Analysis of cancer somatic mutations taking into consideration human genetic variations

Cancer cells accumulate somatic mutations during clonal evolution, but not all the somatic mutations contribute to cancer progression. To identify real cancer related mutations, background or passenger mutations need to be filtered out. This work clusters cancer somatic mutations while taking into account human genetic variations for three types (breast, lung, and kidney) of cancer in TCGA project. Then the biological significance of the clusters was validated with Gene Ontology, biological function, and cancer pathway and cell signaling pathway analysis. Results indicate that clusters composed of variants having low frequencies in the 1000 genomes data but high frequencies in cancer patients are more closely associated with cancer than clusters composed of variants having high frequencies in both 1000 genomes data and cancer patients. Moreover, genes that were clustered based on only their frequencies in human populations and in cancer patients have similar biological functions within the same clusters, supporting the validity of the clustering. This work shows that human genetic variation is an important factor that needs to be taken into account when filtering cancer somatic mutations to determine causal variants.