FineMAV: prioritizing candidate genetic variants driving local adaptations in human populations

Michał Szpak,Qasim Ayub,Yuan Chen,Yali Xue,Chris Tyler-Smith,Massimo Mezzavilla

doi:10.1186/s13059-017-1380-2

Abstract

We present a new method, Fine-Mapping of Adaptive Variation (FineMAV), which combines population differentiation, derived allele frequency, and molecular functionality to prioritize positively selected candidate variants for functional follow-up. We calibrate and test FineMAV using eight experimentally validated “gold standard” positively selected variants and simulations. FineMAV has good sensitivity and a low false discovery rate. Applying FineMAV to the 1000 Genomes Project Phase 3 SNP dataset, we report many novel selected variants, including ones in TGM3 and PRSS53 associated with hair phenotypes that we validate using available independent data. FineMAV is widely applicable to sequence data from both human and other species.

Highlights

The out-of-Africa expansion ~ 60,000 years ago exposed humans to a diverse range of new environments and selective pressures including new pathogens, climatic conditions, and diets [1,2,3]
It has been argued that hard sweeps were rather rare in recent human evolution [2, 22] and that selection may more often operate on pre-existing variation that has evolved neutrally in the population until it becomes advantageous under certain conditions (“selection on standing variation”) [2, 4, 22]
Fine-Mapping of Adaptive Variation (FineMAV) combines a new measure of population differentiation, a measure of allele prevalence (DAF), and a measure of functionality

Summary

Introduction

The out-of-Africa expansion ~ 60,000 years ago exposed humans to a diverse range of new environments and selective pressures including new pathogens, climatic conditions, and diets [1,2,3]. Most methods that have been developed to detect signals of recent and ongoing positive selection are based on the classical hard sweep model [2, 22]. This model assumes that a new advantageous mutation rapidly spreads to fixation or high frequency, purging nearby linked variation due to genetic hitchhiking [2, 20, 23]. It has been argued that hard sweeps were rather rare in recent human evolution [2, 22] and that selection may more often operate on pre-existing variation that has evolved neutrally in the population until it becomes advantageous under certain conditions (“selection on standing variation”) [2, 4, 22]

Objectives

Methods

Results

Discussion

Conclusion