Distinguishing between recent balancing selection and incomplete sweep using deep neural networks

Abstract

1AbstractBalancing selection is an important adaptive mechanism underpinning a wide range of phenotypes. Despite its relevance, the detection of recent balancing selection from genomic data is challenging as its signatures are qualitatively similar to those left by ongoing positive selection. In this study we developed and implemented two deep neural networks and tested their performance to predict loci under recent selection, either due to balancing selection or incomplete sweep, from population genomic data. Specifically, we generated forward-intime simulations to train and test an artificial neural network (ANN) and a convolutional neural network (CNN). ANN received as input multiple summary statistics calculated on the locus of interest, while CNN was applied directly on the matrix of haplotypes. We found that both architectures have high accuracy to identify loci under recent selection. CNN generally outperformed ANN to distinguish between signals of balancing selection and incomplete sweep and was less affected by incorrect training data. We deployed both trained networks on neutral genomic regions in European populations and demonstrated a lower false positive rate for CNN than ANN. We finally deployed CNN within the MEFV gene region and identified several common variants predicted to be under incomplete sweep in a European population. Notably, two of these variants are functional changes and could modulate susceptibility to Familial Mediterranean Fever, possibly as a consequence of past adaptation to pathogens. In conclusion, deep neural networks were able to characterise signals of selection on intermediate-frequency variants, an analysis currently inaccessible by commonly used strategies.

Detecting adaptive introgression in human evolution using convolutional neural networks

“…In summary, we have shown that CNNs are a powerful approach to detecting adaptive introgression and can recover both known and novel selection candidates that were introduced via admixture. As in previous applications to other problems in the field (Sheehan & Song, 2016; Flagel et al ., 2018; Schrider & Kern, 2018; Villanea & Schraiber, 2019; Mondal et al ., 2019; Torada et al ., 2019; Isildak et al ., 2020), this exemplifies how deep learning can serve as a very powerful tool for population genetic inference. This type of technique may thus be a useful resource for future studies aiming to unravel our past history and that of other species, as statistical methodologies and computational resources continue to improve.…”

“…To overcome the need to compress data into summary statistics (which might miss important features) or solve complex analytical theory, deep learning techniques are increasingly becoming a popular solution to address problems in population genetics. These problems include the inference of demographic histories (Sheehan & Song, 2016;Flagel et al, 2018;Villanea & Schraiber, 2019;Mondal et al, 2019;Sanchez et al, 2020), admixture (Blischak et al, 2020), recombination (Chan et al, 2018;Flagel et al, 2018;Adrion et al, 2020b) and natural selection (Schrider & Kern, 2018;Sheehan & Song, 2016;Torada et al, 2019;Isildak et al, 2020). Deep learning is a branch of machine learning that relies on algorithms structured as multi-layered networks, which are trained using known relationships between the input data and the desired output.…”

“…They can be used for classification, prediction or data compression (Aggarwal et al ., 2018). Among the techniques in this field, convolutional neural networks (CNNs) are a family of methods originally designed for image recognition and segmentation (LeCun et al ., 1995; Krizhevsky et al ., 2012), which have been recently applied to population genetic data (Chan et al ., 2018; Flagel et al ., 2018; Torada et al ., 2019; Isildak et al ., 2020; Blischak et al ., 2020; Sanchez et al ., 2020). A CNN can learn complex spatial patterns from large datasets that may be informative for classification or prediction, using a series of linear operations known as convolutions, to compress the data into features that are useful for inference.…”

Understanding the Adaptive Evolutionary Histories of South American Ancient and Present-Day Populations via Genomics

“…These methods can be utilized with large genomic datasets and multiple populations, allowing for hypothesis testing between populations with different attributes. Moreover, the varied modeling techniques employed by these methods permits them to directly account for the expected correlation structure among summary statistics across genomes (such as with Trendsetter [ 59 ] and SURFDAWave [ 60 ]), as well as to even automatically estimate genomic features from raw genotype calls that yield the highest prediction accuracies (such as with the methods of Flagel et al [ 56 ], ImaGene [ 61 ], and BaSe [ 62 ]). One particular caveat of these methods is that an accurate demographic model is generally needed for the production of training data.…”

BalLeRMix +: mixture model approaches for robust joint identification of both positive selection and long-term balancing selection

“…The deluge of polymorphism data available from contemporary sequencing technologies has fueled interest in both method development (e.g. Bitarello et al , 2018 ; Cheng and DeGiorgio, 2019 , 2020 ; DeGiorgio et al , 2014 ; Isildak et al , 2021 ; Sheehan and Song, 2016 ; Siewert and Voight, 2017 , 2020 ) and empirical data analysis (e.g. Croze et al , 2017 ; Leffler et al , 2013 ; Teixeira et al , 2015 ) on balancing selection.…”