Poster #58: Diljeet Kaur
Interaction between mutational processes and coding sequences drive enrichments of specific loss-of-function mutations in tumor suppressor genes |
Diljeet Kaur, BA, Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA; Ziyue Gao, PhD, Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA |
The occurrence of driver mutations varies across cancer types and among tumors of the same type. Although these differences are largely attributed to variation in somatic selection and stochasticity, they may also reflect differences in the baseline mutation probabilities across tumors. Here, we tested whether mutational processes differ in their propensity and sequence-context preferences in generating loss-of-function (LoF) mutations in tumor suppressor genes. By enumerating all possible LoF mutations in 85 tumor suppressor genes, we quantified their mutation susceptibility to each COSMIC single-base-substitution mutational signature. We found that stop-gain and splice-disrupting mutations are both concentrated in a limited subset of trinucleotide mutation types, with stop-gain mutations particularly enriched for C>A and C>T substitutions, reflecting the structure of the genetic code. Furthermore, mutational signatures differ substantially in their propensity to generate LoF mutations, with SBS10a, SBS10d, and SBS36 showing the highest stop-gain propensities due to predominance of C>A mutations in their profiles. Consistent with our predictions, POLE exonuclease domain mutant colorectal and endometrial tumors were enriched for stop-gain mutations occurring in codon contexts preferentially targeted by the POLE-associated signatures SBS10a and SBS10b. Together, our findings show that mutational processes shape not only the mutation rates of cancer driver genes but also the identities of the specific mutations generated, thereby contributing to recurrent, tissue-specific driver mutation hotspots. |


Comments