sharepg - analysing shared regions in pangenomes

Note

Want to contribute? Feel free to open a PR on an issue about a missing, buggy or incomplete feature!

Purpose

Tool to analyse shared sequences between populations in pangenomes. This small command-line tool aims to check regions in a graph that are shared by a collection of genomes, and that are not traversed by another collection.

It can extract:

• Nodes that matches a criterion
• Genomic intervals on each sequences
• Bubbles with intervals

Note

Bubbles detection is made possible thanks to BubbleGun, Fawaz Dabbaghie, Jana Ebler, Tobias Marschall, BubbleGun: enumerating bubbles and superbubbles in genome graphs, Bioinformatics, Volume 38, Issue 17, September 2022, Pages 4217–4219, https://doi.org/10.1093/bioinformatics/btac448

Installation

Installation is made using pip with the provided setup.py file.

git clone https://github.com/Tharos-ux/sharepg.git
git pull
pip install --upgrade pip
python -m pip install .

Once installed, command-line become available out-of-the-box. You can check everything went well by printing the manpage with sharepg -h.

Available commands

Let's assume you happen to have:

• a GFA graph, uncompressed or not
• a list A of paths/walks names you want to follow (g0, g1, g2)
• a list B of paths/walks names you want to avoid (g3, g4)
• a reference name (g0)
• you want to strictly have all paths/walks you follow and not any of the ones you want to avoid (threshold=1.)

# Extract nodes
sharepg disnodes graph.gfa -a g0 g1 g2 -b g3 g4 -r g0 -t 1 > disnodes.log
# Extract intervals
sharepg disintervals graph.gfa -a g0 g1 g2 -b g3 g4 -r g0 -t 1 > disintervals.log
# Extract bubbles
sharepg disbubbles graph.gfa -a g0 g1 g2 -b g3 g4 -r g0 -t 1 > disbubbles.log

Note

Threshold parameter is a float within range 0. to 1.; it represents the minimum proportion of genomes that needs to be in A and in the observed region and 1-threshold maximum proportion of genomes of B that can go through this same region. For instance, a threshold of 1. says that for a segment to be selected it needs to have all genomes of A and no genomes from B. A threshold of .7 says that at least 70% of A must go through the region and at most 30% of B can go through he region for it to be selected.