Detecting Positive Selection in Populations Using Genetic Data

摘要: High-throughput genomic sequencing allows to disentangle the evolutionary forces acting in populations. Among forces, positive selection has received a lot of attention because it is related adaptation populations their environments, both biotic and abiotic. Positive selection, also known as Darwinian occurs when an allele favored by natural selection. The frequency increases population and, due genetic hitchhiking, neighboring linked variation diminishes, creating so-called selective sweeps. Such process leaves traces genomes that can be detected future time point. Detecting achieved searching for signatures introduced sweeps, such regions reduced variation, specific shift site spectrum, particular linkage disequilibrium (LD) patterns region. A variety approaches used detecting ranging from simple implementations compute summary statistics more advanced statistical approaches, e.g., Bayesian maximum-likelihood-based methods, machine learning methods. In this chapter, we discuss sweep detection methodologies on basis capacity analyze whole or just subgenomic regions, polymorphism they exploit signatures. We summarize results comparisons among five open-source software releases (SweeD, SweepFinder, SweepFinder2, OmegaPlus, RAiSD) regarding sensitivity, specificity, execution times. Furthermore, test methods present thorough performance analysis. equilibrium neutral models mild bottlenecks, most are able detect sweeps accurately. Methods tools rely rather than single SNPs exhibit higher true rates spectrum (SFS)-based under model recurrent hitchhiking. However, false rate elevated misspecified demographic build distribution statistic null hypothesis. Both LD SFS-based suffer decreased accuracy localizing target bottleneck scenarios. extensive analysis effects gene flow detection, problem been understudied literature.