VarFish Workflow: Mass Queries

This Snakemake-based workflow allows mass queries that even works with large VarFish projects.

Installing

As a prerequisite, install conda and mamba package. Activate conda.

$ mamba env create -n varfish-wf-queries --file environment.yaml
$ conda activte varfish-wf-queries

Configuring

First of all, ensure that ~/.varfishrc.toml is properly configured, e.g.

[global]
varfish_server_url = "https://varfish.example.com"
varfish_api_token = "1234567890123456789012345678901234567890123456789012345678901234"

Then, copy config/main.yml.example to config/main.yml. Adjust the settings:

• varfish_shortcuts
  • query_presets - the query presets, chose one of de_novo, dominant, recessive, mitochondrial
• varfish_projects list of
  • name - name of the project for folders
  • uuid - UUID of project in VarFish
• genes - optional; list of gene symbols (or entrez/ensembl IDs) to restrict the search for - null for no restriction
• max_cases - opional; number of cases to limit query to - null for no restriction

Running

# snakemake --configfile=config/main.yml --cores 10 -d workflow

Tricks

If anything gets stuck, use the following to for re-running queries:

$ for x in $(find workflow/translate-namse -name query.json); do y=${x/.json/_result.json}; test -e $y || rm -f $x; done