Compute a pairwise SNP distance matrix from one or two alignment(s)

• Motivation
• Install
  • cargo
  • conda
  • Precompiled binaries
  • homebrew
  • Container
  • Local
• Usage
  • Quick
  • Full
• Benchmark
• Contributing
• Cite

A key point of difference for psdm is the pairwise SNP distance between two alignment files. This is particularly beneficial if you have computed a SNP distance matrix for many samples already and want to update the distances with some new samples - without rerunning the analysis for all samples in the original file.

Another potential use is having two alignment files for the same samples but with sequences generated by different techniques. For example, if you have produced consensus sequences from SNP calls from Illumina and Nanopore and want to see how similar the Illumina-to-Nanopore (inter-technology) distances are - compared to the intra-technology distances.

Despite these motivations, psdm can still be used to compute a "traditional" pairwise SNP distance matrix for a single FASTA alignment file.

Prerequisite: rust toolchain (min. v1.55.0)

$ cargo install psdm

Prerequisite: conda (and bioconda channel correctly set up)

$ conda install psdm

tl;dr: Run the following snippet to download the latest binary for your system to the current directory and show the help menu.

version="0.3.0"
OS=$(uname -s)                                                                                                       
if [ "$OS" = "Linux" ]; then                                                                                         
    triple="x86_64-unknown-linux-musl"                                                                              
elif [ "$OS" = "Darwin" ]; then                                                                                        
    triple="x86_64-apple-darwin"                                                         
else                                                      
    echo "ERROR: $OS not a recognised operating system"
fi              
if [ -n "$triple" ]; then   
    URL="https://github.com/mbhall88/psdm/releases/download/${version}/psdm-${version}-${triple}.tar.gz"
    wget "$URL" -O - | tar -xzf -
    ./psdm-${version}-${triple}/psdm --help             
fi

These binaries do not require that you have the rust toolchain installed.

Currently, there are two pre-compiled binaries available:

• Linux kernel x86_64-unknown-linux-musl (works on most Linux distributions I tested)
• OSX kernel x86_64-apple-darwin (works for any post-2007 Mac)

An example of downloading one of these binaries using wget

$ version="0.3.0"
$ URL="https://github.com/mbhall88/psdm/releases/download/${version}/psdm-${version}-x86_64-unknown-linux-musl.tar.gz"
$ wget "$URL" -O - | tar -xzf -
$ ./psdm*/psdm --help

If these binaries do not work on your system please raise an issue and I will potentially add some additional target triples.

Prerequisite: homebrew

The homebrew installation is done via the homebrew-bio tap.

$ brew install brewsci/bio/psdm

Docker images are hosted at quay.io.

Prerequisite: singularity

$ URI="docker://quay.io/mbhall88/psdm"
$ singularity exec "$URI" psdm --help

The above will use the latest version. If you want to specify a version then use a tag (or commit) like so.

$ VERSION="0.3.0"
$ URI="docker://quay.io/mbhall88/psdm:${VERSION}"

Prerequisite: docker

$ docker pull quay.io/mbhall88/psdm
$ docker run quay.io/mbhall88/psdm psdm --help

You can find all the available tags on the quay.io repository.

Prerequisite: rust toolchain

$ git clone https://github.com/mbhall88/psdm.git
$ cd psdm
$ cargo build --release
$ target/release/psdm --help
# if you want to check everything is working ok
$ cargo test

aln1.fa

>s1
ABCDEFGH
>s2
aBN-XFnH
>s0
AbCdEfG-

$ psdm aln1.fa
,s1,s2,s0
s1,0,1,0
s2,1,0,1
s0,0,1,0

aln2.fa.gz

>s2
xXNNfoo=
>s5 description
AB-DEFGG

$ psdm aln1.fa aln2.fa.gz
,s1,s2,s0
s2,6,5,5
s5,1,2,0

The column names represent the first alignment file provided.

I'd like the sequences to be sorted by identifier in the output

$ psdm -s aln1.fa
,s0,s1,s2
s0,0,0,1
s1,0,0,1
s2,1,1,0

I want a tab-delimited (TSV) matrix instead of a comma-delimited (CSV) one

$ psdm -d "\t" aln1.fa
        s1      s2      s0
s1      0       1       0
s2      1       0       1
s0      0       1       0

The case of nucleotides matters to me - i.e., acgt is not the same as ACGT

$ psdm -c aln1.fa
,s1,s2,s0
s1,0,3,3
s2,3,0,5
s0,3,5,0

By default, psdm ignores N's and gaps (-). However, maybe you also want to ignore X's

$ psdm -e NX- aln1.fa
,s1,s2,s0
s1,0,0,0
s2,0,0,0
s0,0,0,0

Or maybe you don't want to ignore anything

$ psdm -e "" aln1.fa
,s1,s2,s0
s1,0,5,4
s2,5,0,8
s0,4,8,0

I'm impatient, use all the threads I have!

$ psdm -t 0 aln1.fa

Give me long-form output instead of a matrix

$ psdm -l aln1.fa
s1,s1,0
s1,s2,1
s1,s0,0
s2,s1,1
s2,s2,0
s2,s0,1
s0,s1,0
s0,s2,1
s0,s0,0

I'd like to know the progress of the pairwise comparisons

$ psdm -P -t 8 aln.fa
[2024-06-20T02:50:32Z INFO  psdm] Using 8 thread(s)
[2024-06-20T02:50:32Z INFO  psdm] Loading first alignment file...
[2024-06-20T02:50:38Z INFO  psdm] Loaded 200 sequences with length 4411532bp
[2024-06-20T02:50:38Z INFO  psdm] Calculating 20100 pairwise distances...
Progress: 50.00% (10050 / 20100)

Write the matrix to a file please

$ psdm -o dists.csv aln1.fa

$ psdm --help
psdm 0.3.0
Michael Hall <michael@mbh.sh>
Compute a pairwise SNP distance matrix from one or two alignment(s)

USAGE:
    psdm [OPTIONS] [ALIGNMENTS]...

ARGS:
    <ALIGNMENTS>...
            FASTA alignment file(s) to compute the pairwise distance for.

            Providing two files will compute the distances for all sequences in one file against all
            sequences from the other file - i.e., not between sequences in the same file. The first
            file will be the column names, while the second is the row names. The alignment file(s)
            can be compressed.

OPTIONS:
    -c, --case-sensitive
            Case matters - i.e., dist(a, A) = 1

    -d, --delim <DELIMITER>
            Delimiting character for the output table

            [default: ,]

    -e, --ignored-chars <IGNORED_CHARS>
            String of characters to ignore - e.g., `-e N-` -> dist(A, N) = 0 and dist(A, -) = 0

            Note, if using `--case-sensitive` the upper- and lower-case form of a character is
            needed. To not ignore any characters, use `-e ''` or `-e ""`

            [default: N-]

    -h, --help
            Print help information

    -l, --long
            Output as long-form ("melted") table

            By default the output is a N x N or N x M table where N is the number of sequences in
            the first alignment and M is the number of sequences in the (optional) second alignment.

    -o, --output <OUTPUT>
            Output file name [default: stdout]

    -P, --progress
            Show a progress bar

    -q, --quiet
            No logging (except progress info if `-P` is given)

    -s, --sort
            Sort the alignment(s) by ID

    -t, --threads <THREADS>
            Number of threads to use. Setting to 0 will use all available

            [default: 1]

    -V, --version
            Print version information

We benchmark against snp-dists (v0.8.2). snp-dists is a brilliant tool for computing SNP distance matrices and is the inspiration for psdm.

As snp-dists doesn't allow inter-alignment comparisons, we will benchmark the intra-alignment mode. We use an alignment file of 151 sequences with a length of 4411532bp.

The benchmark times were recorded with hyperfine with 10 runs for each tool/threads combination.

Contributions are always welcome. For changes to be accepted, they must pass the CI and coverage checks. These include:

• Code is formatted (cargo fmt).
• There are no compiler errors/warnings. cargo clippy --all-features --all-targets -- -D warnings
• Test code coverage has not reduced.

Please also add a succinct description of the contribution in the CHANGELOG.

psdm is archived at Zenodo.

@software{Hall2021psdm,
  author       = {Hall, Michael B.},
  title        = {{psdm: Compute a pairwise SNP distance matrix from 
                   one or two alignment(s)}},
  month        = nov,
  year         = 2021,
  publisher    = {Zenodo},
  version      = {0.3.0},
  doi          = {10.5281/zenodo.5706784},
  url          = {https://doi.org/10.5281/zenodo.5706785}
}