A Snakemake workflow for neural posterior estimation in population genetics.

The main workflow for this package is contained in workflow/training_workflow.smk. This will run a complete neural posterior estimation workflow based on a given config file. For more details, see the docs.

Current environment is set up with conda. To install the environment, run

conda env create -f environment.yaml

To run the workflow, you first need to create a config file containing the details of the simulation and inference you want to perform. An example of such a config file is contained in workflow/config/AraTha_2epoch_cnn.yaml. This config file is used to run the neural posterior estimation of a two-epoch demographic model with a CNN embedding network.

To run the workflow, you can use the following command:

snakemake --configfile workflow/config/AraTha_2epoch_cnn.yaml --snakefile workflow/training_workflow.smk

Example data for the VCF prediction pipeline may be created via

python resources/util/simulate-vcf.py \
  --outpath "example_data/AraTha_2epoch" \
  --window-size 1000000 \
  --configfile workflow/config/AraTha_2epoch_cnn.yaml

To generate predictions along windows in the VCF,

snakemake --configfile workflow/config/AraTha_2epoch_cnn.yaml --snakefile workflow/prediction_workflow.smk

TODO: more details on input files, when this is finalized

To run the workflow on a cluster, you should first create a profile config file for you cluster. We've provided an example profile in example_profile/config.yaml, which is meant to be used with the SLURM cluster at University of Oregon.

To use this profile, you can run the following command:

snakemake --executor slurm --configfile workflow/config/AraTha_2epoch_cnn.yaml --workflow-profile ~/.config/snakemake/yourprofile/ --snakefile workflow/training_workflow.smk

note: The profile used in the workflow assumes snakemake version 8.X. If you are using a different version, you may have to change the profile.