WO2017210635A1

WO2017210635A1 - Method for mass spectrometry acquisition and analysis using overlapping isolation windows

Info

Publication number: WO2017210635A1
Application number: PCT/US2017/035816
Authority: WO
Inventors: Michael R. Heaven; Archibald L. Cobbs
Original assignee: Vulcan Analytical, Llc.
Priority date: 2016-06-03
Filing date: 2017-06-02
Publication date: 2017-12-07

Abstract

The invention provides general methods for data acquisition and analysis of any small organic or biological molecule in mass spectrometric assays. The methods include acquiring overlapping precursor isolation windows combined with a deconvolution approach to increase the specificity of isolation windows and corresponding fragment ion spectra. Applications of the method enable identifying and quantifying analytes with smaller reconstructed isolation windows to provide greater detection sensitivity and less interference during fragment ion chromatogram quantification.

Description

METHOD FOR MASS SPECTROMETRY ACQUISITION AND ANALYSIS USING OVERLAPPING ISOLATION WINDOWS

FIELD OF THE INVENTION

The present invention relates to the operation of a mass spectrometer in data independent acquisition mode for detecting and quantifying known and unknown analytes such as proteins, peptides, metabolites, small molecule chemicals, and other organic or inorganic compounds. The disclosure further provides methods for analyzing data from overlapping precursor isolation windows that is useful for increasing analyte precursor ion specificity and deconvoluting fragment ion spectra.

BACKGROUND OF THE INVENTION

Mass spectrometry-based data independent acquisition (DIA) allows measuring thousands of analytes in a single analysis by systematically fragmenting all precursors within a predetermined mass-to-charge (m/z) range (Venable, J.D. et al. Nat. Methods, 2004; 1: 39-45: Gillet, L.C., et al. Mol. Cell Proteomics, 2012; 11: 1-17: Plumb, R. et al. Rapid Commun Mass Spectrom., 2006; 20: 1989-1994). Most often, this is accomplished by increasing the size of precursor isolation windows applied relative to targeted selected reaction monitoring (SRM) (Unwin, R.D., et al. Mol. Cell Proteomics, 2005; 4: 1134-1144), parallel reaction monitoring (PRM) (Peterson, A.C., et al. Mol. Cell Proteomics, 2012; 11 : 1475-1488) or shotgun proteomics (Stahl, D.C., et al. J. Am. Soc. Mass Spectrom., 1996; 7: 532-540) resulting in less precursor ion selectivity and noisier fragment ion spectra (Egertson, J.D., et al. Nat. Methods, 2013; 10: 744-746). A workaround to avoid large DIA isolation windows was reported in the highly sensitive precursor acquisition independent from ion count (PAcIFIC) method (Panchaud, A., et al. Anal. Chem., 2009; 81 : 6481-6488). In PAcIFIC mode, MS2 spectra are acquired with ± 1.25 m/z-wide isolation windows across a limited precursor range that does not encompass most observable analytes from a complex peptide digest. During data analysis, numerous PAcIFIC injections are combined from various precursor m/z ranges to measure peptides across a more comprehensive precursor analyte range.

High-throughput DIA techniques have been described such as sequential windowed acquisition of all theoretical fragment ions (SWATH) that allow each sample to be analyzed once (Aebersold, R. U.S. patent application 2006/0008851). However, due to scan speed limits, approximately 4-to-75-fold larger isolation windows are applied than shotgun proteomics to measure precursors associated with most detectable peptides, even with recent versions of SWATH that apply variable isolation window techniques (Navarro, P., et al. Nat. Biotechnol., 2016; 34: 1130-1136). The highly complex SWATH MS2 spectra generated contain up to ten peptides in human samples (Wang, J. et al. Nat. Methods, 2015; 12: 1106-1108), producing lower signal-to-noise than DIA analyses with smaller isolation window lengths (Heaven, M.R., et al. J. Mass Spectrom., 2015; 51 : 1-15: Navarro, P., et al. Nat. Biotechnol., 2016; 34: 1130-1136) (herein incorporated by reference). Consequently, most SWATH approaches have the drawback of detecting fewer peptides than shotgun acquisition (Heaven, M.R., et al. J. Mass Spectrom., 2015; 51 : 1-15: Tsou, C.C., et al. Nat. Methods, 2015; 12: 258-264) (herein incorporated by reference), with one exception that applied the SWATH Atlas spectral library (Wang, J. et al. Nat. Methods, 2015; 12: 1106-1108), and another where checking the MS2 chromatograms revealed half the peptides detected were false positives (Li, Y., et al. Nat. Methods, 2015; 12: 1105-1106). To address the problem of low precursor selectivity in DIA analyses, methods for deconvoluting data are known in the art. The DIA approach termed MS^e (Plumb, R. et al. Rapid Commun Mass Spectrom., 2006; 20: 1989-1994), applies chromatographic deconvolution through correlation of intensity and retention time characteristics of precursors analytes in MSI scans to fragment ions in MS^e scans (Geromanos, S.J. et al. patent application PCT/US2005/017743). Other modified applications of grouping precursor and fragment elution profiles have also been used more recently in SWATH software tools (Tsou, C.C., et al. Nat. Methods, 2015; 12: 258-264).

Prior art deconvolution methods for increasing precursor ion selectivity have been described in the MSX DIA approach with the Orbitrap mass spectrometer (Egertson, J.D., et al. Nat. Methods, 2013; 10: 744-746). In MSX DIA, each scan isolates five separate 4 m/z-wide windows simultaneously for fragment ion mass analysis. The five windows isolated in each scan are chosen randomly from the set of 100 possible non-overlapping windows covering the 500-900 precursor m/z range. Each mixed MSX spectrum is deconvoluted into the five component spectra corresponding to each isolated window. The drawbacks with MSX DIA are that it cannot be applied to mass spectrometers that lack the ability to scan multiple precursor windows in a single scan, such as quadrupole time-of-flight (QTOF) mass spectrometers. The method also uses an all or none detection test (e.g., whether a peak is present or absent) to assign fragment ion peaks into a more specific isolation window, which is problematic for analytes with natural isotopes that straddle multiple MSX DIA fragment ion spectra.

SUMMARY OF THE INVENTION The following summary is provided to facilitate an understanding of some of the innovative features unique to the present invention and is not intended to be a full description. A complete appreciation of the various aspects of the invention can be obtained from the entire specification, drawings, and abstract.

In some embodiments, the present invention provides methods for quantifying the relative concentration of many analytes across samples by label-free quantification. The present invention can also be applied to analyze isotopically labeled analytes. Non- limiting examples include applying the present invention to analyze stable isotope labeled amino acids in cell culture (SILAC) (Ong, S.E., et al. Mol. Cell Proteomics, 2002; 1: 376-386). In SILAC applications, a cell culture containing labeled arginine 13C(9) 15N(1) (+10.027 Da) and labeled lysine 13C(6) 15N(2) (+8.014 Da) as well as a separate cell culture with unlabeled media allows two different types of biological conditions to be compared, such as a drug treatment added to the SILAC culture compared to untreated in the unlabeled culture. Upon mixing equal amounts of total protein from each sample, the unlabeled and labeled cell culture proteins are digested with trypsin and the resulting peptides containing lysine and arginine can be compared quantitatively to determine polypeptide abundance in the labeled or unlabeled sample by the light to heavy intensity ratio. In one aspect, the present invention is particularly advantageous with respect to fragment ion spectra intensity quantification by SILAC because the deconvoluted sections of reconstructed MS2 spectra can be designed so the unlabeled and labeled peptide forms are isolated in the same raw MS2 spectra eliminating errors from scan-to-scan intensity artifacts.

In some applications detecting and quantifying unmodified and modified forms of polypeptides in separate deconvoluted MS2 spectra is preferred because it increases identification confidence. In such cases, the present invention provides sufficiently small isolation window MS2 spectra to resolve most modified versions of peptides even when using high-throughput applications that require injecting each sample only once.

Methods of the invention that increase precursor ion selectivity are particularly useful for analyzing highly complex mixtures. In one aspect, protein samples are digested separately with numerous proteolytic enzymes such as trypsin, Glu-C, and Asp-N, followed by quenching the reactions, combining the digests together, and analyzing the samples in a single liquid or gas chromatography-mass spectrometry analysis. The present invention provides a faster method to increase the throughput of sample analysis as compared to analyzing each proteolytic enzyme digest in a one-at-a- time manner. In such applications, each proteolytic enzyme can be used to digest the same protein starting material such as a cell whole lysate. In other applications, each proteolytic enzyme can be used to digest a different protein sample. For instance, trypsin can be used to digest a nucleus enriched cell fractionated sample that predominantly cleaves proteins at the C-terminus of lysine and arginine, Glu-C can be used to digest a cytosolic enriched cell fractionated sample to generate peptides from cleavage at the C-terminus of glutamic and aspartic acid residues, and Asp-N can be used to digest a mitochondrion enriched cell fractionated sample to generate peptides from N-terminal cleavage of aspartic acid. By mixing digests from each cellular enriched fraction together, followed by simultaneous analysis with overlapping isolation windows and deconvolution in accordance with the present invention, relative protein abundance differences in each subcellular organelle can be ascribed to a specific subcellular organelle based on a peptide's observed cleavage site(s) from an organism's reference protein database. Compatible samples in accordance with the present invention include, but are not limited to, the analysis of analytes derived from cell cultures, viruses, biological fluids and tissues, serum, blood, urine, and environmental samples such as soil, plants, and water samples. A non-limiting list of analytes that can be subjected to the acquisition and deconvolution methods of the present disclosure include metabolites, organic molecules, inorganic molecules, peptides, intact protein top-down analyses, metabolites, and lipids.

In some embodiments, biomolecule analytes including peptides are optionally quantified by comparing the concentrations in samples from a healthy subject or pooled sample of healthy subjects compared to a subject suspected of having a disease. The increased sensitivity and specificity of fragment ion spectra generated by the present invention make it particularly useful for assessing and diagnosing disease states or monitoring the therapeutic response to a treatment.

BRIEF DESCRIPTION OF THE DRAWINGS

Various features and advantages of the embodiments herein can be fully appreciated when considered with the accompanying drawings/tables, wherein:

FIG. 1 depicts data acquisition with overlapping precursor isolation windows.

FIG. 2A depicts three raw MS2 spectra N, N+l, and N+2 used to assign fragment ions from raw scan N+l into reconstructed spectra with 3-fold more specific isolation windows.

FIG. 2B depicts the deconvolution of scan N+l .

FIG. 3 shows MS2 chromatograms of the peptide YGWLAAPQAYVSEK with deconvolution (top), as compared to the raw MS2 chromatogram (bottom). FIG. 4 shows MS2 chromatograms of the peptide ENAPAIIFIDEIDAIATK with deconvolution (top), compared to the raw MS2 chromatogram (bottom).

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The following description of the preferred embodiment(s) is merely exemplary in nature and not intended to limit the scope of the invention, its application, or uses, which may, of course, vary. The invention is described with relation to the non-limiting definitions and terminology included herein. These definitions and terminology are not designed to function as a limitation on the scope or practice of the invention, but are presented for illustrative and descriptive purposes only. It should be noted that like elements disclosed below are indicated by like reference numbers in the drawings.

The present invention is an improvement from state-of-the-art mass spectrometry approaches that collect spectra with the smallest isolation windows possible, corresponding to the smallest list of possible analytes that may match a given spectrum. Such an approach is applied in various methods including the PAcIFIC and MSX DIA approaches known in the art (Panchaud, A., et al. Anal. Chem., 2009; 81 : 6481-6488; Egertson, J.D., et al. Nat. Methods, 2013; 10: 744-746). The present invention provides methods that are faster than the PAcIFIC approach by lacking the requirement to inject each sample many times to perform DIA analysis with small precursor isolation windows. Furthermore, in contrast to MSX DIA, the present invention can be applied to a broad class of mass spectrometers such as the Bruker Impact II QTOF, Sciex 6600 QTOF, Waters Synapt QTOF, or Agilent 6545 QTOF.

Applications of the current disclosure use fragment ion intensity combined with a system of overlapping and non-overlapping isolation window sections to sort fragment ions into a more exact precursor range origin. This allows the present invention to identify more analytes such as polypeptides generated from in vitro proteolysis of protein samples. Moreover, quantification of analytes is also improved by removing interfering analytes that can be resolved by the deconvolution process of the present invention.

FIG. 1 shows an example 101 system of overlapping precursor isolation windows of a mass spectrometer in accordance with the present invention. During data acquisition precursor isolation windows are collected that increase by 2/3^rds the length of the isolation window. The precursor isolation windows are also acquired in chronological order so the first MS2 scan N, has a precursor isolation window m/z range from 538-544, the next MS2 scan N+l from 542-548, and the last MS2 scan N+2 from 546-552. Raw scan N+l is deconvoluted into three equally-sized precursor range subsections 102, 103, and 104. Section 102 is an overlapping precursor isolation window to the previous scan N, section 103 is a center region with no precursor range overlap with scans N and N+2, and section 104 is an overlapping region shared between scan N+l and the next sequential scan N+2.

FIGs. 2A and 2B show raw scans N, N+l, and N+2 in 105 from example system 101. The deconvolution procedure used on raw scans in 105 applies the known boundaries of acquired overlapping 102 and 104 sections and non-overlapping 103 precursor isolation window ranges, fragment ion m/z, fragment ion intensity, and scan time. As shown in 106 for deconvoluting scan N+l, fragment ions are assigned to overlapping section 102 from scans N and N+l acquired sequentially in time for the precursor m/z range from 542-544 that contain fragment ions with similar m/z values within an error tolerance depending on the instrument used such as ± 0.015 m/z for QTOF mass spectrometers, and a≤ 3 -fold intensity difference. For overlapping section 104, the deconvoluted scan in 106 is generated by assigning fragment ions from scans N+l and N+2 acquired sequentially in time for the precursor m/z range from 546-548 that have fragment ions with similar m/z values and a≤ 3-fold intensity difference. Following the assignment of fragment ions to overlapping sections 102 and 104, reconstructed spectra are generated for analyte detection or quantification that optionally sums the intensity of each distinct fragment ion m/z in the corresponding overlapping scans as shown in 106.

Fragment ions in scan N+l shown in system 101 are also deconvoluted in center region 103 from precursor m/z 544-546 where the prior raw MS2 scans N and N+2 do not have precursor m/z overlap with scan N+l . Fragment ions originating from an analyte in center region 103 for the raw scan N+l shown in 105 are determined by identifying fragment ions with similar m/z in the preceding raw scan N or next raw scan N+2 shown in 105, and only retaining those with intensity increases > 1.1 -fold in scan N+l relative to scans N or N+2. Following the assignment of fragment ions to center section 103, a reconstructed spectrum shown in 106 is generated for analyte detection or quantification using the intensity and fragment ion m/z values in the corresponding raw scan N+l shown in 105.

Reconstructed MS2 spectra shown in 106 are then analyzed to detect or quantify analytes by a variety of methods known in the art. Suitable detection methods for polypeptides include using an MS/MS database searching program (Craig, R. et al. Bioinformatics, 2004; 20: 1466-1467) or extraction of peptide data based on a spectral library (Gillet, L.C., et al. Mol. Cell Proteomics, 2012; 11 : 1-17).

The term 'overlapping' as used herein refers to the common precursor isolation window range included in fragment ion spectra scans that are collected within a time frame analytes elute from chromatography systems interfaced to a mass spectrometer. Usually, the time an analyte elutes chromatographically is less than a minute, but can be more time depending on the elution gradient and parameters known by those of ordinary skill in the art.

The term 'isolation windows', 'mass-isolation windows', or 'precursor isolation windows' refers to the unfragmented analyte m/z range allowed to pass through a mass spectrometer to subsequently be fragmented by collision induced dissociation or other fragmentation methods such as electron transfer dissociation for detection by another mass spectrometer configured in a tandem configuration. Various non-limiting instrument configurations provide such an ability including QTOF, triple quadrupole, or quadrupole-Orbitrap instruments. In these examples, the quadrupole mass spectrometer is the device performing the mass filtering that determines the isolation window size, length, or range, commonly expressed in m/z units. It is appreciated that other mass spectrometry techniques utilizing ion mobility are also in accordance with the present invention. Moreover, each mass spectrometer in accordance with the present invention may optimally perform the deconvolution steps of the inventive method with a fragment ion m/z tolerance that is appropriate given the resolution of such instruments.

The term 'fragment ion' or 'fragment ions' refers to the mass spectrometry signals obtained from dissociation of a precursor m/z range. Commonly used acronyms include MS/MS, MS2, MS3, and MS".

As used in the disclosure, 'sequential' refers to acquiring a precursor m/z range in chronological and numerical order.

The term 'reconstructed spectrum' refers to m/z and corresponding intensity values of analytes that have been subjected to the deconvolution procedure described in the disclosure. During the reconstruction of a spectrum in an 'overlapping section,' the deconvoluted spectrum can be assigned the average time between the two overlapping fragment ion scans. The intensity of fragment ions in reconstructed scans of overlapping sections 102 and 104 can be the sum, average, lowest, or highest intensity value in the two overlapping scans. Additionally, the fragment ion m/z values written into each overlapping section 102 or 104 reconstructed scan are averaged, or in the preferred method the m/z value is used from the scan wherein the fragment ion had greater intensity.

An 'analyte' as used herein is any charged gaseous molecule detectable by mass spectrometry.

The present invention also includes acquisition and deconvolution methods for use with mass spectrometers that can obtain fragment ion information from multiple precursor isolation window m/z ranges in a single scan such as the Thermo Q-Exactive Orbitrap. For example, each fragment ion scan in the inclusion list contains three distinct and sequential overlapping isolation windows whose boundaries generate 102, 103, and 104 sections as depicted in scans N, N+l, and N+2 in system 101. The resulting data is deconvoluted in the same manner as depicted in 106, except instead of having three sequential scans, only a single scan is simultaneously acquired that includes data from all three precursor isolation windows.

In some embodiments, sequential MS2 scans lack any precursor m/z range that does not overlap, except for the smallest and largest precursor m/z isolation windows analyzed. For example, using 6 m/z-wide isolation windows to analyze precursor m/z values centered at 403, 406, and 409. The fragment ion scan centered at precursor m/z

403 is reconstructed into two reconstructed spectrums. One deconvoluted spectrum assigned the precursor m/z range of 400-403 contains fragment ions not detected in the next sequential scan centered at 406 m/z, or that were increased by more than a threshold intensity value such as 1.1 -fold in the scan centered at precursor m/z 403 compared to 406. A second reconstructed spectrum is also synthesized containing fragment ions assigned to the more exact precursor m/z range of 403-406 that were detected in both overlapping MS2 scans centered at 403 and 406 precursor m/z with a maximum intensity difference of≤ 3-fold. The next overlapping precursor m/z range from 406-409 precursor m/z is assigned fragment ions during the deconvolution process that were detected in both overlapping MS2 scans centered at 406 and 409 precursor m/z with a maximum intensity difference≤ 3-fold. Next, fragment ions are assigned to the more exact precursor m/z range of 409-412 that were undetected in the prior MS2 scan centered at precursor m/z 406, or had an increase in intensity of more than 1.1 -fold in the MS2 scan with a precursor center of 409 compared to 406 m/z.

It is appreciated that various isolation window sizes and modifications to the amount of overlap between sequential fragment ion scans will provide optimal results with the present invention using various mass spectrometers and sample types. This includes both isolation windows with fixed width for all MS2 scans (Gillet, L.C., et al. Mol. Cell Proteomics, 2012; 11: 1-17), and those with variable isolation widths (Navarro, P., et al. Nat. Biotechnol., 2016; 34: 1130-1136). In such implementations, as long as the isolation window boundaries are known, deconvolution of the resulting MS2 spectra can be applied.

Improvements to the intensity comparison steps in overlapping 102 and 104 sections as well as center 103 sections are within the scope of the present invention. It is optionally preferred that analytes containing larger precursor m/z values, and hence generally contain more atoms with natural isotopes such as carbon, have a maximum allowable intensity difference in 102 and 104 overlapping sections that expands according to precursor m/z size to account for larger intensity differences caused by more substantial natural isotope contributions.

The following example is provided to demonstrate an application of the invention, but is not be construed as the only type of application possible from the disclosure.

Working Example 1

An example using overlapping isolation windows and the deconvolution method to increase the number of detectable peptides and proteins in HeLa tryptic peptide lysate.

HeLa cell protein lysates were reduced with dithiothreitol and alkylated with iodoacetamide followed by trypsin proteolysis. The tryptic peptides were suspended in

0.1% formic acid and analyzed by liquid chromatography electrospray ionization tandem mass spectrometry (LC-MS/MS). Data were acquired using positive electrospray ionization via liquid junction into a Bruker Impact II QTOF mass spectrometer operating in DIA mode. The sequential MS2 scan inclusion list stepped

+4 m/z per scan with 6 m/z isolation windows centered at 433, 437, 441, 445, 449, 453,

457, 461, 465, 469, 473, 477, 481, 485, 489, 493, 497, 501, 505, 509, 513, 517, 521,

525, 529, 533, 537, 541, 545, 549, 553, 557, 561, 565, 569, 573, 577, 581, 585, 589,

593, 597, 601, 605, 609, 613, 617, 621, 625, 629, 633, 637, 641, 645, 649, 653, 657,

661, 665, 669, 673, 677, 681, 685, 689, 693, 697, 701, 705, 709, 713, 717, 721, 725,

729, 733, 737, 741, 745, 749, 753, 757, 761, 765, 769, 773, 777, 781, 785, 789, 793,

797, 801, 805, 809, 813, 817, 821, 825, 829, 833, 837, 841, 845, 849, 853, 857, 861, 865, 869, 873, 877, 881, 885, 889, 893, 897, 901, 905, and 909 m/z. A precursor scan was also acquired from 430-910 m/z. All scans were acquired with a 37 Hertz scan rate. Collision induced dissociation (CID) was used for peptide fragmentation. 500 nanograms of peptides were analyzed with a 30-minute linear acetonitrile gradient.

All data analysis was performed by the Vulcan Analytical software tool, Protalizer version 1.1.2917. Each fragment ion spectrum was deconvoluted by determining center 103 sections sandwiched between 102 and 104 overlapping sections. To determine fragment ions belonging to 103 center sections in scan N+l, MS2 peaks were matched in the preceding scan N and following scan N+2 within ± 0.015 m/z and those with various minimum fold intensity increases shown in Table 1 in scan N+l were retained in a reconstructed spectrum assigned a more specific isolation window of ± 1 m/z at the center of the original raw MS2 scan. Fragment ions at the flanking end of a raw isolation window are assigned to 102 and 104 overlapping sections when detected within ± 0.015 m/z with a maximum fold intensity difference shown in Table 1 in the raw MS2 scans that overlap. The 102 and 104 overlapping section fragment ions were each merged into reconstructed spectra with the intensity of the fragment ions summed together and isolation window adjusted to ± 1 m/z. Applying the deconvolution resulted in the following reconstructed isolation window ranges 434-436, 436-438, 438-440, 440-442, 442-444, 444-446, 446-448, 448-450, 450-452, 452-454, 454-456, 456-458, 458-460, 460-462, 462-464, 464-466, 466-468, 468-470, 470-472, 472-474, 474-476, 476-478, 478-480, 480-482, 482-484, 484-486, 486-488, 488-490, 490-492, 492-494, 494-496, 496-498, 498-500, 500-502, 502-504, 504-506, 506-508, 508-510, 510-512, 512-514, 514-516, 516-518, 518-520, 520-522, 522-524, 524-526, 526-528, 528-530, 530-532, 532-534, 534-536, 536-538, 538-540, 540-542, 542-544, 544-546, 546-548, 548-550, 550-552, 552-554, 554-556, 556-558, 558-560, 560-562, 562-564, 564-566, 566-568, 568-570, 570-572, 572-574, 574-576, 576-578, 578-580, 580-582, 582-584, 584-586, 586-588, 588-590, 590-592, 592-594, 594-596, 596-598, 598-600, 600-602, 602-604, 604-606, 606-608, 608-610, 610-612, 612-614, 614-616, 616-618, 618-620, 620-622, 622-624, 624-626, 626-628, 628-630, 630-632, 632-634, 634-636, 636-638, 638-640, 640-642, 642-644, 644-646, 646-648, 648-650, 650-652, 652-654, 654-656, 656-658, 658-660, 660-662, 662-664, 664-666, 666-668, 668-670, 670-672, 672-674, 674-676, 676-678, 678-680, 680-682, 682-684, 684-686, 686-688, 688-690, 690-692, 692-694, 694-696, 696-698, 698-700, 700-702, 702-704, 704-706, 706-708, 708-710, 710-712, 712-714, 714-716, 716-718, 718-720, 720-722, 722-724, 724-726, 726-728, 728-730, 730-732, 732-734, 734-736, 736-738, 738-740, 740-742, 742-744, 744-746, 746-748, 748-750, 750-752, 752-754, 754-756, 756-758, 758-760, 760-762, 762-764, 764-766, 766-768, 768-770, 770-772, 772-774, 774-776, 776-778, 778-780, 780-782, 782-784, 784-786, 786-788, 788-790, 790-792, 792-794, 794-796, 796-798, 798-800, 800-802, 802-804, 804-806, 806-808, 808-810, 810-812, 812-814, 814-816, 816-818, 818-820, 820-822, 822-824, 824-826, 826-828, 828-830, 830-832, 832-834, 834-836, 836-838, 838-840, 840-842, 842-844, 844-846, 846-848, 848-850, 850-852, 852-854, 854-856, 856-858, 858-860, 860-862, 862-864, 864-866, 866-868, 868-870, 870-872, 872-874, 874-876, 876-878, 878-880, 880-882, 882-884, 884-886, 886-888, 888-890, 890-892, 892-894, 894-896, 896-898, 898-900, 900-902, 902-904, 904-906, and 906- 908 m/z.

Elution based demultiplexing of MS2 spectra was then applied in a similar manner as described previously (Bern, M., et al. Anal. Chem., 2010; 82: 833). Spectra were then searched for isotopic clusters with an m/z shift of 1.003355 and 0.501677 from the monoisotopic peak corresponding to +1 and +2 fragment ions, respectively, with an m/z tolerance of ± 0.015 m/z. Isotopic peak series identified with these m/z differences were removed to deisotope MS2 spectra. The remaining monoisotopic fragment ions with isotope clusters meeting these criteria were assigned a charge state and the spectra were searched with a ± 0.015 m/z tolerance to identify peptides and source proteins in the forward and reversed human Swiss-Prot reference database downloaded March 17, 2015 containing (40,392 sequences including reversed decoys). Peptides with > 5 fragment ions were retained as candidate peptide-matches. Each candidate match was checked to determine if the theoretical precursor monoisotopic and first isotopic peak in the MSI scan with the most similar retention time were observed using a ± 0.015 m/z tolerance. The first isotopic peak was also required to have the expected m/z shift based on the precursor charge (+3 peptides had +0.33445 m/z, and +2 peptides +0.50167 m/z). Peptide-matches were then assigned an expectation score by:

For example, a peptide with b6, b7, b8, y3, y4, y8, y6 was assigned an expectation score of 0.02 assuming no match to a corresponding precursor in an MSI scan, however if a precursor was also detected the score was calculated as 0.0175. A cutoff filter then eliminated all matches with scores > 0.06. Peptide sequences specific to a single protein entry in the Swiss-Prot database were assigned to that protein. Razor peptides present in multiple proteins in the database were assigned to a single match with the greatest protein percent sequence coverage. In cases where the protein sequence coverage did not delineate a razor peptide assignment, due to multiple proteins with equal sequence coverage, the razor peptide was removed from the search results. The false-positive identification rate was calculated using the target-decoy strategy as described previously (Elias, J.E., et al. Nat. Methods, 2007; 4: 207-214).

Results showing the number of non-redundant peptides and proteins detected at ≤ 1% protein FDR are presented in Table 1 from analyses of the exact same file with various center and overlapping section intensity thresholds compared to not applying the deconvolution steps of the present invention indicated by 'NA'.

Table 1. The present invention allows identifying up to 4,121 (25%) more peptides and

777 (29%) more proteins than without deconvolution.

Notably, the most peptide and protein identifications were obtained using a 103 center section minimum intensity increase of 1.1 -fold coupled with a 102 and 104 overlapping section maximum intensity difference of 3-fold, resulting in 20,740 peptides and 3,440 proteins. In contrast, without using the deconvolution method resulted in only 16,619 peptide and 2,663 protein identifications. To demonstrate the correct assignment of an analyte into the expected precursor isolation window, FIG. 3. shows MS2 chromatograms of the peptide sequence YGWLAAPQAYVSEK with a theoretical precursor m/z of 791.90 in a +2 charge state from the human protein 1 ,4-alpha-glucan-branching enzyme using the deconvolution method intensity configuration in Table 1 that provided the most identifications, compared to the raw file without deconvolution. The MS2 chromatogram from this example shows the overlapping 102 section from 790-792 precursor isolation window reconstructed spectra that has substantially fewer background peaks compared to the raw MS2 chromatogram.

FIG. 4. shows a second example of a correctly assigned analyte into the expected precursor isolation window after deconvolution. MS2 chromatograms of the peptide sequence, ENAPAIIFIDEIDAIATK, with a theoretical precursor monoisotopic m/z of 648.68 in a +3 charge state from the human protein 26S protease regulatory subunit 6B with the most sensitive deconvolution intensity parameters shown in Table 1, compared to the raw file without deconvolution. The MS2 chromatogram from the deconvoluted center 102 section from the 648-650 precursor m/z isolation window contains less interfering background peaks than the raw MS2 chromatogram.

Claims

1. A method for mass spectrometry acquisition and deconvolution for characterizing analytes, said method comprising:

(a) subjecting analytes to mass spectrometry acquisition with overlapping precursor isolation windows containing distinct regions of overlap and non-overlap in fragment ion scans;

(b) identifying distinct fragment ion peaks based on a m/z similarity in the raw fragment ion scans that are present or absent in other overlapping isolation window scans;

(c) comparing an intensity of the fragment ion peaks present in the overlapping isolation windows from (b), and retaining a first subset of the fragment ion peaks with intensity differences less than 3 -fold in a reconstructed spectrum assigned a precursor m/z corresponding to the overlapping precursor range of the raw scan;

(d) comparing the intensity of fragment ion peaks present in other overlapping isolation window scans from (b), and retaining a second subset of the fragment ion peaks with intensity increases more than 1.1 -fold in a reconstructed spectrum assigned a precursor m/z corresponding to a non-overlapping precursor range of the raw scan; and

(e) retaining the fragment ion peaks absent in other overlapping isolation window scans from (b) in a reconstructed spectrum assigned a precursor m/z corresponding to a non-overlapping precursor range of the raw scan.

2. The method of claim 1, wherein the mass spectrometry acquisition is data independent acquisition.

3. The method of claim 1, wherein the m/z similarity is within an error tolerance that the mass spectrometer can detect the same fragment ion in multiple fragment ion scans.

4. The method of claim 3, wherein the m/z similarity is within 0.015 m/z for quadrupole time-of-flight mass spectrometers.

5. The method of claim 1, wherein the overlapping isolation window fragment ion scans are separated in time by less than half the full elution width of an analyte.

6. The method of claim 1, wherein deconvolution is used to increase the precursor ion specificity of fragment ions.

7. The method of claim 1, wherein characterizing is identifying or quantifying an analyte.

8. The method of claim 1, wherein the analytes are peptides, proteins, metabolites, lipids, organic, or inorganic molecules.

9. The method of claim 1, wherein the deconvolution steps are implemented by a software program.

10. A method for mass spectrometry acquisition and deconvolution for characterizing analytes, said method comprising:

(a) subjecting analytes to mass spectrometry acquisition with overlapping precursor isolation windows containing distinct regions of overlap in fragment ion scans;

(b) identifying distinct fragment ion peaks based on a m/z similarity in raw fragment ion scans that are present or absent in other overlapping isolation window scans; and

(c) comparing an intensity of the fragment ion peaks present in the overlapping isolation windows from (b), and retaining a subset of the fragment ion peaks with intensity differences less than 3-fold in a reconstructed spectrum assigned a precursor m/z corresponding to the overlapping precursor range of the raw scan.

11. The method of claim 10, wherein the mass spectrometry acquisition is data independent acquisition.

12. The method of claim 10, wherein the m/z similarity is within an error tolerance that the mass spectrometer can detect the same fragment ion in multiple fragment ion scans.

13. The method of claim 12, wherein the m/z similarity is within 0.015 m/z for quadrupole time-of-flight mass spectrometers.

14. The method of claim 10, wherein the overlapping isolation window fragment ion scans are separated in time by less than half the full elution width of an analyte.

15. The method of claim 10, wherein deconvolution is used to increase the precursor ion specificity of fragment ions.

16. The method of claim 10, wherein characterizing is identifying or quantifying an analyte.

17. The method of claim 10, wherein the analytes are peptides, proteins, metabolites, lipids, organic, or inorganic molecules.

18. The method of claim 10, wherein the deconvolution steps are implemented by a software program.