To identify genes that may be relevant to breast cancer, we looked for those that harboured multiple recurrent or inactivating mutations, as these are mutation patterns typical of oncogenes and tumour suppressors. Recurrent mutations were defined as missense SNVs and in-frame substitutions that affected the same codon of the annotation transcript, whereas inactivating mutations included nonsense SNVs, frameshift substitutions and variants that affected splice sites. The proportions of recurrent (ONC) and inactivating (TSG) mutations for each gene (out of the total number of mutations) were computed, and a threshold of 0.2 was used (20/20 rule). Genes with an ONC score >0.2 and with a TSG score >0.05 were classified as tumour suppressors. A minimum of five recurrent or inactivating mutations was required for a gene to be selected as putative drivers. The method was adapted from the study by Vogelstein et al.16 (link)
Full text: Click here