SIA: Selection Inference Using the Ancestral Recombination Graph

Abstract

Detecting signals of selection from genomic data is a central problem in population genetics. Coupling the rich information in the ancestral recombination graph (ARG) with a powerful and scalable deep learning framework, we developed a novel method to detect and quantify positive selection: Selection Inference using the Ancestral recombination graph (SIA). Built on a Long Short-Term Memory (LSTM) architecture, a particular type of a Recurrent Neural Network (RNN), SIA can be trained to explicitly infer a full range of selection coefficients, as well as the allele frequency trajectory and time of selection onset. We benchmarked SIA extensively on simulations under a European human demographic model, and found that it performs as well or better as some of the best available methods, including state-of-the-art machine-learning and ARG-based methods. In addition, we used SIA to estimate selection coefficients at several loci associated with human phenotypes of interest. SIA detected novel signals of selection particular to the European (CEU) population at the MC1R and ABCC11 loci. In addition, it recapitulated signals of selection at the LCT locus and several pigmentation-related genes. Finally, we reanalyzed polymorphism data of a collection of recently radiated southern capuchino seedeater taxa in the genus Sporophila to quantify the strength of selection and improved the power of our previous methods to detect partial soft sweeps. Overall, SIA uses deep learning to leverage the ARG and thereby provides new insight into how selective sweeps shape genomic diversity.

Ultra-fast genome-wide inference of pairwise coalescence times

“…When selection is recent, this causes an enrichment for recent coalescent times in the distribution of pairwise TMRCAs at positively selected loci. This skew can be used as a basis for a test for selection [11, 22].…”

“…The length of these chains, measured by the (scaled) number of generations until coalescence, is called the coalescence time , or the time to the most recent common ancestor (TMRCA). Coalescence times reflect the genealogical and genetic history of a sample, and therefore have been used in many population and statistical genetic analyses, such as reconstructing demographic histories [15, 25, 27, 28], detecting selection signatures [11, 21, 22], genome-wide association studies [20, 29], genotype imputation [29] and more.…”

Neural networks for self-adjusting mutation rate estimation when the recombination rate is unknown

“…Machine learning methods based on the raw genetic data will require more complex network architectures and could help to automatically extract this additional information. Recent advances in the development of machine learning methods for population genetics [ 22 , 23 , 35 – 37 ], indicate that expanding the toolbox of neural network-based inference tools to more complex estimation tasks and larger data sets is a promising approach.…”

Neural Networks for self-adjusting Mutation Rate Estimation when the Recombination Rate is unknown

“…Machine learning methods based on the raw genetic data will require more complex network archi-tectures and could help to automatically extract this additional information. Recent advances in the development of machine learning methods for population genetics (22, 23, 35–37), indicate that expanding the toolbox of neural network-based inference tools to more complex estimation tasks and larger data sets is a promising approach.…”