Cross-species analysis of enhancer logic using deep learning

Large-scale genomic analysis of the domestic dog informs biological discovery

Recent advances in genomics, coupled with a unique population structure and remarkable levels of variation, have propelled the domestic dog to new levels as a system for understanding fundamental principles in mammalian biology. Central to this advance are more than 350 recognized breeds, each a closed population that has undergone selection for unique features. Genetic variation in the domestic dog is particularly well characterized compared with other domestic mammals, with almost 3000 high-coverage genomes publicly available. Importantly, as the number of sequenced genomes increases, new avenues for analysis are becoming available. Herein, we discuss recent discoveries in canine genomics regarding behavior, morphology, and disease susceptibility. We explore the limitations of current data sets for variant interpretation, tradeoffs between sequencing strategies, and the burgeoning role of long-read genomes for capturing structural variants. In addition, we consider how large-scale collections of whole-genome sequence data drive rare variant discovery and assess the geographic distribution of canine diversity, which identifies Asia as a major source of missing variation. Finally, we review recent comparative genomic analyses that will facilitate annotation of the noncoding genome in dogs.

Multiplex profiling of developmental cis-regulatory elements with quantitative single-cell expression reporters

The inability to scalably and precisely measure the activity of developmental cis-regulatory elements (CREs) in multicellular systems is a bottleneck in genomics. Here we develop a dual RNA cassette that decouples the detection and quantification tasks inherent to multiplex single-cell reporter assays. The resulting measurement of reporter expression is accurate over multiple orders of magnitude, with a precision approaching the limit set by Poisson counting noise. Together with RNA barcode stabilization via circularization, these scalable single-cell quantitative expression reporters provide high-contrast readouts, analogous to classic in situ assays but entirely from sequencing. Screening >200 regions of accessible chromatin in a multicellular in vitro model of early mammalian development, we identify 13 (8 previously uncharacterized) autonomous and cell-type-specific developmental CREs. We further demonstrate that chimeric CRE pairs generate cognate two-cell-type activity profiles and assess gain- and loss-of-function multicellular expression phenotypes from CRE variants with perturbed transcription factor binding sites. Single-cell quantitative expression reporters can be applied in developmental and multicellular systems to quantitatively characterize native, perturbed and synthetic CREs at scale, with high sensitivity and at single-cell resolution.

Controllers of histone methylation-modifying enzymes in gastrointestinal cancers

Gastrointestinal (GI) cancers have been considered primarily genetic malignancies, caused by a series of progressive genetic alterations. Accumulating evidence shows that histone methylation, an epigenetic modification program, plays an essential role in the different pathological stages of GI cancer progression, such as precancerous lesions, tumorigenesis, and tumor metastasis. Histone methylation-modifying enzymes, including histone methyltransferases (HMTs) and demethylases (HDMs), are the main executor of post-transcriptional modification. The abnormal expression of histone methylation-modifying enzymes characterizes GI cancers with complex pathogenesis and progression. Interactions between upstream controllers and histone methylation-modifying enzymes have recently been revealed, and have provided numerous opportunities to elucidate the pathogenesis of GI cancers in depth and clearly. Here we focus on the association between histone methylation-modifying enzymes and their controllers, aiming to provide a new perspective on the molecular research and clinical management of GI cancers.

Improving the performance of supervised deep learning for regulatory genomics using phylogenetic augmentation

MOTIVATION

Supervised deep learning is used to model the complex relationship between genomic sequence and regulatory function. Understanding how these models make predictions can provide biological insight into regulatory functions. Given the complexity of the sequence to regulatory function mapping (the cis-regulatory code), it has been suggested that the genome contains insufficient sequence variation to train models with suitable complexity. Data augmentation is a widely used approach to increase the data variation available for model training, however current data augmentation methods for genomic sequence data are limited.

RESULTS

Inspired by the success of comparative genomics, we show that augmenting genomic sequences with evolutionarily related sequences from other species, which we term phylogenetic augmentation, improves the performance of deep learning models trained on regulatory genomic sequences to predict high-throughput functional assay measurements. Additionally, we show that phylogenetic augmentation can rescue model performance when the training set is down-sampled and permits deep learning on a real-world small dataset, demonstrating that this approach improves data efficiency. Overall, this data augmentation method represents a solution for improving model performance that is applicable to many supervised deep-learning problems in genomics.

AVAILABILITY AND IMPLEMENTATION

The open-source GitHub repository agduncan94/phylogenetic_augmentation_paper includes the code for rerunning the analyses here and recreating the figures.

Construction of single-cell cross-species chromatin accessibility landscapes with combinatorial-hybridization-based ATAC-seq

Despite recent advances in single-cell genomics, the lack of maps for single-cell candidate cis-regulatory elements (cCREs) in non-mammal species has limited our exploration of conserved regulatory programs across vertebrates and invertebrates. Here, we developed a combinatorial-hybridization-based method for single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq) named CH-ATAC-seq, enabling the construction of single-cell accessible chromatin landscapes for zebrafish, Drosophila, and earthworms (Eisenia andrei). By integrating scATAC censuses of humans, monkeys, and mice, we systematically identified 152 distinct main cell types and around 0.8 million cell-type-specific cCREs. Our analysis provided insights into the conservation of neural, muscle, and immune lineages across species, while epithelial cells exhibited a higher organ-origin heterogeneity. Additionally, a large-scale gene regulatory network (GRN) was constructed in four vertebrates by integrating scRNA-seq censuses. Overall, our study provides a valuable resource for comparative epigenomics, identifying the evolutionary conservation and divergence of gene regulation across different species.

Cell-type-directed design of synthetic enhancers

Transcriptional enhancers act as docking stations for combinations of transcription factors and thereby regulate spatiotemporal activation of their target genes1. It has been a long-standing goal in the field to decode the regulatory logic of an enhancer and to understand the details of how spatiotemporal gene expression is encoded in an enhancer sequence. Here we show that deep learning models2-6, can be used to efficiently design synthetic, cell-type-specific enhancers, starting from random sequences, and that this optimization process allows detailed tracing of enhancer features at single-nucleotide resolution. We evaluate the function of fully synthetic enhancers to specifically target Kenyon cells or glial cells in the fruit fly brain using transgenic animals. We further exploit enhancer design to create 'dual-code' enhancers that target two cell types and minimal enhancers smaller than 50 base pairs that are fully functional. By examining the state space searches towards local optima, we characterize enhancer codes through the strength, combination and arrangement of transcription factor activator and transcription factor repressor motifs. Finally, we apply the same strategies to successfully design human enhancers, which adhere to enhancer rules similar to those of Drosophila enhancers. Enhancer design guided by deep learning leads to better understanding of how enhancers work and shows that their code can be exploited to manipulate cell states.

Single-cell spatial multi-omics and deep learning dissect enhancer-driven gene regulatory networks in liver zonation

In the mammalian liver, hepatocytes exhibit diverse metabolic and functional profiles based on their location within the liver lobule. However, it is unclear whether this spatial variation, called zonation, is governed by a well-defined gene regulatory code. Here, using a combination of single-cell multiomics, spatial omics, massively parallel reporter assays and deep learning, we mapped enhancer-gene regulatory networks across mouse liver cell types. We found that zonation affects gene expression and chromatin accessibility in hepatocytes, among other cell types. These states are driven by the repressors TCF7L1 and TBX3, alongside other core hepatocyte transcription factors, such as HNF4A, CEBPA, FOXA1 and ONECUT1. To examine the architecture of the enhancers driving these cell states, we trained a hierarchical deep learning model called DeepLiver. Our study provides a multimodal understanding of the regulatory code underlying hepatocyte identity and their zonation state that can be used to engineer enhancers with specific activity levels and zonation patterns.

Artificial intelligence for neurodegenerative experimental models

INTRODUCTION

Experimental models are essential tools in neurodegenerative disease research. However, the translation of insights and drugs discovered in model systems has proven immensely challenging, marred by high failure rates in human clinical trials.

METHODS

Here we review the application of artificial intelligence (AI) and machine learning (ML) in experimental medicine for dementia research.

RESULTS

Considering the specific challenges of reproducibility and translation between other species or model systems and human biology in preclinical dementia research, we highlight best practices and resources that can be leveraged to quantify and evaluate translatability. We then evaluate how AI and ML approaches could be applied to enhance both cross-model reproducibility and translation to human biology, while sustaining biological interpretability.

DISCUSSION

AI and ML approaches in experimental medicine remain in their infancy. However, they have great potential to strengthen preclinical research and translation if based upon adequate, robust, and reproducible experimental data.

HIGHLIGHTS

There are increasing applications of AI in experimental medicine. We identified issues in reproducibility, cross-species translation, and data curation in the field. Our review highlights data resources and AI approaches as solutions. Multi-omics analysis with AI offers exciting future possibilities in drug discovery.

Predictive analyses of regulatory sequences with EUGENe

Deep learning has become a popular tool to study cis-regulatory function. Yet efforts to design software for deep-learning analyses in regulatory genomics that are findable, accessible, interoperable and reusable (FAIR) have fallen short of fully meeting these criteria. Here we present elucidating the utility of genomic elements with neural nets (EUGENe), a FAIR toolkit for the analysis of genomic sequences with deep learning. EUGENe consists of a set of modules and subpackages for executing the key functionality of a genomics deep learning workflow: (1) extracting, transforming and loading sequence data from many common file formats; (2) instantiating, initializing and training diverse model architectures; and (3) evaluating and interpreting model behavior. We designed EUGENe as a simple, flexible and extensible interface for streamlining and customizing end-to-end deep-learning sequence analyses, and illustrate these principles through application of the toolkit to three predictive modeling tasks. We hope that EUGENe represents a springboard towards a collaborative ecosystem for deep-learning applications in genomics research.

Landscape of enhancer disruption and functional screen in melanoma cells

BACKGROUND

The high mutation rate throughout the entire melanoma genome presents a major challenge in stratifying true driver events from the background mutations. Numerous recurrent non-coding alterations, such as those in enhancers, can shape tumor evolution, thereby emphasizing the importance in systematically deciphering enhancer disruptions in melanoma.

RESULTS

Here, we leveraged 297 melanoma whole-genome sequencing samples to prioritize highly recurrent regions. By performing a genome-scale CRISPR interference (CRISPRi) screen on highly recurrent region-associated enhancers in melanoma cells, we identified 66 significant hits which could have tumor-suppressive roles. These functional enhancers show unique mutational patterns independent of classical significantly mutated genes in melanoma. Target gene analysis for the essential enhancers reveal many known and hidden mechanisms underlying melanoma growth. Utilizing extensive functional validation experiments, we demonstrate that a super enhancer element could modulate melanoma cell proliferation by targeting MEF2A, and another distal enhancer is able to sustain PTEN tumor-suppressive potential via long-range interactions.

CONCLUSIONS

Our study establishes a catalogue of crucial enhancers and their target genes in melanoma growth and progression, and illuminates the identification of novel mechanisms of dysregulation for melanoma driver genes and new therapeutic targeting strategies.

A comparative atlas of single-cell chromatin accessibility in the human brain

Recent advances in single-cell transcriptomics have illuminated the diverse neuronal and glial cell types within the human brain. However, the regulatory programs governing cell identity and function remain unclear. Using a single-nucleus assay for transposase-accessible chromatin using sequencing (snATAC-seq), we explored open chromatin landscapes across 1.1 million cells in 42 brain regions from three adults. Integrating this data unveiled 107 distinct cell types and their specific utilization of 544,735 candidate cis-regulatory DNA elements (cCREs) in the human genome. Nearly a third of the cCREs demonstrated conservation and chromatin accessibility in the mouse brain cells. We reveal strong links between specific brain cell types and neuropsychiatric disorders including schizophrenia, bipolar disorder, Alzheimer's disease (AD), and major depression, and have developed deep learning models to predict the regulatory roles of noncoding risk variants in these disorders.

Chromatin accessibility in the Drosophila embryo is determined by transcription factor pioneering and enhancer activation

Chromatin accessibility is integral to the process by which transcription factors (TFs) read out cis-regulatory DNA sequences, but it is difficult to differentiate between TFs that drive accessibility and those that do not. Deep learning models that learn complex sequence rules provide an unprecedented opportunity to dissect this problem. Using zygotic genome activation in Drosophila as a model, we analyzed high-resolution TF binding and chromatin accessibility data with interpretable deep learning and performed genetic validation experiments. We identify a hierarchical relationship between the pioneer TF Zelda and the TFs involved in axis patterning. Zelda consistently pioneers chromatin accessibility proportional to motif affinity, whereas patterning TFs augment chromatin accessibility in sequence contexts where they mediate enhancer activation. We conclude that chromatin accessibility occurs in two tiers: one through pioneering, which makes enhancers accessible but not necessarily active, and the second when the correct combination of TFs leads to enhancer activation.

SCENIC+: single-cell multiomic inference of enhancers and gene regulatory networks

Joint profiling of chromatin accessibility and gene expression in individual cells provides an opportunity to decipher enhancer-driven gene regulatory networks (GRNs). Here we present a method for the inference of enhancer-driven GRNs, called SCENIC+. SCENIC+ predicts genomic enhancers along with candidate upstream transcription factors (TFs) and links these enhancers to candidate target genes. To improve both recall and precision of TF identification, we curated and clustered a motif collection with more than 30,000 motifs. We benchmarked SCENIC+ on diverse datasets from different species, including human peripheral blood mononuclear cells, ENCODE cell lines, melanoma cell states and Drosophila retinal development. Next, we exploit SCENIC+ predictions to study conserved TFs, enhancers and GRNs between human and mouse cell types in the cerebral cortex. Finally, we use SCENIC+ to study the dynamics of gene regulation along differentiation trajectories and the effect of TF perturbations on cell state. SCENIC+ is available at scenicplus.readthedocs.io .

Decoding enhancer complexity with machine learning and high-throughput discovery

Enhancers are genomic DNA elements controlling spatiotemporal gene expression. Their flexible organization and functional redundancies make deciphering their sequence-function relationships challenging. This article provides an overview of the current understanding of enhancer organization and evolution, with an emphasis on factors that influence these relationships. Technological advancements, particularly in machine learning and synthetic biology, are discussed in light of how they provide new ways to understand this complexity. Exciting opportunities lie ahead as we continue to unravel the intricacies of enhancer function.

Relating enhancer genetic variation across mammals to complex phenotypes using machine learning

Protein-coding differences between species often fail to explain phenotypic diversity, suggesting the involvement of genomic elements that regulate gene expression such as enhancers. Identifying associations between enhancers and phenotypes is challenging because enhancer activity can be tissue-dependent and functionally conserved despite low sequence conservation. We developed the Tissue-Aware Conservation Inference Toolkit (TACIT) to associate candidate enhancers with species' phenotypes using predictions from machine learning models trained on specific tissues. Applying TACIT to associate motor cortex and parvalbumin-positive interneuron enhancers with neurological phenotypes revealed dozens of enhancer-phenotype associations, including brain size-associated enhancers that interact with genes implicated in microcephaly or macrocephaly. TACIT provides a foundation for identifying enhancers associated with the evolution of any convergently evolved phenotype in any large group of species with aligned genomes.

Comparative single cell epigenomic analysis of gene regulatory programs in the rodent and primate neocortex

Sequence divergence of cis- regulatory elements drives species-specific traits, but how this manifests in the evolution of the neocortex at the molecular and cellular level remains to be elucidated. We investigated the gene regulatory programs in the primary motor cortex of human, macaque, marmoset, and mouse with single-cell multiomics assays, generating gene expression, chromatin accessibility, DNA methylome, and chromosomal conformation profiles from a total of over 180,000 cells. For each modality, we determined species-specific, divergent, and conserved gene expression and epigenetic features at multiple levels. We find that cell type-specific gene expression evolves more rapidly than broadly expressed genes and that epigenetic status at distal candidate cis -regulatory elements (cCREs) evolves faster than promoters. Strikingly, transposable elements (TEs) contribute to nearly 80% of the human-specific cCREs in cortical cells. Through machine learning, we develop sequence-based predictors of cCREs in different species and demonstrate that the genomic regulatory syntax is highly preserved from rodents to primates. Lastly, we show that epigenetic conservation combined with sequence similarity helps uncover functional cis -regulatory elements and enhances our ability to interpret genetic variants contributing to neurological disease and traits.

AdaLiftOver: high-resolution identification of orthologous regulatory elements with Adaptive liftOver

MOTIVATION

Elucidating functionally similar orthologous regulatory regions for human and model organism genomes is critical for exploiting model organism research and advancing our understanding of results from genome-wide association studies (GWAS). Sequence conservation is the de facto approach for finding orthologous non-coding regions between human and model organism genomes. However, existing methods for mapping non-coding genomic regions across species are challenged by the multi-mapping, low precision, and low mapping rate issues.

RESULTS

We develop Adaptive liftOver (AdaLiftOver), a large-scale computational tool for identifying functionally similar orthologous non-coding regions across species. AdaLiftOver builds on the UCSC liftOver framework to extend the query regions and prioritizes the resulting candidate target regions based on the conservation of the epigenomic and the sequence grammar features. Evaluations of AdaLiftOver with multiple case studies, spanning both genomic intervals from epigenome datasets across a wide range of model organisms and GWAS SNPs, yield AdaLiftOver as a versatile method for deriving hard-to-obtain human epigenome datasets as well as reliably identifying orthologous loci for GWAS SNPs.

AVAILABILITY AND IMPLEMENTATION

The R package and the data for AdaLiftOver is available from https://github.com/keleslab/AdaLiftOver.

Evolution of homology: From archetype towards a holistic concept of cell type

The concept of homology lies in the heart of comparative biological science. The distinction between homology as structure and analogy as function has shaped the evolutionary paradigm for a century and formed the axis of comparative anatomy and embryology, which accept the identity of structure as a ground measure of relatedness. The advent of single-cell genomics overturned the classical view of cell homology by establishing a backbone regulatory identity of cell types, the basic biological units bridging the molecular and phenotypic dimensions, to reveal that the cell is the most flexible unit of living matter and that many approaches of classical biology need to be revised to understand evolution and diversity at the cellular level. The emerging theory of cell types explicitly decouples cell identity from phenotype, essentially allowing for the divergence of evolutionarily related morphotypes beyond recognition, as well as it decouples ontogenetic cell lineage from cell-type phylogeny, whereby explicating that cell types can share common descent regardless of their structure, function or developmental origin. The article succinctly summarizes current progress and opinion in this field and formulates a more generalistic view of biological cell types as avatars, transient or terminal cell states deployed in a continuum of states by the developmental programme of one and the same omnipotent cell, capable of changing or combining identities with distinct evolutionary histories or inventing ad hoc identities that never existed in evolution or development. It highlights how the new logic grounded in the regulatory nature of cell identity transforms the concepts of cell homology and phenotypic stability, suggesting that cellular evolution is inherently and massively network-like, with one-to-one homologies being rather uncommon and restricted to shallower levels of the animal tree of life.

Enhancer-FRL: Improved and Robust Identification of Enhancers and Their Activities Using Feature Representation Learning

Enhancers are crucial for precise regulation of gene expression, while enhancer identification and strength prediction are challenging because of their free distribution and tremendous number of similar fractions in the genome. Although several bioinformatics tools have been developed, shortfalls in these models remain, and their performances need further improvement. In the present study, a two-layer predictor called Enhancer-FRL was proposed for identifying enhancers (enhancers or nonenhancers) and their activities (strong and weak). More specifically, to build an efficient model, the feature representation learning scheme was applied to generate a 50D probabilistic vector based on 10 feature encodings and five machine learning algorithms. Subsequently, the multiview probabilistic features were integrated to construct the final prediction model. Compared with the single feature-based model, Enhancer-FRL showed significant performance improvement and model robustness. Performance assessment on the independent test dataset indicated that the proposed model outperformed state-of-the-art available toolkits. The webserver Enhancer-FRL is freely accessible at http://lab.malab.cn/∼wangchao/softwares/Enhancer-FRL/, The code and datasets can be downloaded at the webserver page or at the Github https://github.com/wangchao-malab/Enhancer-FRL/.

Deciphering the multi-scale, quantitative cis-regulatory code

Uncovering the cis-regulatory code that governs when and how much each gene is transcribed in a given genome and cellular state remains a central goal of biology. Here, we discuss major layers of regulation that influence how transcriptional outputs are encoded by DNA sequence and cellular context. We first discuss how transcription factors bind specific DNA sequences in a dosage-dependent and cooperative manner and then proceed to the cofactors that facilitate transcription factor function and mediate the activity of modular cis-regulatory elements such as enhancers, silencers, and promoters. We then consider the complex and poorly understood interplay of these diverse elements within regulatory landscapes and its relationships with chromatin states and nuclear organization. We propose that a mechanistically informed, quantitative model of transcriptional regulation that integrates these multiple regulatory layers will be the key to ultimately cracking the cis-regulatory code.

Applications of deep learning in understanding gene regulation

Gene regulation is a central topic in cell biology. Advances in omics technologies and the accumulation of omics data have provided better opportunities for gene regulation studies than ever before. For this reason deep learning, as a data-driven predictive modeling approach, has been successfully applied to this field during the past decade. In this article, we aim to give a brief yet comprehensive overview of representative deep-learning methods for gene regulation. Specifically, we discuss and compare the design principles and datasets used by each method, creating a reference for researchers who wish to replicate or improve existing methods. We also discuss the common problems of existing approaches and prospectively introduce the emerging deep-learning paradigms that will potentially alleviate them. We hope that this article will provide a rich and up-to-date resource and shed light on future research directions in this area.

maxATAC: Genome-scale transcription-factor binding prediction from ATAC-seq with deep neural networks

Transcription factors read the genome, fundamentally connecting DNA sequence to gene expression across diverse cell types. Determining how, where, and when TFs bind chromatin will advance our understanding of gene regulatory networks and cellular behavior. The 2017 ENCODE-DREAM in vivo Transcription-Factor Binding Site (TFBS) Prediction Challenge highlighted the value of chromatin accessibility data to TFBS prediction, establishing state-of-the-art methods for TFBS prediction from DNase-seq. However, the more recent Assay-for-Transposase-Accessible-Chromatin (ATAC)-seq has surpassed DNase-seq as the most widely-used chromatin accessibility profiling method. Furthermore, ATAC-seq is the only such technique available at single-cell resolution from standard commercial platforms. While ATAC-seq datasets grow exponentially, suboptimal motif scanning is unfortunately the most common method for TFBS prediction from ATAC-seq. To enable community access to state-of-the-art TFBS prediction from ATAC-seq, we (1) curated an extensive benchmark dataset (127 TFs) for ATAC-seq model training and (2) built "maxATAC", a suite of user-friendly, deep neural network models for genome-wide TFBS prediction from ATAC-seq in any cell type. With models available for 127 human TFs, maxATAC is the largest collection of high-performance TFBS prediction models for ATAC-seq. maxATAC performance extends to primary cells and single-cell ATAC-seq, enabling improved TFBS prediction in vivo. We demonstrate maxATAC's capabilities by identifying TFBS associated with allele-dependent chromatin accessibility at atopic dermatitis genetic risk loci.

Enhancer Function and Evolutionary Roles of Human Accelerated Regions

Human accelerated regions (HARs) are the fastest-evolving sequences in the human genome. When HARs were discovered in 2006, their function was mysterious due to scant annotation of the noncoding genome. Diverse technologies, from transgenic animals to machine learning, have consistently shown that HARs function as gene regulatory enhancers with significant enrichment in neurodevelopment. It is now possible to quantitatively measure the enhancer activity of thousands of HARs in parallel and model how each nucleotide contributes to gene expression. These strategies have revealed that many human HAR sequences function differently than their chimpanzee orthologs, though individual nucleotide changes in the same HAR may have opposite effects, consistent with compensatory substitutions. To fully evaluate the role of HARs in human evolution, it will be necessary to experimentally and computationally dissect them across more cell types and developmental stages.

Deep cis-regulatory homology of the butterfly wing pattern ground plan

Butterfly wing patterns derive from a deeply conserved developmental ground plan yet are diverse and evolve rapidly. It is poorly understood how gene regulatory architectures can accommodate both deep homology and adaptive change. To address this, we characterized the cis-regulatory evolution of the color pattern gene WntA in nymphalid butterflies. Comparative assay for transposase-accessible chromatin using sequencing (ATAC-seq) and in vivo deletions spanning 46 cis-regulatory elements across five species revealed deep homology of ground plan-determining sequences, except in monarch butterflies. Furthermore, noncoding deletions displayed both positive and negative regulatory effects that were often broad in nature. Our results provide little support for models predicting rapid enhancer turnover and suggest that deeply ancestral, multifunctional noncoding elements can underlie rapidly evolving trait systems.

Model Comparison of Heritability Enrichment Analysis in Livestock Population

Heritability enrichment analysis is an important means of exploring the genetic architecture of complex traits in human genetics. Heritability enrichment is typically defined as the proportion of an SNP subset explained heritability, divided by the proportion of SNPs. Heritability enrichment enables better study of underlying complex traits, such as functional variant/gene subsets, biological networks and metabolic pathways detected through integrating explosively increased omics data. This would be beneficial for genomic prediction of disease risk in humans and genetic values estimation of important economical traits in livestock and plant species. However, in livestock, factors affecting the heritability enrichment estimation of complex traits have not been examined. Previous studies on humans reported that the frequencies, effect sizes, and levels of linkage disequilibrium (LD) of underlying causal variants (CVs) would affect the heritability enrichment estimation. Therefore, the distribution of heritability across the genome should be fully considered to obtain the unbiased estimation of heritability enrichment. To explore the performance of different heritability enrichment models in livestock populations, we used the VanRaden, GCTA and α models, assuming different α values, and the LDAK model, considering LD weight. We simulated three types of phenotypes, with CVs from various minor allele frequency (MAF) ranges: genome-wide (0.005 ≤ MAF ≤ 0.5), common (0.05 ≤ MAF ≤ 0.5), and uncommon (0.01 ≤ MAF < 0.05). The performances of the models with two different subsets (one of which contained known CVs and the other consisting of randomly selected markers) were compared to verify the accuracy of heritability enrichment estimation of functional variant sets. Our results showed that models with known CV subsets provided more robust enrichment estimation. Models with different α values tended to provide stable and accurate estimates for common and genome-wide CVs (relative deviation 0.5−2.2%), while tending to underestimate the enrichment of uncommon CVs. As the α value increased, enrichments from 15.73% higher than true value (i.e., 3.00) to 48.93% lower than true value for uncommon CVs were observed. In addition, the long-range LD windows (e.g., 5000 kb) led to large bias of the enrichment estimations for both common and uncommon CVs. Overall, heritability enrichment estimations were sensitive for the α value assumption and LD weight consideration of different models. Accuracy would be greatly improved by using a suitable model. This study would be helpful in understanding the genetic architecture of complex traits and provides a reference for genetic analysis in the livestock population.

Ranking reprogramming factors for cell differentiation

Transcription factor over-expression is a proven method for reprogramming cells to a desired cell type for regenerative medicine and therapeutic discovery. However, a general method for the identification of reprogramming factors to create an arbitrary cell type is an open problem. Here we examine the success rate of methods and data for differentiation by testing the ability of nine computational methods (CellNet, GarNet, EBseq, AME, DREME, HOMER, KMAC, diffTF and DeepAccess) to discover and rank candidate factors for eight target cell types with known reprogramming solutions. We compare methods that use gene expression, biological networks and chromatin accessibility data, and comprehensively test parameter and preprocessing of input data to optimize performance. We find the best factor identification methods can identify an average of 50-60% of reprogramming factors within the top ten candidates, and methods that use chromatin accessibility perform the best. Among the chromatin accessibility methods, complex methods DeepAccess and diffTF have higher correlation with the ranked significance of transcription factor candidates within reprogramming protocols for differentiation. We provide evidence that AME and diffTF are optimal methods for transcription factor recovery that will allow for systematic prioritization of transcription factor candidates to aid in the design of new reprogramming protocols.

Genomics enters the deep learning era

The tremendous amount of biological sequence data available, combined with the recent methodological breakthrough in deep learning in domains such as computer vision or natural language processing, is leading today to the transformation of bioinformatics through the emergence of deep genomics, the application of deep learning to genomic sequences. We review here the new applications that the use of deep learning enables in the field, focusing on three aspects: the functional annotation of genomes, the sequence determinants of the genome functions and the possibility to write synthetic genomic sequences.

Machine learning sequence prioritization for cell type-specific enhancer design

Recent discoveries of extreme cellular diversity in the brain warrant rapid development of technologies to access specific cell populations within heterogeneous tissue. Available approaches for engineering-targeted technologies for new neuron subtypes are low yield, involving intensive transgenic strain or virus screening. Here, we present Specific Nuclear-Anchored Independent Labeling (SNAIL), an improved virus-based strategy for cell labeling and nuclear isolation from heterogeneous tissue. SNAIL works by leveraging machine learning and other computational approaches to identify DNA sequence features that confer cell type-specific gene activation and then make a probe that drives an affinity purification-compatible reporter gene. As a proof of concept, we designed and validated two novel SNAIL probes that target parvalbumin-expressing (PV+) neurons. Nuclear isolation using SNAIL in wild-type mice is sufficient to capture characteristic open chromatin features of PV+ neurons in the cortex, striatum, and external globus pallidus. The SNAIL framework also has high utility for multispecies cell probe engineering; expression from a mouse PV+ SNAIL enhancer sequence was enriched in PV+ neurons of the macaque cortex. Expansion of this technology has broad applications in cell type-specific observation, manipulation, and therapeutics across species and disease models.

DeepSTARR predicts enhancer activity from DNA sequence and enables the de novo design of synthetic enhancers

Enhancer sequences control gene expression and comprise binding sites (motifs) for different transcription factors (TFs). Despite extensive genetic and computational studies, the relationship between DNA sequence and regulatory activity is poorly understood, and de novo enhancer design has been challenging. Here, we built a deep-learning model, DeepSTARR, to quantitatively predict the activities of thousands of developmental and housekeeping enhancers directly from DNA sequence in Drosophila melanogaster S2 cells. The model learned relevant TF motifs and higher-order syntax rules, including functionally nonequivalent instances of the same TF motif that are determined by motif-flanking sequence and intermotif distances. We validated these rules experimentally and demonstrated that they can be generalized to humans by testing more than 40,000 wildtype and mutant Drosophila and human enhancers. Finally, we designed and functionally validated synthetic enhancers with desired activities de novo.

Inferring mammalian tissue-specific regulatory conservation by predicting tissue-specific differences in open chromatin

BACKGROUND

Evolutionary conservation is an invaluable tool for inferring functional significance in the genome, including regions that are crucial across many species and those that have undergone convergent evolution. Computational methods to test for sequence conservation are dominated by algorithms that examine the ability of one or more nucleotides to align across large evolutionary distances. While these nucleotide alignment-based approaches have proven powerful for protein-coding genes and some non-coding elements, they fail to capture conservation of many enhancers, distal regulatory elements that control spatial and temporal patterns of gene expression. The function of enhancers is governed by a complex, often tissue- and cell type-specific code that links combinations of transcription factor binding sites and other regulation-related sequence patterns to regulatory activity. Thus, function of orthologous enhancer regions can be conserved across large evolutionary distances, even when nucleotide turnover is high.

RESULTS

We present a new machine learning-based approach for evaluating enhancer conservation that leverages the combinatorial sequence code of enhancer activity rather than relying on the alignment of individual nucleotides. We first train a convolutional neural network model that can predict tissue-specific open chromatin, a proxy for enhancer activity, across mammals. Next, we apply that model to distinguish instances where the genome sequence would predict conserved function versus a loss of regulatory activity in that tissue. We present criteria for systematically evaluating model performance for this task and use them to demonstrate that our models accurately predict tissue-specific conservation and divergence in open chromatin between primate and rodent species, vastly out-performing leading nucleotide alignment-based approaches. We then apply our models to predict open chromatin at orthologs of brain and liver open chromatin regions across hundreds of mammals and find that brain enhancers associated with neuron activity have a stronger tendency than the general population to have predicted lineage-specific open chromatin.

CONCLUSION

The framework presented here provides a mechanism to annotate tissue-specific regulatory function across hundreds of genomes and to study enhancer evolution using predicted regulatory differences rather than nucleotide-level conservation measurements.

JASPAR 2022: the 9th release of the open-access database of transcription factor binding profiles

JASPAR (http://jaspar.genereg.net/) is an open-access database containing manually curated, non-redundant transcription factor (TF) binding profiles for TFs across six taxonomic groups. In this 9th release, we expanded the CORE collection with 341 new profiles (148 for plants, 101 for vertebrates, 85 for urochordates, and 7 for insects), which corresponds to a 19% expansion over the previous release. We added 298 new profiles to the Unvalidated collection when no orthogonal evidence was found in the literature. All the profiles were clustered to provide familial binding profiles for each taxonomic group. Moreover, we revised the structural classification of DNA binding domains to consider plant-specific TFs. This release introduces word clouds to represent the scientific knowledge associated with each TF. We updated the genome tracks of TFBSs predicted with JASPAR profiles in eight organisms; the human and mouse TFBS predictions can be visualized as native tracks in the UCSC Genome Browser. Finally, we provide a new tool to perform JASPAR TFBS enrichment analysis in user-provided genomic regions. All the data is accessible through the JASPAR website, its associated RESTful API, the R/Bioconductor data package, and a new Python package, pyJASPAR, that facilitates serverless access to the data.

Decoding gene regulation in the fly brain

The Drosophila brain is a frequently used model in neuroscience. Single-cell transcriptome analysis^1-6, three-dimensional morphological classification⁷ and electron microscopy mapping of the connectome^8,9 have revealed an immense diversity of neuronal and glial cell types that underlie an array of functional and behavioural traits in the fly. The identities of these cell types are controlled by gene regulatory networks (GRNs), involving combinations of transcription factors that bind to genomic enhancers to regulate their target genes. Here, to characterize GRNs at the cell-type level in the fly brain, we profiled the chromatin accessibility of 240,919 single cells spanning 9 developmental timepoints and integrated these data with single-cell transcriptomes. We identify more than 95,000 regulatory regions that are used in different neuronal cell types, of which 70,000 are linked to developmental trajectories involving neurogenesis, reprogramming and maturation. For 40 cell types, uniquely accessible regions were associated with their expressed transcription factors and downstream target genes through a combination of motif discovery, network inference and deep learning, creating enhancer GRNs. The enhancer architectures revealed by DeepFlyBrain lead to a better understanding of neuronal regulatory diversity and can be used to design genetic driver lines for cell types at specific timepoints, facilitating their characterization and manipulation.

Analysis of long and short enhancers in melanoma cell states

Understanding how enhancers drive cell-type specificity and efficiently identifying them is essential for the development of innovative therapeutic strategies. In melanoma, the melanocytic (MEL) and the mesenchymal-like (MES) states present themselves with different responses to therapy, making the identification of specific enhancers highly relevant. Using massively parallel reporter assays (MPRAs) in a panel of patient-derived melanoma lines (MM lines), we set to identify and decipher melanoma enhancers by first focusing on regions with state-specific H3K27 acetylation close to differentially expressed genes. An in-depth evaluation of those regions was then pursued by investigating the activity of overlapping ATAC-seq peaks along with a full tiling of the acetylated regions with 190 bp sequences. Activity was observed in more than 60% of the selected regions, and we were able to precisely locate the active enhancers within ATAC-seq peaks. Comparison of sequence content with activity, using the deep learning model DeepMEL2, revealed that AP-1 alone is responsible for the MES enhancer activity. In contrast, SOX10 and MITF both influence MEL enhancer function with SOX10 being required to achieve high levels of activity. Overall, our MPRAs shed light on the relationship between long and short sequences in terms of their sequence content, enhancer activity, and specificity across melanoma cell states.

Addiction-Associated Genetic Variants Implicate Brain Cell Type- and Region-Specific Cis-Regulatory Elements in Addiction Neurobiology

Recent large genome-wide association studies have identified multiple confident risk loci linked to addiction-associated behavioral traits. Most genetic variants linked to addiction-associated traits lie in noncoding regions of the genome, likely disrupting cis-regulatory element (CRE) function. CREs tend to be highly cell type-specific and may contribute to the functional development of the neural circuits underlying addiction. Yet, a systematic approach for predicting the impact of risk variants on the CREs of specific cell populations is lacking. To dissect the cell types and brain regions underlying addiction-associated traits, we applied stratified linkage disequilibrium score regression to compare genome-wide association studies to genomic regions collected from human and mouse assays for open chromatin, which is associated with CRE activity. We found enrichment of addiction-associated variants in putative CREs marked by open chromatin in neuronal (NeuN+) nuclei collected from multiple prefrontal cortical areas and striatal regions known to play major roles in reward and addiction. To further dissect the cell type-specific basis of addiction-associated traits, we also identified enrichments in human orthologs of open chromatin regions of female and male mouse neuronal subtypes: cortical excitatory, D1, D2, and PV. Last, we developed machine learning models to predict mouse cell type-specific open chromatin, enabling us to further categorize human NeuN+ open chromatin regions into cortical excitatory or striatal D1 and D2 neurons and predict the functional impact of addiction-associated genetic variants. Our results suggest that different neuronal subtypes within the reward system play distinct roles in the variety of traits that contribute to addiction.SIGNIFICANCE STATEMENT We combine statistical genetic and machine learning techniques to find that the predisposition to for nicotine, alcohol, and cannabis use behaviors can be partially explained by genetic variants in conserved regulatory elements within specific brain regions and neuronal subtypes of the reward system. Our computational framework can flexibly integrate open chromatin data across species to screen for putative causal variants in a cell type- and tissue-specific manner for numerous complex traits.

INFIMA leverages multi-omics model organism data to identify effector genes of human GWAS variants

Genome-wide association studies reveal many non-coding variants associated with complex traits. However, model organism studies largely remain as an untapped resource for unveiling the effector genes of non-coding variants. We develop INFIMA, Integrative Fine-Mapping, to pinpoint causal SNPs for diversity outbred (DO) mice eQTL by integrating founder mice multi-omics data including ATAC-seq, RNA-seq, footprinting, and in silico mutation analysis. We demonstrate INFIMA's superior performance compared to alternatives with human and mouse chromatin conformation capture datasets. We apply INFIMA to identify novel effector genes for GWAS variants associated with diabetes. The results of the application are available at http://www.statlab.wisc.edu/shiny/INFIMA/ .

CDK7 and MITF repress a transcription program involved in survival and drug tolerance in melanoma

Melanoma cell phenotype switching between differentiated melanocytic and undifferentiated mesenchymal-like states drives metastasis and drug resistance. CDK7 is the serine/threonine kinase of the basal transcription factor TFIIH. We show that dedifferentiation of melanocytic-type melanoma cells into mesenchymal-like cells and acquisition of tolerance to targeted therapies is achieved through chronic inhibition of CDK7. In addition to emergence of a mesenchymal-type signature, we identify a GATA6-dependent gene expression program comprising genes such as AMIGO2 or ABCG2 involved in melanoma survival or targeted drug tolerance, respectively. Mechanistically, we show that CDK7 drives expression of the melanocyte lineage transcription factor MITF that in turn binds to an intronic region of GATA6 to repress its expression in melanocytic-type cells. We show that GATA6 expression is activated in MITF-low melanoma cells of patient-derived xenografts. Taken together, our data show how the poorly characterized repressive function of MITF in melanoma participates in a molecular cascade regulating activation of a transcriptional program involved in survival and drug resistance in melanoma.

AP-2α-Mediated Activation of E2F and EZH2 Drives Melanoma Metastasis

In melanoma metastasis, the role of the AP-2α transcription factor, which is encoded by TFAP2A, is controversial as some findings have suggested tumor suppressor activity while other studies have shown high TFAP2A expression in node-positive melanoma associated with poor prognosis. Here we demonstrate that AP-2α facilitates melanoma metastasis through transcriptional activation of genes within the E2F pathway including EZH2. A BioID screen found that AP-2α interacts with members of the nucleosome remodeling and deacetylase (NuRD) complex. Loss of AP-2α removed activating chromatin marks in the promoters of EZH2 and other E2F target genes through activation of the NuRD repression complex. In melanoma cells, treatment with tazemetostat, an FDA-approved and highly specific EZH2 inhibitor, substantially reduced anchorage-independent colony formation and demonstrated heritable antimetastatic effects, which were dependent on AP-2α. Single-cell RNA sequencing analysis of a metastatic melanoma mouse model revealed hyperexpansion of Tfap2a ^High/E2F-activated cell populations in transformed melanoma relative to progenitor melanocyte stem cells. These findings demonstrate that melanoma metastasis is driven by the AP-2α/EZH2 pathway and suggest that AP-2α expression can be used as a biomarker to predict responsiveness to EZH2 inhibitors for the treatment of advanced melanomas. SIGNIFICANCE: AP-2α drives melanoma metastasis by upregulating E2F pathway genes including EZH2 through inhibition of the NuRD repression complex, serving as a biomarker to predict responsiveness to EZH2 inhibitors.

Annotating the Insect Regulatory Genome

An ever-growing number of insect genomes is being sequenced across the evolutionary spectrum. Comprehensive annotation of not only genes but also regulatory regions is critical for reaping the full benefits of this sequencing. Driven by developments in sequencing technologies and in both empirical and computational discovery strategies, the past few decades have witnessed dramatic progress in our ability to identify cis-regulatory modules (CRMs), sequences such as enhancers that play a major role in regulating transcription. Nevertheless, providing a timely and comprehensive regulatory annotation of newly sequenced insect genomes is an ongoing challenge. We review here the methods being used to identify CRMs in both model and non-model insect species, and focus on two tools that we have developed, REDfly and SCRMshaw. These resources can be paired together in a powerful combination to facilitate insect regulatory annotation over a broad range of species, with an accuracy equal to or better than that of other state-of-the-art methods.

Evolutionary Cell Type Mapping with Single-Cell Genomics

A fundamental characteristic of animal multicellularity is the spatial coexistence of functionally specialized cell types that are all encoded by a single genome sequence. Cell type transcriptional programs are deployed and maintained by regulatory mechanisms that control the asymmetric, differential access to genomic information in each cell. This genome regulation ultimately results in specific cellular phenotypes. However, the emergence, diversity, and evolutionary dynamics of animal cell types remain almost completely unexplored beyond a few species. Single-cell genomics is emerging as a powerful tool to build comprehensive catalogs of cell types and their associated gene regulatory programs in non-traditional model species. We review the current state of sampling efforts across the animal tree of life and challenges ahead for the comparative study of cell type programs. We also discuss how the phylogenetic integration of cell atlases can lead to the development of models of cell type evolution and a phylogenetic taxonomy of cells.

Comparative Transcriptomic Analysis of the Hematopoietic System between Human and Mouse by Single Cell RNA Sequencing

(1) Background: mouse models are fundamental to the study of hematopoiesis, but comparisons between mouse and human in single cells have been limited in depth. (2) Methods: we constructed a single-cell resolution transcriptomic atlas of hematopoietic stem and progenitor cells (HSPCs) of human and mouse, from a total of 32,805 single cells. We used Monocle to examine the trajectories of hematopoietic differentiation, and SCENIC to analyze gene networks underlying hematopoiesis. (3) Results: After alignment with Seurat 2, the cells of mouse and human could be separated by same cell type categories. Cells were grouped into 17 subpopulations; cluster-specific genes were species-conserved and shared functional themes. The clustering dendrogram indicated that cell types were highly conserved between human and mouse. A visualization of the Monocle results provided an intuitive representation of HSPC differentiation to three dominant branches (Erythroid/megakaryocytic, Myeloid, and Lymphoid), derived directly from the hematopoietic stem cell and the long-term hematopoietic stem cells in both human and mouse. Gene regulation was similarly conserved, reflected by comparable transcriptional factors and regulatory sequence motifs in subpopulations of cells. (4) Conclusions: our analysis has confirmed evolutionary conservation in the hematopoietic systems of mouse and human, extending to cell types, gene expression and regulatory elements.

Interpretation of allele-specific chromatin accessibility using cell state-aware deep learning

Genomic sequence variation within enhancers and promoters can have a significant impact on the cellular state and phenotype. However, sifting through the millions of candidate variants in a personal genome or a cancer genome, to identify those that impact cis-regulatory function, remains a major challenge. Interpretation of noncoding genome variation benefits from explainable artificial intelligence to predict and interpret the impact of a mutation on gene regulation. Here we generate phased whole genomes with matched chromatin accessibility, histone modifications, and gene expression for 10 melanoma cell lines. We find that training a specialized deep learning model, called DeepMEL2, on melanoma chromatin accessibility data can capture the various regulatory programs of the melanocytic and mesenchymal-like melanoma cell states. This model outperforms motif-based variant scoring, as well as more generic deep learning models. We detect hundreds to thousands of allele-specific chromatin accessibility variants (ASCAVs) in each melanoma genome, of which 15%-20% can be explained by gains or losses of transcription factor binding sites. A considerable fraction of ASCAVs are caused by changes in AP-1 binding, as confirmed by matched ChIP-seq data to identify allele-specific binding of JUN and FOSL1. Finally, by augmenting the DeepMEL2 model with ChIP-seq data for GABPA, the TERT promoter mutation, as well as additional ETS motif gains, can be identified with high confidence. In conclusion, we present a new integrative genomics approach and a deep learning model to identify and interpret functional enhancer mutations with allelic imbalance of chromatin accessibility and gene expression.

On the importance of evolutionary constraint for regulatory sequence identification

Regulation of gene expression relies on the activity of specialized genomic elements, enhancers or silencers, distributed over sometimes large distance from their target gene promoters. A significant part of vertebrate genomes consists in such regulatory elements, but their identification and that of their target genes remains challenging, due to the lack of clear signature at the nucleotide level. For many years the main hallmark used for identifying functional elements has been their sequence conservation between genomes of distant species, indicative of purifying selection. More recently, genome-wide biochemical assays have opened new avenues for detecting regulatory regions, shifting attention away from evolutionary constraints. Here, we review the respective contributions of comparative genomics and biochemical assays for the definition of regulatory elements and their targets and advocate that both sequence conservation and preserved synteny, taken as signature of functional constraint, remain essential tools in this task.

Enhancer grammar in development, evolution, and disease: dependencies and interplay

Each language has standard books describing that language's grammatical rules. Biologists have searched for similar, albeit more complex, principles relating enhancer sequence to gene expression. Here, we review the literature on enhancer grammar. We introduce dependency grammar, a model where enhancers encode information based on dependencies between enhancer features shaped by mechanistic, evolutionary, and biological constraints. Classifying enhancers based on the types of dependencies may identify unifying principles relating enhancer sequence to gene expression. Such rules would allow us to read the instructions for development within genomes and pinpoint causal enhancer variants underlying disease and evolutionary changes.

Chromatin accessibility profiling methods

Chromatin accessibility, or the physical access to chromatinized DNA, is a widely studied characteristic of the eukaryotic genome. As active regulatory DNA elements are generally 'accessible', the genome-wide profiling of chromatin accessibility can be used to identify candidate regulatory genomic regions in a tissue or cell type. Multiple biochemical methods have been developed to profile chromatin accessibility, both in bulk and at the single-cell level. Depending on the method, enzymatic cleavage, transposition or DNA methyltransferases are used, followed by high-throughput sequencing, providing a view of genome-wide chromatin accessibility. In this Primer, we discuss these biochemical methods, as well as bioinformatics tools for analysing and interpreting the generated data, and insights into the key regulators underlying developmental, evolutionary and disease processes. We outline standards for data quality, reproducibility and deposition used by the genomics community. Although chromatin accessibility profiling is invaluable to study gene regulation, alone it provides only a partial view of this complex process. Orthogonal assays facilitate the interpretation of accessible regions with respect to enhancer-promoter proximity, functional transcription factor binding and regulatory function. We envision that technological improvements including single-molecule, multi-omics and spatial methods will bring further insight into the secrets of genome regulation.