MSpangepop is a workflow for simulating pangenome variation graphs from coalescent simulations.
A simplified description of the algorithm can be found here.

The official MSpangepop repository can be found at the INRAE forge.
A GitHub mirror can be found at INRAE GitHub.
The mirror is especially useful for people with no Renater account to submit issues.

Input Files Requirements and specifications for input data Master Configuration How to set up configuration files and parameters Demographic model Adapt or create a model Output Files Description of generated results and outputs Visualizations Understanding the generated plots and charts Advanced Topics In-depth information for power users

Clone the Git repository

git clone https://forge.inrae.fr/pangepop/MSpangepop 

• Create an environement for snakemake (from the provided envfile):

conda env create -n wf_env -f dependencies/wf_env.yaml

Two elements are needed to run the simulation :

• The masterconfig -> Master Configuration
• The demographic_file -> Demographic model configuration

Edit the masterconfig file in the .config/ directory with your sample information. (Master Configuration)

nano .config/masterconfig.yaml

Example config:

samples:
  test_run:
    model: "simulation_data/Panmictic_Model.json"
    replicates: 1

• model is the demographic scenario the simulation will run on. You can create your own or tailor the ones in ./simulation_data (Demographic model configuration)

• ⚠️ Don't want to create your own model?⚠️ Use the provided Panmictic_Model.json - simply edit it to specify your genome, then adjust mutation_rate and recombination_rate (start with low values)

• Run the workflow :

sbatch mspangepop dry # Check for warnings
sbatch mspangepop run # Then

Nb : If your account name can't be automatically determined, add it in the .config/snakemake/profiles/slurm/config.yaml file.

Nb : Use the command squeue --format="%.10i %.9P %.6j %.10k %.8u %.2t %.10M %.6D %.20R" -A $user to see job names

./mspangepop dry # Check for warnings
./mspangepop local-run # Then

mspangepop [dry|run|local-run|dag|rulegraph|unlock|touch] [additional snakemake args]
    dry - run in dry-run mode
    run - run the workflow with SLURM
    local-run - run the workflow localy (on a single node)
    dag - generate the directed acyclic graph for the workflow
    rulegraph - generate the rulegraph for the workflow
    unlock - Unlock the directory if snakemake crashed
    touch - Tell snakemake that all files are up to date (use with caution)
    [additional snakemake args] - for any snakemake arg, like --until hifiasm

MSpangepop implements graph path operations to add variants by modifying how lineages traverse the graph.

Core Features:

• Multi-path targeting - Operations apply to single or multiple lineage paths simultaneously, enabling both unique and shared variants
• Orientation-aware - All operations preserve node directionality using edges that track exit and entry node sides, creating orientation-aware links (++, +-, -+, --)
• Composable - Operations can be nested and overlapping (e.g., deletion within inversion), representing complex compound variants

These operations modify paths through existing nodes rather than altering the graph structure, maintaining shared sequences while creating alternative routes for different lineages. New nodes (e.g., for insertions) are generated using an order 1 Markov model to produce realistic sequences.

MSpangepop is developed at INRAE as part of the PangenOak project.