FastaDist 1.0.1

Package to calculate a distance matrix from a multiple sequence file

FastaDist

Github repository

This small utility package will calculate number of differences between all samples in a fasta alignment file. It will count any position where there is a G,A,T or C (case insensitive) in both sequences that differ as 1 SNV.

Output formats are a square distance matrix in tsv, csv or phylip formats It is fast since it first converts sequences to bit arrays and then uses fast bit operations to calculate the differences.

On a mid-range laptop a distance matrix was produced in 11 minutes from a 764 sequence alignment of length 1,082,859 using -p 1 and 4.5 minutes with -p 4

Installation

FastaDist is available as PyPi package for Python3

pip3 install fastadist

Usage

usage: fastadist [-h] -i ALIGNMENT_FILEPATH [-t TREE_FILEPATH] -o
                 OUTPUT_FILEPATH [-f FORMAT] [-p PARALLEL_PROCESSES] [-v]

    A script to calculate distances by converting sequences to bit arrays.
    Specify number of processes as -p N to speed up the calculation


optional arguments:
  -h, --help            show this help message and exit
  -i ALIGNMENT_FILEPATH, --alignment_filepath ALIGNMENT_FILEPATH
                        path to multiple sequence alignment input file
  -t TREE_FILEPATH, --tree_filepath TREE_FILEPATH
                        path to newick tree for distance matrix ordering
  -o OUTPUT_FILEPATH, --output_filepath OUTPUT_FILEPATH
                        path to distance matrix output file
  -f FORMAT, --format FORMAT
                        output format for distance matrix (one of tsv
                        [default], csv and phylip
  -p PARALLEL_PROCESSES, --parallel_processes PARALLEL_PROCESSES
                        number of parallel processes to run (default 1)
  -v, --version         print out software version