The frequency of somatic mutations in cancer predicts the phenotypic relevance of germline mutations.

ABSTRACT: Genomic sequence mutations can be pathogenic in both germline and somatic cells. Several authors have observed that often the same genes are involved in cancer when mutated in somatic cells and in genetic diseases when mutated in the germline. Recent advances in high-throughput sequencing techniques have provided us with large databases of both types of mutations, allowing us to investigate this issue in a systematic way. Hence, we applied a machine learning based framework to this problem, comparing multiple models. The models achieved significant predictive power as shown by both cross-validation and their application to recently discovered gene/phenotype associations not used for training. We found that genes characterized by high frequency of somatic mutations in the most common cancer

SUBMITTER: Draetta EL 

PROVIDER: S-EPMC9868957 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature