Run SComatic with a matched normal single-cell RNA-seq data

[gongyuTang123]

I am trying to detect somatic mutations from my single-cell RNA-seq data for lung cancer samples. I knew SComatic relies on different cell types' data to remove possible germline mutations and artifacts. But I happened to have the matched normal single-cell data isolated from the same cancer patient. How could these data be used to help generate better results by SComatic?

Should I use the 'scripts/PoN/PoN.py' to generate a novel PON for each patient or for each cell subpopulation and use it as a reference to remove possible germline mutations?

[Francesc-Muyas]

Dear user,
Sorry for not getting back to you sooner. The question that you bring is quite interesting and can have different ways of being approached:

1. You should first run the full SComatic workflow in the normal sample. Next, take the unfiltered output of the Step4.2 of this matched normal sample and used it as --pon in the Step4.2 of the tumour sample.
2. Alternatively, compute the PoN script using the basecalling.step1.tsv file obtained for the normal sample and set the parameter --min_samples 1. You can then use this PoN for the tumour sample.

I hope it helps,
Fran