How to generate Panel of Normals for the pipeline?

[YingYa]

Hello,

I would like to use the pipeline to call somatic mutations from tumor-only RNAseq data. Would you mind to detail that how to generate the Panel of Normals?

Thanks,
Ya

[pablo-gar]

Hi Ya,

In this study I used two types of PoN

1. Tissue-specific PoN
2. Whole-data PoN

Those were constructed as follows:

Tissue-specific PoN eliminates all mutations that occur at least 40% across samples of the same tissue
Whole-data PoN eliminates all mutations that occur at least 4% across samples of all tissues

You can calculate the frequency of unique mutations from a mutation map file using this shell command. A mutation map has to have at least these tab-separated columns: chromosome, position, reference base, mutation. For example:
chr1 25451781 G A