[TCGA] Statistical analysis of somatic variant calling

Question

[umn_bist]

I am a undergraduate student assisting in cancer research. I am hoping to learn and become a better computational biologist.

I have tumor and matched normal RNA-seq samples from TCGA to call and annotate somatic variants. What more analysis do you think can be conducted next? What analysis can I conduct on the detected somatic variants? I should mention that I am trying to find mutation "patterns" across multiple cancer types.

I was also wondering in what direction I should expand from variant calling. I was considering differential gene expression but I would appreciate additional downstream analysis ideas.

Thank you very, very much for your time and help.

Answer 1

[TriS]

I think you might want to start by looking into the TCGA Pan cancer analysis here, here and here.

then, you CAN call mutations from RNASeq, however, it is commonly done with exome sequencing data.

lastly, since you need access to get the bam files, are you using the MAF files ? if so, the variant calls are already done and you can annotate with i.e. Annovar or SnpEff