Bayesian inference of recombination rates and hotspots using population genomic data

As more large-scale human genomic data become available, recombination rate variation can be inferred on a genome-wide scale. Several statistical methods for estimating fine-scale recombination rates using population samples have been developed. However, current statistical methods to infer recombination rates that can be applied to moderate, or large, genomic regions are limited to approximated-likelihoods. We developed a full-likelihood Markov Chain Monte Carlo method for estimating recombination rate under a Bayesian framework. Genealogies underlying a sampling of chromosomes are effectively modeled by using marginal individual SNP genealogies related through an ancestral recombination graph. Simulation studies show that our method performs well for different simulation scenarios.;Sperm-typing studies have revealed that recombination hotspots are a general feature of the human genome. We developed a new model of recombination hotspots taking account of the variations in background recombination rates across regions. The probability model was inspired by the observed patterns of recombination at several genomic regions analyzed in sperm-typing studies. Posterior probabilities and Bayes factors of recombination hotspots and background rates along chromosomes are inferred. Different criteria for identifying hotspots can be used to adjust false positive rates and power.;Simulation analyses show that our method can accurately estimate the variation in recombination rates across genomic regions. In particular, clusters of hotspots can be distinguished even though weaker hotspots are present. The method was applied to SNP data from the human leukocyte antigen (HLA) region, the MS32, and chromosome 19.