Background: The observed mutations in cancer are the result of ~30 mutational processes, which stamp particular mutational signatures (MS). Nevertheless, it is still not clear which genomic alterations correlate to several MS. Here, a method to analyze associations of genomic data with MS is presented and applied to The Cancer Genome Atlas breast cancer data revealing promising associations.
Methods: The MS were discretized into clusters whose extremes were statistically associated with mutations, copy number, and gene expression data.
Results: Known associations for apolipoprotein B editing complex (APOBEC) and for BRCA1 and BRCA2 support the proposal. For BRCA1/2, mutations in ARAP3, three focal deletions, and one amplification were detected. Around 50 mutated genes for the two APOBEC signatures were identified including three kinesins (KIF13A, KIF1B, KIF4A), three ubiquitins (USP45, UBR4, UBR1), and two demethylases (KDM5B, KDM5C) among other genes also connected to DNA damage pathways. The results suggest novel roles for other genes currently not involved in DNA repair. The altered expression program was very high for the BRCA1/2 signature, high for APOBEC signature 13 clearly associated to immune response, and low for APOBEC signature 2. The remaining signatures show scarce associations.
Conclusion: Specific genetic alterations can be associated with particular MS.
Keywords: APOBEC; BRCA1; BRCA2; cancer genomics; demethylases; kinesins; mutational signatures; mutations; ubiquitins.
© 2019 The Authors. Molecular Genetics & Genomic Medicine published by Wiley Periodicals, Inc.