Annotations new features

[Marcel-Mueck]

What

The new update to the annotation pipeline includes all annotation tools used for the published version of deeprvat, as well as all processing steps needed to create ready-to-run annotation data for the published deeprvat version.

Updates since last push:

• inclusion of AlphaMissense annotations as plugin in VEP
• changed DAG in order to optimize parallelisation
• Allele Frequency, MAF and AF_MB calculated from genotype file
• pipeline includes filtering step for variants based on distance to the nearest exon
• selection of required columns for deeprvat as well as renaming of columns to fit required input is now performed
• abSplice pipeline is now integrated into the DAG
• spliceAI scores are aggregated to spliceAI_delta_sores

Testing

• since the annotation tools rely on pre-scored data, automatic tests via GitHub actions is not completely possible
• the annotation pipeline was instead tested locally using example data

[meyerkm]

@Marcel-Mueck There seems to be duplicate functions in annotations.py for the following functions:

• deepripe_score_variant_onlyseq_all
• merge_abscore
• process_annotations

Also, looks like the Read-the-docs build is failing as the docstrings aren't in the correct Sphinx format. See here to adapt: https://sphinx-rtd-tutorial.readthedocs.io/en/latest/docstrings.html