S/HIC: Robust Identification of Soft and Hard Sweeps Using Machine Learning

Abstract

Detecting the targets of adaptive natural selection from whole genome sequencing data is a central problem for population genetics. However, to date most methods have shown sub-optimal performance under realistic demographic scenarios. Moreover, over the past decade there has been a renewed interest in determining the importance of selection from standing variation in adaptation of natural populations, yet very few methods for inferring this model of adaptation at the genome scale have been introduced. Here we introduce a new method, S/HIC, which uses supervised machine learning to precisely infer the location of both hard and soft selective sweeps. We show that S/HIC has unrivaled accuracy for detecting sweeps under demographic histories that are relevant to human populations, and distinguishing sweeps from linked as well as neutrally evolving regions. Moreover, we show that S/HIC is uniquely robust among its competitors to model misspecification. Thus, even if the true demographic model of a population differs catastrophically from that specified by the user, S/HIC still retains impressive discriminatory power. Finally, we apply S/HIC to the case of resequencing data from human chromosome 18 in a European population sample, and demonstrate that we can reliably recover selective sweeps that have been identified earlier using less specific and sensitive methods.

Author Summary

The genetic basis of recent adaptation can be uncovered from genomic patterns of variation, which are perturbed in predictable ways when a beneficial mutation “sweeps” through a population. However, the detection of such “selective sweeps” is complicated by demographic events, such as population expansion, which can produce similar skews in genetic diversity. Here, we present a method for detecting selective sweeps that is remarkably powerful and robust to potentially confounding demographic histories. This method, called S/HIC, operates using a machine learning paradigm to combine many different features of population genetic variation, and examine their values across a large genomic region in order to infer whether a selective sweep has recently occurred near its center. S/HIC is also able to accurately distinguish between selection acting on de novo beneficial mutations (“hard sweeps”) and selection on previously standing variants (“soft sweeps”). We demonstrate S/HIC’s power on a variety of simulated datasets as well as human population data wherein we recover several previously discovered targets of recent adaptation as well as a novel selective sweep.