Weighted likelihood inference of genomic autozygosity patterns in dense genotype data

摘要: Background: Genomic regions of autozygosity (ROA) arise when an individual is homozygous for haplotypes inherited identical-by-descent from ancestors shared by both parents. Over the past decade, they have gained importance understanding evolutionary history and genetic basis complex diseases traits. However, methods to detect ROA in dense genotype data not evolved step with advances genome technology that now enable us rapidly create large high-resolution datasets, limiting our ability investigate their constituent patterns. Results: We report a weighted likelihood approach identifying accounts autocorrelation among genotyped positions possibilities unobserved mutation recombination events, variability confidence calls whole sequence (WGS) data. Forward-time simulations under two demographic scenarios reflect situations where inbreeding its effect on fitness are interest suggest this better powered than existing state-of-the-art at marker densities consistent WGS popular microarray genotyping platforms used human non-human studies. Moreover, we present evidence suggests able distinguish arising via consanguinity endogamy. Using subsets The 1000 Genomes Project Phase 3 show that, relative WGS, intermediate long captured robustly platforms, while detection short more variable improves density. Worldwide patterns inferred found accord well those previously reported Finally, highlight potential genomic enriched signals one group another based upon comparisons per-individual likelihoods instead frequencies. Conclusions: This can assist population- disease-geneticists working wide variety types species explore identify differential groups, thereby advancing role recessive variation phenotypic disease.