Feature/create pon

[anoronh4]

---

•  Draft PR      *
  

---

Addressing #126. New workflow mirroring steps in the main pipeline to generate PoN as per the instructions here: https://ccstempo.netlify.app/wes-panel-of-normals.html#strelka2. It capitalizes on the resources, configurations and libraries already existing for similar steps in the main pipeline. It is not meant to be run very often or in step with the main workflow, which is why we are not replicating extra functions like the watch mode.

Currently it takes in the --bamMapping file, in the same format that the main workflow takes. However we may want to consider adding another column to indicate sequencer type (Novaseq vs Hiseq, for example).

Remaining steps include annotation of normal calls. This is necessary to look at certain features, such as clonal hematopoiesis mutations, to decide if the germline sample should be included in the final PoN.

Also, variants that are not present in a minimum number of samples should be filtered out (gatk best practices suggests variants present in less than 2 samples should be filtered out). This can be achieved using bcftools isec

[anoronh4]

To-do:
Annotate with databases like gnomAD and oncokb in order to flag common germline variants and pathogenic variants. We should aim to isolate artifacts foremost, because this panel of normals will not replace a matched normal, just provide an additional flag.