d2ssect 0.0.2

Calculate d2s scores from short reads

d2ssect

A tool to calculate d2s scores using short fastq reads This repo will test and benchmark the existing alignment-free tools and the improving versions.

The originally version of this pipeline including three big steps:

1. get jellyfish count results
2. calculate d2s using jellyfish dump results of every pair of samples
3. generate a matrix

Our goal is to integrate these three steps and try to increase the speed of d2s calculation.

Installation

1. Install dependencies
   • Jellyfish 2.3.0
   • python 3.8
   • pandas

Usage

Lets say we have a collection of fasta files corresponding to sequencing reads from samples that we want to compare with d2ssect. First count kmers in these files using jellyfish

for f in *.fasta;do jellyfish count -m 21 -s 10000000 $f -o ${f%.fasta}.jf ;done

Note that the command above will create a corresponding .jf file for every .fasta file in the current directory. By keeping the base names of the jf and fasta files identical we can then run d2ssect as follows;

python3 ../d2ssect/d2ssect/main.py -l *.jf -f *.fasta

Building from source

CC=g++ pip install .