stacksPairwise

Calculate pairwise sequence divergence between samples from Stacks RAD genotyping output. Simply supply the path to a samples.fa file generated from a Stacks run and a text file with individuals to calculate divergence between.

Pairwise divergence is estimated seperately per-RAD tag. It is calculated as the average divergence between haplotypes from different samples. Within sample diversity (i.e., heterozygosity) is also calculated.

Requirements

• Biopython

Install

To install the program + biopython, run:

git clone https://github.com/gibsonMatt/stacks-pairwise.git
cd stacks-pairwise
python setup.py install

Usage

usage: stacksPairwise [-h] [-names] [-o] samples

Calculate pairwise divergence (pairwise pi) from Stacks `samples.fa` output
fle

positional arguments:
  samples            Path to `samples.fa` file (from Stacks output)

optional arguments:
  -h, --help         show this help message and exit
  -names , --names   Names of samples to analyze. Either a text file or comma
                     seperated list.
  -o , --outputdir   Output directory/prefix

Inputs

samples.fa file

To generate a samples.fa file, you inclcude the --fasta-samples flag to the populations program in Stacks. For example:

populations -P ./stacks_output/ --fasta-samples

Note that no internal filtering is applied, so any filtering should be done prior to calculating pairwise divergence.

Sample list

A text file containing the samples to be compared. One name per line.

Outputs

Four files will be written:

1. .diffs.csv, containing the average number of nucleotide differences for each RAD tag
2. .sites.csv, containing the total number of sites (per locus) for which each pairwise comparision has data
3. .estimates.csv, containing the estimated per-site divergence (per locus) for each pairwise comparison
4. .summary.txt, containing genome-wide summaries