Tracing the evolution of the human mutation rate

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): All genetic variation is created by mutations, changes that arise due to DNA damage or copying mistakes during DNA replication. Mutations are frequent enough that, on average, a child's 3-billion base pair genome contains 74 new genetic variants that are not present in the genome of either parent. Such new mutations confer a higher disease risk than older mutations because they have not passed the test of surviving through several generations of parents and offspring. We aim to pinpoint how the human mutation rate has evolved as humans left Africa and adapted to diverse new environments across the globe. One specific aim will follow up on my preliminary research which showed that Europeans experienced a mutation rate change after diverging from Africans and Asians. The primary evidence for this change is that European genomes have a higher burden than African or Asian genomes of the mutation type TCC→TTC, where the trinucleotide "TCC" has experienced a mutation from "C" to "T" at its central site. We wish to and the genetic basis of this mutation rate change by looking at rare variants in mixed- ancestry Latino and African-American individuals. Specially, we will isolate young genetic variants that probably arose via mutation within the past 10-15 generations, after gene ow from Europe into the Americas had already begun. We will infer the genetic background (European, African, or Native American) upon which each new mutation arose and look for genomic regions where European ances- try correlates strongly with an excess of TCC→TTC mutations. These will be the regions most likely to harbor a causal allele that changed the process of mutation accumulation in Europeans. This work has the potential to yield valuable insights into melanoma, a cancer that predominantly affects individuals of European ancestry and whose somatic mutational signature is dominated by TCC→TTC. A second specific aim is to look for other signatures of mutation rate change that have occurred within the human species or, more broadly, within the great apes. We will use a natural language processing technique called Latent Dirichlet Allocation (LDA) to identify collections of mutation types whose rates appear to be under common genetic control. A few mutation types besides TCC→TTC show weak signals of rate differentiation between populations, and we will attempt to infer how many separate mutation rate change events are necessary to explain these signals. The admixture mapping technique from Specific Aim I can also be adapted to interrogate the genetic basis of other mutation rate changes that might have occurred in the recent past. These efforts should improve our understanding of the human mutation rate's genetic architecture and how mutation rates differ between populations.