconds, most preferably between 1 and 3 seconds.
23. A method according to any of the preceding claims, wherein in step (b) the electric field sequence is repeated a plurality of times, the first and second electric field gradients and the duration of each cycle being configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range 0.1 to 10 mm, preferably 0.5 to 5 mm, more preferably 1 to 3 mm.
24. A method according to any of the preceding claims, wherein the first region of the main axis of the chamber has a length in the range 0.1 to 5 cm, preferably 0.5 to 2.5 cm, more preferably 0.5 to 1.5 cm.
25. A method according to any of the preceding claims, wherein in step (a) the charged particles are of a plurality of substances and include charged particles of different mobility, and during step (b) the group of charged particles additionally forms into a plurality of bands due to the influence of the electric field sequence, each band containing charged particles with the same mobility, and the width of each band is successively reduced.
26. A method according to any of the preceding claims, wherein in step (b) the electric field sequence is repeated a plurality of times, the size of the first and/or second regions being reduced as the method proceeds by reducing the width of the main axis along which the respective electric field gradients are applied between successive cycles at least once, preferably every N cycles where N is an integer number greater than or equal to 1.
27. A method according to claim 6, wherein step (b) includes a particle-collection phase and a subsequent particle-converging phase, the size of the first and/or second regions being greater in the particle-collection phase than in the particle-converging phase.
28. A method according to any of the preceding claims, wherein the spatial rate of change of electric field magnitude with distance along the main axis of the respective electric field gradients is increased as the method proceeds, preferably being lower in a particle-collection phase than in a subsequent particle-converging phase.
29. A method according to any of the preceding claims further comprising, after step (a) and before step (b): a') collecting the charged particles together to place a major proportion of the group of charged particles, preferably substantially all, within the first portion of the main axis of the chamber.
30. A method according to claim 29, wherein step (a') comprises: ) applying a static electric field gradient along a second portion of the main axis which is larger than and includes the first portion, the static electric field gradient including a zero crossing point begin which is located in the first portion, whereby the charged particles move under the influence of the static electric field gradient towards the zero crossing point.
31. A device for converging charged particles, comprising: a chamber into which charged particles of at least one substance are injected in use, the chamber having a main axis; an electric field generator configured to apply electric fields along at least a first portion of the main axis of the chamber; and a controller configured to control the electric field generator, and programmed to control the electric field generator to apply an electric field sequence along at least the first portion of the main axis of the chamber, the electric field sequence comprising the steps: b1) applying a first electric field gradient, in which the sign of the electric field is one of either positive or negative at substantially all points within a first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a first direction of the main axis; and then b2) applying a second electric field gradient, in which the sign of the electric field is the other of either positive or negative at substantially all points within the first region of the main axis and the magnitude of the electric field decreases along the main axis in the first region in a second direction of the main axis which is opposite to the first direction; whereby in use, during step (b1), a group of the charged particles moves under the influence of the first electric field gradient in the first direction along the main axis, and in step (b2) the group of charged particles moves under the influence of the second electric field gradient in the second, opposite, direction along the main axis of the chamber, and during each of steps (b1) and (b2) the front edge of the group moves more slowly than the trailing edge of the group due to the smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is successively reduced.
32. A device according to claim 31, wherein the controller is further programmed to control the electric field generator to perform the methods of any of claims 1 to 30.
33. A device according to claim 31 or 32, wherein the electric field generator comprises an array of electrodes disposed along at least the first portion of the main axis of the chamber.
34. A device according to any of claims 31 to 33, wherein the chamber is provided with at least one input port for injecting a sample into the chamber, the sample comprising at least the charged particles of at least one substance.
35. A device according to any of claims 31 to 34, wherein the chamber is provided with at least one exit port for extraction of groups or bands of charged particles from the chamber
36. A device according to any of claims 31 to 35, further comprising a detector adapted to detect groups or bands of charged particles in the chamber, the detector preferably being adapted to image the groups or bands.
37. A method of carrying out a reaction between at least two substances, comprising: (i) providing charged particles of the at least two substances; and (ii) converging the charged particles within a chamber using the method of any of claims Al to A25, whereby the charged particles of each substance form a respective band, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting the reaction between the substances.
38. A method according to claim 37, wherein in step (ii) the electric field sequence is repeated a plurality of times, preferably at least 30 times, more preferably at least 50 times, still preferably at least 100 times.
39. A method according to claim 38, wherein in step (ii), the electric field sequence is repeated until a predetermined amount of one of the at least two substances has reacted, the predetermined amount preferably being 30%, 50%, 70%, 80%, 90%, 99% or 100%.
40. A method according to claim 38 or 39, wherein the cycle time is between 0.1 and 100 seconds, preferably between 0.1 and 10 seconds, still preferably between 1 and 5 seconds.
41. A method according to any of claims 38 to 40, wherein the first and second electric field gradients and the duration of each cycle are configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range of less than 10 mm, preferably less than 3 mm, more preferably less than 2 mm.
42. A method according to any of claims 37 to 41, wherein the parameters of the electric field sequence and the respective motilities of the charged particles of each substance are such that the respective bands of each substance cross over one another within the chamber during step b, preferably repeatedly.
43. A method according to any of claims 37 to 42, further comprising, after step (ii): (iii) separating the charged particles according to their type by performing electrophoresis on the content of the reaction chamber, whereby a reaction product is separated from any remaining amounts of the at least two substances.
44. A method according to any of claims 37 to 43, further comprising, after step (ii) or after step (Hi): (iv) extracting a reaction product, preferably by applying an electric field configured to cause charged particles of the reaction product to move to a port through which the reaction product is removed.
45. A method of detecting an analyte with a binding protein that specifically binds the analyte, said method comprising: (i) providing charged particles of the analyte and the binding protein; and (ii) converging the charged particles within a chamber using the method of any of claims 1 to 30, whereby the charged particles of each of the analyte and the binding protein form a respective band, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting the reaction between the analyte and the binding protein.
46. A reactor for performing a reaction between at least two substances, the reactor comprising a device for converging charged particles according to any of claims 31 to 36, whereby in use charged particles of each substance form a respective band in the chamber, the width of which is successively reduced, the respective bands overlapping one another during at least part of the electric field sequence thereby promoting the reaction between the substances.
47. A reactor according to claim 46, wherein the controller is further programmed to control the electric field generator to perform the methods of any of claims 37 to 45.
48. A reactor according to claim 46 or 47, further comprising a separation channel fluidicly connected to the chamber, for performing electrophoresis therein, the electric field generator being further adapted to apply electric fields along the separation channel and the controller being further configured to control the electric field generator to apply a particle-separation electric field along the separation channel.
49. A reactor according to claim 48, further comprising an extraction port for extracting a separated band of particles from the separation channel.
50. A reactor according to claim 48 or 49, wherein the separation channel is connected to an exit port of the chamber or the separation channel is contiguous with the chamber and aligned along the main axis of the chamber
51. A reactor according to any of claims 46 to 50, further comprising a detector adapted to detect separated bands of charged particles in the separation channel, the detector preferably being adapted to image the bands.
52. A method of separating at least two substances, comprising: (I) providing charged particles of the at least two substances; (II) converging the charged particles within a chamber containing a sieving medium using the method of any of claims Al to A25, whereby the width of the group of charged particles is reduced; and then (Ill) performing electrophoresis on the group of charged particles along a separation channel containing a sieving medium, the separation channel being contiguous with the chamber and being aligned with the main axis of the chamber.
53. A method according to claim 52, wherein in step (II) the electric field sequence is repeated a plurality of times, preferably at least 10 times, more preferably at least 30 times, still preferably at least 50 times.
54. A method according to claim 52 or 53, wherein the electric field sequence is repeated until the width of the group of charged particles reaches a predetermined threshold, the predetermined threshold preferably being lOmm or less, more preferably lmm or less, still preferably 0.1mm or less.
55. A method according to claim 53 or 54, wherein the cycle time is between 0.1 and 10 seconds, preferably between 0.1 and 5 seconds, still preferably between 1 and 3 seconds most preferably between 1.5 and 2.5 seconds.
56. A method according to any of claims 53 to 55, wherein the first and second electric field gradients and the duration of each cycle are configured such that the charged particles oscillate along the main axis of the chamber, the width of the oscillation being in the range 0.1 to 10 mm, preferably 0.5 to 5 mm, more preferably 1 to 3 mm.
57. A method according to any of claims 52 to 56, wherein in step (Ill), electrophoresis is performed by: Ill') applying a particle-separation electric field along the separation channel, the particle-separation electric field having a field profile and thereby causing the charged particles to move relative to the sieving medium; and III") varying the applied particle-separation electric field so as to adjust the field profile relative to the separation channel, thereby causing the charged particles to separate into a respective band of each of the at least two substances under the combined influences of an electric force due to the electric field and a hydrodynamic force due to the sieving medium.
58. A method according to claim 57 wherein the field profile of the particle-separation electric field in step (III) has an electric field gradient of which the spatial rate of change of electric field magnitude with distance along the separation channel is less than that of the first and second electric field gradients in step (II).
59. An electrophoresis device for separating charged particles, comprising a device for converging charged particles according to any of claims 31 to 36 and a separation channel which is contiguous with the chamber and aligned along the main axis thereof, the electric field generator being further adapted to apply electric fields along the separation channel and the controller being further configured to control the electric field generator to apply a particle-separation electric field along the separation channel.
60. An electrophoresis device according to claim 59, wherein the controller is further programmed to control the electric field generator to perform the methods of any of claims 52 to 58.
61. An electrophoresis device according to claim 59 or 60, further comprising an extraction port for extracting a separated band of particles from the separation channel.
62. An electrophoresis device according to any of claims 59 to 61, further comprising a detector adapted to detect separated bands of charged particles in the separation channel, the detector preferably being adapted to image the bands.
63. A method of separating at least two substances, comprising: (I) preparing a sample containing the at least two substances to provide charged particles of the at least two substances; (II) collecting the charged particles within a chamber containing a sieving medium by applying a particle-collection electric field gradient along at least a portion of a main axis of the chamber, whereby the charged particles move under the influence of the electric field gradient to form at least one group of charged particles, the front edge of the group moving more slowly than the trailing edge of the group due to a smaller electric field magnitude acting on the particles at the front edge" such that the width of the group of charged particles is reduced; and then (III) performing electrophoresis on the group of charged particles along a separation channel containing a sieving medium, the separation channel being contiguous with the chamber and being aligned with the main axis of the chamber.
64. A method according to claim 63, wherein the particle-collection electricfield gradient is static.
65. A method according to claim 63 or 64, wherein the particle-collection electric field gradient includes a zero crossing point within the portion of the main axis of the chamber, whereby the at least one group of charged particles moves towards the zero crossing point.
66. A method according to claim 63, 64 or 65, wherein in step (III), electrophoresis is performed by: III') applying a particle-separation electric field along the separation channel, the particle-separation electric field having a field profile and thereby causing the charged particles to move relative to the sieving medium; and III") varying the applied particle-separation electric field so as to adjust the field profile relative to the separation channel, thereby causing the charged particles to separate into a respective band of each of the at least two substances under the combined influences of an electric force due to the electric field and a hydrodynamic force due to the sieving medium.
67. A method according to claim 66, wherein the field profile of the particle-separation electric field in step (III) has an electric field gradient of which the spatial rate of change of electric field magnitude with distance along the separation channel is less than that of the particle-collection electric field gradient in step (II).
68. A method according to any of claims 66 or 67, wherein the electric field is changed discretely or continuously from the particle-collection electric field gradient to the particle-separation electric field profile between step (II) and step (III), preferably while being moved from the chamber to the separation channel.
69. An electrophoresis device for separating charged particles of at least two substances, comprising: a chamber into which charged particles are injected in a sieving medium in use, the chamber having a main axis; a separation channel contiguous with the chamber and aligned with the main axis of the chamber; an electric field generator configured to apply electric fields along at least a first portion of the main axis of the chamber and along the separation channel; and a controller configured to control the electric field generator, and programmed to control the electric field generator first to apply a particle-collection electric field gradient along at least a portion of a main axis of the chamber, whereby in use the charged particles move under the influence of the electric field gradient to form at least one group of charged particles, the front edge of the group moving more slowly than the trailing edge of the group due to a smaller electric field magnitude acting on the particles at the front edge, such that the width of the group of charged particles is reduced; and then to apply a particle-separation electric field along the separation channel whereby electrophoresis is performed on the group of charged particles.
70. An electrophoresis device according to claim 69, wherein the controller is further configured to control the electric field generator to perform the method of any of claims 63 to 68.
71. An electrophoresis device according to claim 69 or 70, wherein the electric field generator comprises an array of electrodes disposed along at least the first portion of the main axis of the chamber and along the separation channel.
72. An electrophoresis device according to any of claims 69 to 71, wherein the chamber is provided with at least one input port for injecting a sample into the chamber, the sample comprising at least the charged particles of at least two 15 substances.
73. An electrophoresis device according to any of claims 69 to 72, wherein the separation channel is provided with a least one extraction port for extraction of bands of charged particles from the channel.
74. An electrophoresis device according to any of claims 69 to 73, further comprising a detector adapted to detect bands of charged particles in the separation channel, the detector preferably being adapted to image the bands.