CWL workflow (bird_worklow) to generate count tables from alignment files and run BIRD. This includes the following sub-workflows / steps:

1. mpileups. Generate variant pileup counts from alignment files (BAMs) over locations (single basepair genomic coordinates).
2. combine_counts. Create a combined count table following the expected BIRD input file format.
3. bird_scatter_values. Add logic to divide the set of variants in order to parallelize and speed up running BIRD.
4. bird. BIRD CWL wrapper to identify allele-specific regulatory activity using a bayesian framework.

A BIRD Docker file and image are provided.