Computational methods for analysis of somatic background mutational processes in cancer

Authors

• A Goncearenco
• AL Brown
• M Li
• C Cunningham
• IB Rogozin
• AR Panchenko

Abstract

Mutational rate and patterns in the DNA sequence context of somatic mutations are highly dependent on the responsible mutagen. We derived statistical models for background mutagenic processes, including exposure to chemical agents, enzymes, radiation, and infidelity of DNA replication and repair machinery from the sequences of multiple tumor samples. These models of background mutagenesis are used to identify mutations that drive cancer development and progression. The combinations of mutational processes detected in tumor samples help to distinguish cancer subtypes, particularly in highly heterogeneous cancers. We developed novel computational methods that can handle particularly difficult cases with low numbers of mutations and mixed etiology. These methods were implemented in a freely available online resource MutaGene https://www.ncbi.nlm.nih.gov/projects/mutagene/. MutaGene simplifies analysis of user-submitted tumor samples and provides an annotated collection of cancer-type specific mutational profiles, mutational signatures and motifs of various mutational processes.

Scientific Focus Area: Computational Biology

This page was last updated on Friday, March 26, 2021