Learning the properties of adaptive regions with functional data analysis

Efficient Detection and Characterization of Targets of Natural Selection Using Transfer Learning

Natural selection leaves detectable patterns of altered spatial diversity within genomes, and identifying affected regions is crucial for understanding species evolution. Recently, machine learning approaches applied to raw population genomic data have been developed to uncover these adaptive signatures. Convolutional neural networks (CNNs) are particularly effective for this task, as they handle large data arrays while maintaining element correlations. However, shallow CNNs may miss complex patterns due to their limited capacity, while deep CNNs can capture these patterns but require extensive data and computational power. Transfer learning addresses these challenges by utilizing a deep CNN pretrained on a large dataset as a feature extraction tool for downstream classification and evolutionary parameter prediction. This approach reduces extensive training data generation requirements and computational needs while maintaining high performance. In this study, we developed TrIdent, a tool that uses transfer learning to enhance detection of adaptive genomic regions from image representations of multilocus variation. We evaluated TrIdent across various genetic, demographic, and adaptive settings, in addition to unphased data and other confounding factors. TrIdent demonstrated improved detection of adaptive regions compared to recent methods using similar data representations. We further explored model interpretability through class activation maps and adapted TrIdent to infer selection parameters for identified adaptive candidates. Using whole-genome haplotype data from European and African populations, TrIdent effectively recapitulated known sweep candidates and identified novel cancer, and other disease-associated genes as potential sweeps.

Efficient detection and characterization of targets of natural selection using transfer learning

Natural selection leaves detectable patterns of altered spatial diversity within genomes, and identifying affected regions is crucial for understanding species evolution. Recently, machine learning approaches applied to raw population genomic data have been developed to uncover these adaptive signatures. Convolutional neural networks (CNNs) are particularly effective for this task, as they handle large data arrays while maintaining element correlations. However, shallow CNNs may miss complex patterns due to their limited capacity, while deep CNNs can capture these patterns but require extensive data and computational power. Transfer learning addresses these challenges by utilizing a deep CNN pre-trained on a large dataset as a feature extraction tool for downstream classification and evolutionary parameter prediction. This approach reduces extensive training data generation requirements and computational needs while maintaining high performance. In this study, we developed TrIdent, a tool that uses transfer learning to enhance detection of adaptive genomic regions from image representations of multilocus variation. We evaluated TrIdent across various genetic, demographic, and adaptive settings, in addition to unphased data and other confounding factors. TrIdent demonstrated improved detection of adaptive regions compared to recent methods using similar data representations. We further explored model interpretability through class activation maps and adapted TrIdent to infer selection parameters for identified adaptive candidates. Using whole-genome haplotype data from European and African populations, TrIdent effectively recapitulated known sweep candidates and identified novel cancer, and other disease-associated genes as potential sweeps.

Data preprocessing methods for selective sweep detection using convolutional neural networks

The identification of positive selection has been framed as a classification task, with Convolutional Neural Networks (CNNs) already outperforming summary statistics and likelihood-based approaches in accuracy. Despite the prevalence of CNN-based methods that manipulate the pixels of images representing raw genomic data as a preprocessing step to improve classification accuracy, the efficacy of these pixel-rearrangement techniques remains inadequately examined, particularly in the presence of confounding factors like population bottlenecks, migration and recombination hotspots. We introduce a set of pixel rearrangement algorithms aimed at enhancing CNN classification accuracy in detecting selective sweeps. These algorithms are employed to assess the performance of four CNN models for selective sweep detection. Our findings illustrate that the judicious application of rearrangement algorithms notably enhances the overall classification accuracy of a CNN across various datasets simulating confounding factors. We observed that sorting the columns of the genomic matrices has higher on CNN performance than rearranging the sequences. To some extent, these rearrangement algorithms are more robust to misspecified demographic models compared with the utilization of the default preprocessing algorithm as suggested by the respective authors of each CNN architecture. We provide the data rearrangement algorithms as a distinct package available for download at: https://github.com/Zhaohq96/Genetic-data-rearrangement.

Digital Image Processing to Detect Adaptive Evolution

In recent years, advances in image processing and machine learning have fueled a paradigm shift in detecting genomic regions under natural selection. Early machine learning techniques employed population-genetic summary statistics as features, which focus on specific genomic patterns expected by adaptive and neutral processes. Though such engineered features are important when training data are limited, the ease at which simulated data can now be generated has led to the recent development of approaches that take in image representations of haplotype alignments and automatically extract important features using convolutional neural networks. Digital image processing methods termed α-molecules are a class of techniques for multiscale representation of objects that can extract a diverse set of features from images. One such α-molecule method, termed wavelet decomposition, lends greater control over high-frequency components of images. Another α-molecule method, termed curvelet decomposition, is an extension of the wavelet concept that considers events occurring along curves within images. We show that application of these α-molecule techniques to extract features from image representations of haplotype alignments yield high true positive rate and accuracy to detect hard and soft selective sweep signatures from genomic data with both linear and nonlinear machine learning classifiers. Moreover, we find that such models are easy to visualize and interpret, with performance rivaling those of contemporary deep learning approaches for detecting sweeps.

Tree Sequences as a General-Purpose Tool for Population Genetic Inference

As population genetic data increase in size, new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks applied to population genetic alignments. To better utilize these new data structures, we propose and implement a graph convolutional network to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard convolutional neural network approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a graph convolutional network approach and can be used to perform well on these common population genetic inference tasks with accuracies roughly matching or even exceeding that of a convolutional neural network-based method. As tree sequences become more widely used in population genetic research, we foresee developments and optimizations of this work to provide a foundation for population genetic inference moving forward.

Population Genomic Scans for Natural Selection and Demography

Uncovering the fundamental processes that shape genomic variation in natural populations is a primary objective of population genetics. These processes include demographic effects such as past changes in effective population size or gene flow between structured populations. Furthermore, genomic variation is affected by selection on nonneutral genetic variants, for example, through the adaptation of beneficial alleles or balancing selection that maintains genetic variation. In this article, we discuss the characterization of these processes using population genetic models, and we review methods developed on the basis of these models to unravel the underlying processes from modern population genomic data sets. We briefly discuss the conditions in which these approaches can be used to infer demography or identify specific nonneutral genetic variants and cases in which caution is warranted. Moreover, we summarize the challenges of jointly inferring demography and selective processes that affect neutral variation genome-wide.

Tree sequences as a general-purpose tool for population genetic inference

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Assessing the Presence of Recent Adaptation in the Human Genome With Mixture Density Regression

How much genome differences between species reflect neutral or adaptive evolution is a central question in evolutionary genomics. In humans and other mammals, the presence of adaptive versus neutral genomic evolution has proven particularly difficult to quantify. The difficulty notably stems from the highly heterogeneous organization of mammalian genomes at multiple levels (functional sequence density, recombination, etc.) which complicates the interpretation and distinction of adaptive versus neutral evolution signals. In this study, we introduce mixture density regressions (MDRs) for the study of the determinants of recent adaptation in the human genome. MDRs provide a flexible regression model based on multiple Gaussian distributions. We use MDRs to model the association between recent selection signals and multiple genomic factors likely to affect the occurrence/detection of positive selection, if the latter was present in the first place to generate these associations. We find that an MDR model with two Gaussian distributions provides an excellent fit to the genome-wide distribution of a common sweep summary statistic (integrated haplotype score), with one of the two distributions likely enriched in positive selection. We further find several factors associated with signals of recent adaptation, including the recombination rate, the density of regulatory elements in immune cells, GC content, gene expression in immune cells, the density of mammal-wide conserved elements, and the distance to the nearest virus-interacting gene. These results support the presence of strong positive selection in recent human evolution and highlight MDRs as a powerful tool to make sense of signals of recent genomic adaptation.

Tensor Decomposition-based Feature Extraction and Classification to Detect Natural Selection from Genomic Data

Inferences of adaptive events are important for learning about traits, such as human digestion of lactose after infancy and the rapid spread of viral variants. Early efforts toward identifying footprints of natural selection from genomic data involved development of summary statistic and likelihood methods. However, such techniques are grounded in simple patterns or theoretical models that limit the complexity of settings they can explore. Due to the renaissance in artificial intelligence, machine learning methods have taken center stage in recent efforts to detect natural selection, with strategies such as convolutional neural networks applied to images of haplotypes. Yet, limitations of such techniques include estimation of large numbers of model parameters under nonconvex settings and feature identification without regard to location within an image. An alternative approach is to use tensor decomposition to extract features from multidimensional data although preserving the latent structure of the data, and to feed these features to machine learning models. Here, we adopt this framework and present a novel approach termed T-REx, which extracts features from images of haplotypes across sampled individuals using tensor decomposition, and then makes predictions from these features using classical machine learning methods. As a proof of concept, we explore the performance of T-REx on simulated neutral and selective sweep scenarios and find that it has high power and accuracy to discriminate sweeps from neutrality, robustness to common technical hurdles, and easy visualization of feature importance. Therefore, T-REx is a powerful addition to the toolkit for detecting adaptive processes from genomic data.

Uncovering Footprints of Natural Selection Through Spectral Analysis of Genomic Summary Statistics

Natural selection leaves a spatial pattern along the genome, with a haplotype distribution distortion near the selected locus that fades with distance. Evaluating the spatial signal of a population-genetic summary statistic across the genome allows for patterns of natural selection to be distinguished from neutrality. Considering the genomic spatial distribution of multiple summary statistics is expected to aid in uncovering subtle signatures of selection. In recent years, numerous methods have been devised that consider genomic spatial distributions across summary statistics, utilizing both classical machine learning and deep learning architectures. However, better predictions may be attainable by improving the way in which features are extracted from these summary statistics. We apply wavelet transform, multitaper spectral analysis, and S-transform to summary statistic arrays to achieve this goal. Each analysis method converts one-dimensional summary statistic arrays to two-dimensional images of spectral analysis, allowing simultaneous temporal and spectral assessment. We feed these images into convolutional neural networks and consider combining models using ensemble stacking. Our modeling framework achieves high accuracy and power across a diverse set of evolutionary settings, including population size changes and test sets of varying sweep strength, softness, and timing. A scan of central European whole-genome sequences recapitulated well-established sweep candidates and predicted novel cancer-associated genes as sweeps with high support. Given that this modeling framework is also robust to missing genomic segments, we believe that it will represent a welcome addition to the population-genomic toolkit for learning about adaptive processes from genomic data.

Genome-wide scans for selective sweeps using convolutional neural networks

MOTIVATION

Recent methods for selective sweep detection cast the problem as a classification task and use summary statistics as features to capture region characteristics that are indicative of a selective sweep, thereby being sensitive to confounding factors. Furthermore, they are not designed to perform whole-genome scans or to estimate the extent of the genomic region that was affected by positive selection; both are required for identifying candidate genes and the time and strength of selection.

RESULTS

We present ASDEC (https://github.com/pephco/ASDEC), a neural-network-based framework that can scan whole genomes for selective sweeps. ASDEC achieves similar classification performance to other convolutional neural network-based classifiers that rely on summary statistics, but it is trained 10× faster and classifies genomic regions 5× faster by inferring region characteristics from the raw sequence data directly. Deploying ASDEC for genomic scans achieved up to 15.2× higher sensitivity, 19.4× higher success rates, and 4× higher detection accuracy than state-of-the-art methods. We used ASDEC to scan human chromosome 1 of the Yoruba population (1000Genomes project), identifying nine known candidate genes.

Tensor decomposition based feature extraction and classification to detect natural selection from genomic data

Inferences of adaptive events are important for learning about traits, such as human digestion of lactose after infancy and the rapid spread of viral variants. Early efforts toward identifying footprints of natural selection from genomic data involved development of summary statistic and likelihood methods. However, such techniques are grounded in simple patterns or theoretical models that limit the complexity of settings they can explore. Due to the renaissance in artificial intelligence, machine learning methods have taken center stage in recent efforts to detect natural selection, with strategies such as convolutional neural networks applied to images of haplotypes. Yet, limitations of such techniques include estimation of large numbers of model parameters under non-convex settings and feature identification without regard to location within an image. An alternative approach is to use tensor decomposition to extract features from multidimensional data while preserving the latent structure of the data, and to feed these features to machine learning models. Here, we adopt this framework and present a novel approach termed T-REx , which extracts features from images of haplotypes across sampled individuals using tensor decomposition, and then makes predictions from these features using classical machine learning methods. As a proof of concept, we explore the performance of T-REx on simulated neutral and selective sweep scenarios and find that it has high power and accuracy to discriminate sweeps from neutrality, robustness to common technical hurdles, and easy visualization of feature importance. Therefore, T-REx is a powerful addition to the toolkit for detecting adaptive processes from genomic data.

A spatially aware likelihood test to detect sweeps from haplotype distributions

The inference of positive selection in genomes is a problem of great interest in evolutionary genomics. By identifying putative regions of the genome that contain adaptive mutations, we are able to learn about the biology of organisms and their evolutionary history. Here we introduce a composite likelihood method that identifies recently completed or ongoing positive selection by searching for extreme distortions in the spatial distribution of the haplotype frequency spectrum along the genome relative to the genome-wide expectation taken as neutrality. Furthermore, the method simultaneously infers two parameters of the sweep: the number of sweeping haplotypes and the "width" of the sweep, which is related to the strength and timing of selection. We demonstrate that this method outperforms the leading haplotype-based selection statistics, though strong signals in low-recombination regions merit extra scrutiny. As a positive control, we apply it to two well-studied human populations from the 1000 Genomes Project and examine haplotype frequency spectrum patterns at the LCT and MHC loci. We also apply it to a data set of brown rats sampled in NYC and identify genes related to olfactory perception. To facilitate use of this method, we have implemented it in user-friendly open source software.

A Likelihood Approach for Uncovering Selective Sweep Signatures from Haplotype Data

Selective sweeps are frequent and varied signatures in the genomes of natural populations, and detecting them is consequently important in understanding mechanisms of adaptation by natural selection. Following a selective sweep, haplotypic diversity surrounding the site under selection decreases, and this deviation from the background pattern of variation can be applied to identify sweeps. Multiple methods exist to locate selective sweeps in the genome from haplotype data, but none leverages the power of a model-based approach to make their inference. Here, we propose a likelihood ratio test statistic T to probe whole-genome polymorphism data sets for selective sweep signatures. Our framework uses a simple but powerful model of haplotype frequency spectrum distortion to find sweeps and additionally make an inference on the number of presently sweeping haplotypes in a population. We found that the T statistic is suitable for detecting both hard and soft sweeps across a variety of demographic models, selection strengths, and ages of the beneficial allele. Accordingly, we applied the T statistic to variant calls from European and sub-Saharan African human populations, yielding primarily literature-supported candidates, including LCT, RSPH3, and ZNF211 in CEU, SYT1, RGS18, and NNT in YRI, and HLA genes in both populations. We also searched for sweep signatures in Drosophila melanogaster, finding expected candidates at Ace, Uhg1, and Pimet. Finally, we provide open-source software to compute the T statistic and the inferred number of presently sweeping haplotypes from whole-genome data.

Learning Retention Mechanisms and Evolutionary Parameters of Duplicate Genes from Their Expression Data

Learning about the roles that duplicate genes play in the origins of novel phenotypes requires an understanding of how their functions evolve. A previous method for achieving this goal, CDROM, employs gene expression distances as proxies for functional divergence and then classifies the evolutionary mechanisms retaining duplicate genes from comparisons of these distances in a decision tree framework. However, CDROM does not account for stochastic shifts in gene expression or leverage advances in contemporary statistical learning for performing classification, nor is it capable of predicting the parameters driving duplicate gene evolution. Thus, here we develop CLOUD, a multi-layer neural network built on a model of gene expression evolution that can both classify duplicate gene retention mechanisms and predict their underlying evolutionary parameters. We show that not only is the CLOUD classifier substantially more powerful and accurate than CDROM, but that it also yields accurate parameter predictions, enabling a better understanding of the specific forces driving the evolution and long-term retention of duplicate genes. Further, application of the CLOUD classifier and predictor to empirical data from Drosophila recapitulates many previous findings about gene duplication in this lineage, showing that new functions often emerge rapidly and asymmetrically in younger duplicate gene copies, and that functional divergence is driven by strong natural selection. Hence, CLOUD represents a major advancement in classifying retention mechanisms and predicting evolutionary parameters of duplicate genes, thereby highlighting the utility of incorporating sophisticated statistical learning techniques to address long-standing questions about evolution after gene duplication.

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

The South American continent is remarkably diverse in its ecological zones, spanning the Amazon rainforest, the high-altitude Andes, and Tierra del Fuego. Yet the original human populations of the continent successfully inhabited all these zones, well before the buffering effects of modern technology. Therefore, it is likely that the various cultures were successful, in part, due to positive natural selection that allowed them to successfully establish populations for thousands of years. Detecting positive selection in these populations is still in its infancy, as the ongoing effects of European contact have decimated many of these populations and introduced gene flow from outside of the continent. In this review, we explore hypotheses of possible human biological adaptation, methods to identify positive selection, the utilization of ancient DNA, and the integration of modern genomes through the identification of genomic tracts that reflect the ancestry of the first populations of the Americas.