The somatic variants / mutations that you have found have likely already been identified by some other study (think of TCGA and ICGC).

You can try one or more of the following:

1. use cBioPortal to search for mutations in genes that have already been identified by TCGA
2. overlap your mutations with the COSMIC database
3. download the open access TCGA Mutation Annotation Format (MAF) files from Genomic Data Commons (GDC) for each TCGA cancer, curate these files into ready-to-use datasets, and then overlap these with your own data
4. check the functionality of your variants with one of many algorithms ( many which I list here: A: pathogenicity predictors of cancer mutations ) - GWAVA and Funseq2 are tailoured for somatic variants

Kevin

You can take a look to what genes are related to a disease with some databases or tools, here are some:

• Clinvar
• Gemini
• Exomiser (if you have phenotypical data)