Base-resolution models of transcription-factor binding reveal soft motif syntax

Transfer learning for cross-context prediction of protein expression from 5'UTR sequence

Model-guided DNA sequence design can accelerate the reprogramming of living cells. It allows us to engineer more complex biological systems by removing the need to physically assemble and test each potential design. While mechanistic models of gene expression have seen some success in supporting this goal, data-centric, deep learning-based approaches often provide more accurate predictions. This accuracy, however, comes at a cost - a lack of generalization across genetic and experimental contexts that has limited their wider use outside the context in which they were trained. Here, we address this issue by demonstrating how a simple transfer learning procedure can effectively tune a pre-trained deep learning model to predict protein translation rate from 5' untranslated region (5'UTR) sequence for diverse contexts in Escherichia coli using a small number of new measurements. This allows for important model features learnt from expensive massively parallel reporter assays to be easily transferred to new settings. By releasing our trained deep learning model and complementary calibration procedure, this study acts as a starting point for continually refined model-based sequence design that builds on previous knowledge and future experimental efforts.

Position-dependent function of human sequence-specific transcription factors

Patterns of transcriptional activity are encoded in our genome through regulatory elements such as promoters or enhancers that, paradoxically, contain similar assortments of sequence-specific transcription factor (TF) binding sites1-3. Knowledge of how these sequence motifs encode multiple, often overlapping, gene expression programs is central to understanding gene regulation and how mutations in non-coding DNA manifest in disease4,5. Here, by studying gene regulation from the perspective of individual transcription start sites (TSSs), using natural genetic variation, perturbation of endogenous TF protein levels and massively parallel analysis of natural and synthetic regulatory elements, we show that the effect of TF binding on transcription initiation is position dependent. Analysing TF-binding-site occurrences relative to the TSS, we identified several motifs with highly preferential positioning. We show that these patterns are a combination of a TF's distinct functional profiles-many TFs, including canonical activators such as NRF1, NFY and Sp1, activate or repress transcription initiation depending on their precise position relative to the TSS. As such, TFs and their spacing collectively guide the site and frequency of transcription initiation. More broadly, these findings reveal how similar assortments of TF binding sites can generate distinct gene regulatory outcomes depending on their spatial configuration and how DNA sequence polymorphisms may contribute to transcription variation and disease and underscore a critical role for TSS data in decoding the regulatory information of our genome.

TFscope: systematic analysis of the sequence features involved in the binding preferences of transcription factors

Characterizing the binding preferences of transcription factors (TFs) in different cell types and conditions is key to understand how they orchestrate gene expression. Here, we develop TFscope, a machine learning approach that identifies sequence features explaining the binding differences observed between two ChIP-seq experiments targeting either the same TF in two conditions or two TFs with similar motifs (paralogous TFs). TFscope systematically investigates differences in the core motif, nucleotide environment and co-factor motifs, and provides the contribution of each key feature in the two experiments. TFscope was applied to > 305 ChIP-seq pairs, and several examples are discussed.

Systematic dissection of sequence features affecting binding specificity of a pioneer factor reveals binding synergy between FOXA1 and AP-1

Despite the unique ability of pioneer factors (PFs) to target nucleosomal sites in closed chromatin, they only bind a small fraction of their genomic motifs. The underlying mechanism of this selectivity is not well understood. Here, we design a high-throughput assay called chromatin immunoprecipitation with integrated synthetic oligonucleotides (ChIP-ISO) to systematically dissect sequence features affecting the binding specificity of a classic PF, FOXA1, in human A549 cells. Combining ChIP-ISO with in vitro and neural network analyses, we find that (1) FOXA1 binding is strongly affected by co-binding transcription factors (TFs) AP-1 and CEBPB; (2) FOXA1 and AP-1 show binding cooperativity in vitro; (3) FOXA1's binding is determined more by local sequences than chromatin context, including eu-/heterochromatin; and (4) AP-1 is partially responsible for differential binding of FOXA1 in different cell types. Our study presents a framework for elucidating genetic rules underlying PF binding specificity and reveals a mechanism for context-specific regulation of its binding.

Advances in computational and experimental approaches for deciphering transcriptional regulatory networks: Understanding the roles of cis-regulatory elements is essential, and recent research utilizing MPRAs, STARR-seq, CRISPR-Cas9, and machine learning has yielded valuable insights

Understanding the influence of cis-regulatory elements on gene regulation poses numerous challenges given complexities stemming from variations in transcription factor (TF) binding, chromatin accessibility, structural constraints, and cell-type differences. This review discusses the role of gene regulatory networks in enhancing understanding of transcriptional regulation and covers construction methods ranging from expression-based approaches to supervised machine learning. Additionally, key experimental methods, including MPRAs and CRISPR-Cas9-based screening, which have significantly contributed to understanding TF binding preferences and cis-regulatory element functions, are explored. Lastly, the potential of machine learning and artificial intelligence to unravel cis-regulatory logic is analyzed. These computational advances have far-reaching implications for precision medicine, therapeutic target discovery, and the study of genetic variations in health and disease.

Gauge fixing for sequence-function relationships

Quantitative models of sequence-function relationships are ubiquitous in computational biology, e.g., for modeling the DNA binding of transcription factors or the fitness landscapes of proteins. Interpreting these models, however, is complicated by the fact that the values of model parameters can often be changed without affecting model predictions. Before the values of model parameters can be meaningfully interpreted, one must remove these degrees of freedom (called "gauge freedoms" in physics) by imposing additional constraints (a process called "fixing the gauge"). However, strategies for fixing the gauge of sequence-function relationships have received little attention. Here we derive an analytically tractable family of gauges for a large class of sequence-function relationships. These gauges are derived in the context of models with all-order interactions, but an important subset of these gauges can be applied to diverse types of models, including additive models, pairwise-interaction models, and models with higher-order interactions. Many commonly used gauges are special cases of gauges within this family. We demonstrate the utility of this family of gauges by showing how different choices of gauge can be used both to explore complex activity landscapes and to reveal simplified models that are approximately correct within localized regions of sequence space. The results provide practical gauge-fixing strategies and demonstrate the utility of gauge-fixing for model exploration and interpretation.

Multiplexed single-cell characterization of alternative polyadenylation regulators

Most mammalian genes have multiple polyA sites, representing a substantial source of transcript diversity regulated by the cleavage and polyadenylation (CPA) machinery. To better understand how these proteins govern polyA site choice, we introduce CPA-Perturb-seq, a multiplexed perturbation screen dataset of 42 CPA regulators with a 3' scRNA-seq readout that enables transcriptome-wide inference of polyA site usage. We develop a framework to detect perturbation-dependent changes in polyadenylation and characterize modules of co-regulated polyA sites. We find groups of intronic polyA sites regulated by distinct components of the nuclear RNA life cycle, including elongation, splicing, termination, and surveillance. We train and validate a deep neural network (APARENT-Perturb) for tandem polyA site usage, delineating a cis-regulatory code that predicts perturbation response and reveals interactions between regulatory complexes. Our work highlights the potential for multiplexed single-cell perturbation screens to further our understanding of post-transcriptional regulation.

Iterative deep learning-design of human enhancers exploits condensed sequence grammar to achieve cell type-specificity

An important and largely unsolved problem in synthetic biology is how to target gene expression to specific cell types. Here, we apply iterative deep learning to design synthetic enhancers with strong differential activity between two human cell lines. We initially train models on published datasets of enhancer activity and chromatin accessibility and use them to guide the design of synthetic enhancers that maximize predicted specificity. We experimentally validate these sequences, use the measurements to re-optimize the predictor, and design a second generation of enhancers with improved specificity. Our design methods embed relevant transcription factor binding site (TFBS) motifs with higher frequencies than comparable endogenous enhancers while using a more selective motif vocabulary, and we show that enhancer activity is correlated with transcription factor expression at the single cell level. Finally, we characterize causal features of top enhancers via perturbation experiments and show enhancers as short as 50bp can maintain specificity.

An updated compendium and reevaluation of the evidence for nuclear transcription factor occupancy over the mitochondrial genome

In most eukaryotes, mitochondrial organelles contain their own genome, usually circular, which is the remnant of the genome of the ancestral bacterial endosymbiont that gave rise to modern mitochondria. Mitochondrial genomes are dramatically reduced in their gene content due to the process of endosymbiotic gene transfer to the nucleus; as a result most mitochondrial proteins are encoded in the nucleus and imported into mitochondria. This includes the components of the dedicated mitochondrial transcription and replication systems and regulatory factors, which are entirely distinct from the information processing systems in the nucleus. However, since the 1990s several nuclear transcription factors have been reported to act in mitochondria, and previously we identified 8 human and 3 mouse transcription factors (TFs) with strong localized enrichment over the mitochondrial genome using ChIP-seq (Chromatin Immunoprecipitation) datasets from the second phase of the ENCODE (Encyclopedia of DNA Elements) Project Consortium. Here, we analyze the greatly expanded in the intervening decade ENCODE compendium of TF ChIP-seq datasets (a total of 6,153 ChIP experiments for 942 proteins, of which 763 are sequence-specific TFs) combined with interpretative deep learning models of TF occupancy to create a comprehensive compendium of nuclear TFs that show evidence of association with the mitochondrial genome. We find some evidence for chrM occupancy for 50 nuclear TFs and two other proteins, with bZIP TFs emerging as most likely to be playing a role in mitochondria. However, we also observe that in cases where the same TF has been assayed with multiple antibodies and ChIP protocols, evidence for its chrM occupancy is not always reproducible. In the light of these findings, we discuss the evidential criteria for establishing chrM occupancy and reevaluate the overall compendium of putative mitochondrial-acting nuclear TFs.

Automated classification of cellular expression in multiplexed imaging data with Nimbus

Multiplexed imaging offers a powerful approach to characterize the spatial topography of tissues in both health and disease. To analyze such data, the specific combination of markers that are present in each cell must be enumerated to enable accurate phenotyping, a process that often relies on unsupervised clustering. We constructed the Pan-Multiplex (Pan-M) dataset containing 197 million distinct annotations of marker expression across 15 different cell types. We used Pan-M to create Nimbus, a deep learning model to predict marker positivity from multiplexed image data. Nimbus is a pre-trained model that uses the underlying images to classify marker expression across distinct cell types, from different tissues, acquired using different microscope platforms, without requiring any retraining. We demonstrate that Nimbus predictions capture the underlying staining patterns of the full diversity of markers present in Pan-M. We then show how Nimbus predictions can be integrated with downstream clustering algorithms to robustly identify cell subtypes in image data. We have open-sourced Nimbus and Pan-M to enable community use at https://github.com/angelolab/Nimbus-Inference.

SuPreMo: a computational tool for streamlining in silico perturbation using sequence-based predictive models

SUMMARY

The increasing development of sequence-based machine learning models has raised the demand for manipulating sequences for this application. However, existing approaches to edit and evaluate genome sequences using models have limitations, such as incompatibility with structural variants, challenges in identifying responsible sequence perturbations, and the need for vcf file inputs and phased data. To address these bottlenecks, we present Sequence Mutator for Predictive Models (SuPreMo), a scalable and comprehensive tool for performing and supporting in silico mutagenesis experiments. We then demonstrate how pairs of reference and perturbed sequences can be used with machine learning models to prioritize pathogenic variants or discover new functional sequences.

AVAILABILITY AND IMPLEMENTATION

SuPreMo was written in Python, and can be run using only one line of code to generate both sequences and 3D genome disruption scores. The codebase, instructions for installation and use, and tutorials are on the GitHub page: https://github.com/ketringjoni/SuPreMo.

Systematic assessment of ISWI subunits shows that NURF creates local accessibility for CTCF

Catalytic activity of the imitation switch (ISWI) family of remodelers is critical for nucleosomal organization and DNA binding of certain transcription factors, including the insulator protein CTCF. Here we define the contribution of individual subcomplexes by deriving a panel of isogenic mouse stem cell lines, each lacking one of six ISWI accessory subunits. Individual deletions of subunits of either CERF, RSF, ACF, WICH or NoRC subcomplexes only moderately affect the chromatin landscape, while removal of the NURF-specific subunit BPTF leads to a strong reduction in chromatin accessibility and SNF2H ATPase localization around CTCF sites. This affects adjacent nucleosome occupancy and CTCF binding. At a group of sites with reduced chromatin accessibility, CTCF binding persists but cohesin occupancy is reduced, resulting in decreased insulation. These results suggest that CTCF binding can be separated from its function as an insulator in nuclear organization and identify a specific role for NURF in mediating SNF2H localization and chromatin opening at bound CTCF sites.

Decoding polygenic diseases: advances in noncoding variant prioritization and validation

Genome-wide association studies (GWASs) provide a key foundation for elucidating the genetic underpinnings of common polygenic diseases. However, these studies have limitations in their ability to assign causality to particular genetic variants, especially those residing in the noncoding genome. Over the past decade, technological and methodological advances in both analytical and empirical prioritization of noncoding variants have enabled the identification of causative variants by leveraging orthogonal functional evidence at increasing scale. In this review, we present an overview of these approaches and describe how this workflow provides the groundwork necessary to move beyond associations toward genetically informed studies on the molecular and cellular mechanisms of polygenic disease.

Dissecting the cis-regulatory syntax of transcription initiation with deep learning

Despite extensive characterization of mammalian Pol II transcription, the DNA sequence determinants of transcription initiation at a third of human promoters and most enhancers remain poorly understood. Hence, we trained and interpreted a neural network called ProCapNet that accurately models base-resolution initiation profiles from PRO-cap experiments using local DNA sequence. ProCapNet learns sequence motifs with distinct effects on initiation rates and TSS positioning and uncovers context-specific cryptic initiator elements intertwined within other TF motifs. ProCapNet annotates predictive motifs in nearly all actively transcribed regulatory elements across multiple cell-lines, revealing a shared cis-regulatory logic across promoters and enhancers mediated by a highly epistatic sequence syntax of cooperative and competitive motif interactions. ProCapNet models of RAMPAGE profiles measuring steady-state RNA abundance at TSSs distill initiation signals on par with models trained directly on PRO-cap profiles. ProCapNet learns a largely cell-type-agnostic cis-regulatory code of initiation complementing sequence drivers of cell-type-specific chromatin state critical for accurate prediction of cell-type-specific transcription initiation.

Transfer learning reveals sequence determinants of the quantitative response to transcription factor dosage

Deep learning approaches have made significant advances in predicting cell type-specific chromatin patterns from the identity and arrangement of transcription factor (TF) binding motifs. However, most models have been applied in unperturbed contexts, precluding a predictive understanding of how chromatin state responds to TF perturbation. Here, we used transfer learning to train and interpret deep learning models that use DNA sequence to predict, with accuracy approaching experimental reproducibility, how the concentration of two dosage-sensitive TFs (TWIST1, SOX9) affects regulatory element (RE) chromatin accessibility in facial progenitor cells. High-affinity motifs that allow for heterotypic TF co-binding and are concentrated at the center of REs buffer against quantitative changes in TF dosage and strongly predict unperturbed accessibility. In contrast, motifs with low-affinity or homotypic binding distributed throughout REs lead to sensitive responses with minimal contributions to unperturbed accessibility. Both buffering and sensitizing features show signatures of purifying selection. We validated these predictive sequence features using reporter assays and showed that a biophysical model of TF-nucleosome competition can explain the sensitizing effect of low-affinity motifs. Our approach of combining transfer learning and quantitative measurements of the chromatin response to TF dosage therefore represents a powerful method to reveal additional layers of the cis-regulatory code.

Using a comprehensive atlas and predictive models to reveal the complexity and evolution of brain-active regulatory elements

Most genetic variants associated with psychiatric disorders are located in noncoding regions of the genome. To investigate their functional implications, we integrate epigenetic data from the PsychENCODE Consortium and other published sources to construct a comprehensive atlas of candidate brain cis-regulatory elements. Using deep learning, we model these elements' sequence syntax and predict how binding sites for lineage-specific transcription factors contribute to cell type-specific gene regulation in various types of glia and neurons. The elements' evolutionary history suggests that new regulatory information in the brain emerges primarily via smaller sequence mutations within conserved mammalian elements rather than entirely new human- or primate-specific sequences. However, primate-specific candidate elements, particularly those active during fetal brain development and in excitatory neurons and astrocytes, are implicated in the heritability of brain-related human traits. Additionally, we introduce PsychSCREEN, a web-based platform offering interactive visualization of PsychENCODE-generated genetic and epigenetic data from diverse brain cell types in individuals with psychiatric disorders and healthy controls.

Predictive Biophysical Neural Network Modeling of a Compendium of in vivo Transcription Factor DNA Binding Profiles for Escherichia coli

The DNA binding of most Escherichia coli Transcription Factors (TFs) has not been comprehensively mapped, and few have models that can quantitatively predict binding affinity. We report the global mapping of in vivo DNA binding for 139 E. coli TFs using ChIP-Seq. We used these data to train BoltzNet, a novel neural network that predicts TF binding energy from DNA sequence. BoltzNet mirrors a quantitative biophysical model and provides directly interpretable predictions genome-wide at nucleotide resolution. We used BoltzNet to quantitatively design novel binding sites, which we validated with biophysical experiments on purified protein. We have generated models for 125 TFs that provide insight into global features of TF binding, including clustering of sites, the role of accessory bases, the relevance of weak sites, and the background affinity of the genome. Our paper provides new paradigms for studying TF-DNA binding and for the development of biophysically motivated neural networks.

Dual-role transcription factors stabilize intermediate expression levels

Precise control of gene expression levels is essential for normal cell functions, yet how they are defined and tightly maintained, particularly at intermediate levels, remains elusive. Here, using a series of newly developed sequencing, imaging, and functional assays, we uncover a class of transcription factors with dual roles as activators and repressors, referred to as condensate-forming level-regulating dual-action transcription factors (TFs). They reduce high expression but increase low expression to achieve stable intermediate levels. Dual-action TFs directly exert activating and repressing functions via condensate-forming domains that compartmentalize core transcriptional unit selectively. Clinically relevant mutations in these domains, which are linked to a range of developmental disorders, impair condensate selectivity and dual-action TF activity. These results collectively address a fundamental question in expression regulation and demonstrate the potential of level-regulating dual-action TFs as powerful effectors for engineering controlled expression levels.

A statistical approach for identifying single nucleotide variants that affect transcription factor binding

Non-coding variants located within regulatory elements may alter gene expression by modifying transcription factor (TF) binding sites, thereby leading to functional consequences. Different TF models are being used to assess the effect of DNA sequence variants, such as single nucleotide variants (SNVs). Often existing methods are slow and do not assess statistical significance of results. We investigated the distribution of absolute maximal differential TF binding scores for general computational models that affect TF binding. We find that a modified Laplace distribution can adequately approximate the empirical distributions. A benchmark on in vitro and in vivo datasets showed that our approach improves upon an existing method in terms of performance and speed. Applications on eQTLs and on a genome-wide association study illustrate the usefulness of our statistics by highlighting cell type-specific regulators and target genes. An implementation of our approach is freely available on GitHub and as bioconda package.

Genomic context sensitizes regulatory elements to genetic disruption

Genomic context critically modulates regulatory function but is difficult to manipulate systematically. The murine insulin-like growth factor 2 (Igf2)/H19 locus is a paradigmatic model of enhancer selectivity, whereby CTCF occupancy at an imprinting control region directs downstream enhancers to activate either H19 or Igf2. We used synthetic regulatory genomics to repeatedly replace the native locus with 157-kb payloads, and we systematically dissected its architecture. Enhancer deletion and ectopic delivery revealed previously uncharacterized long-range regulatory dependencies at the native locus. Exchanging the H19 enhancer cluster with the Sox2 locus control region (LCR) showed that the H19 enhancers relied on their native surroundings while the Sox2 LCR functioned autonomously. Analysis of regulatory DNA actuation across cell types revealed that these enhancer clusters typify broader classes of context sensitivity genome wide. These results show that unexpected dependencies influence even well-studied loci, and our approach permits large-scale manipulation of complete loci to investigate the relationship between regulatory architecture and function.

Functional dissection of complex and molecular trait variants at single nucleotide resolution

Identifying the causal variants and mechanisms that drive complex traits and diseases remains a core problem in human genetics. The majority of these variants have individually weak effects and lie in non-coding gene-regulatory elements where we lack a complete understanding of how single nucleotide alterations modulate transcriptional processes to affect human phenotypes. To address this, we measured the activity of 221,412 trait-associated variants that had been statistically fine-mapped using a Massively Parallel Reporter Assay (MPRA) in 5 diverse cell-types. We show that MPRA is able to discriminate between likely causal variants and controls, identifying 12,025 regulatory variants with high precision. Although the effects of these variants largely agree with orthogonal measures of function, only 69% can plausibly be explained by the disruption of a known transcription factor (TF) binding motif. We dissect the mechanisms of 136 variants using saturation mutagenesis and assign impacted TFs for 91% of variants without a clear canonical mechanism. Finally, we provide evidence that epistasis is prevalent for variants in close proximity and identify multiple functional variants on the same haplotype at a small, but important, subset of trait-associated loci. Overall, our study provides a systematic functional characterization of likely causal common variants underlying complex and molecular human traits, enabling new insights into the regulatory grammar underlying disease risk.

Genotype sampling for deep-learning assisted experimental mapping of a combinatorially complete fitness landscape

MOTIVATION

Experimental characterization of fitness landscapes, which map genotypes onto fitness, is important for both evolutionary biology and protein engineering. It faces a fundamental obstacle in the astronomical number of genotypes whose fitness needs to be measured for any one protein. Deep learning may help to predict the fitness of many genotypes from a smaller neural network training sample of genotypes with experimentally measured fitness. Here I use a recently published experimentally mapped fitness landscape of more than 260 000 protein genotypes to ask how such sampling is best performed.

RESULTS

I show that multilayer perceptrons, recurrent neural networks, convolutional networks, and transformers, can explain more than 90% of fitness variance in the data. In addition, 90% of this performance is reached with a training sample comprising merely ≈103 sequences. Generalization to unseen test data is best when training data is sampled randomly and uniformly, or sampled to minimize the number of synonymous sequences. In contrast, sampling to maximize sequence diversity or codon usage bias reduces performance substantially. These observations hold for more than one network architecture. Simple sampling strategies may perform best when training deep learning neural networks to map fitness landscapes from experimental data.

AVAILABILITY AND IMPLEMENTATION

The fitness landscape data analyzed here is publicly available as described previously (Papkou et al. 2023). All code used to analyze this landscape is publicly available at https://github.com/andreas-wagner-uzh/fitness_landscape_sampling.

How DNA encodes the start of transcription

A deep-learning model reveals the rules that define transcription initiation.

Sequence basis of transcription initiation in the human genome

Transcription initiation is a process that is essential to ensuring the proper function of any gene, yet we still lack a unified understanding of sequence patterns and rules that explain most transcription start sites in the human genome. By predicting transcription initiation at base-pair resolution from sequences with a deep learning-inspired explainable model called Puffin, we show that a small set of simple rules can explain transcription initiation at most human promoters. We identify key sequence patterns that contribute to human promoter activity, each activating transcription with distinct position-specific effects. Furthermore, we explain the sequence basis of bidirectional transcription at promoters, identify the links between promoter sequence and gene expression variation across cell types, and explore the conservation of sequence determinants of transcription initiation across mammalian species.

Deep learning the cis-regulatory code for gene expression in selected model plants

Elucidating the relationship between non-coding regulatory element sequences and gene expression is crucial for understanding gene regulation and genetic variation. We explored this link with the training of interpretable deep learning models predicting gene expression profiles from gene flanking regions of the plant species Arabidopsis thaliana, Solanum lycopersicum, Sorghum bicolor, and Zea mays. With over 80% accuracy, our models enabled predictive feature selection, highlighting e.g. the significant role of UTR regions in determining gene expression levels. The models demonstrated remarkable cross-species performance, effectively identifying both conserved and species-specific regulatory sequence features and their predictive power for gene expression. We illustrated the application of our approach by revealing causal links between genetic variation and gene expression changes across fourteen tomato genomes. Lastly, our models efficiently predicted genotype-specific expression of key functional gene groups, exemplified by underscoring known phenotypic and metabolic differences between Solanum lycopersicum and its wild, drought-resistant relative, Solanum pennellii.

The dynamic landscape of enhancer-derived RNA during mouse early embryo development

Enhancer-derived RNAs (eRNAs) play critical roles in diverse biological processes by facilitating their target gene expression. However, the abundance and function of eRNAs in early embryos are not clear. Here, we present a comprehensive eRNA atlas by systematically integrating publicly available datasets of mouse early embryos. We characterize the transcriptional and regulatory network of eRNAs and show that different embryo developmental stages have distinct eRNA expression and regulatory profiles. Paternal eRNAs are activated asymmetrically during zygotic genome activation (ZGA). Moreover, we identify an eRNA, MZGAe1, which plays an important function in regulating mouse ZGA and early embryo development. MZGAe1 knockdown leads to a developmental block from 2-cell embryo to blastocyst. We create an online data portal, M2ED2, to query and visualize eRNA expression and regulation. Our study thus provides a systematic landscape of eRNA and reveals the important role of eRNAs in regulating mouse early embryo development.

Epigenomic insights into common human disease pathology

The epigenome-the chemical modifications and chromatin-related packaging of the genome-enables the same genetic template to be activated or repressed in different cellular settings. This multi-layered mechanism facilitates cell-type specific function by setting the local sequence and 3D interactive activity level. Gene transcription is further modulated through the interplay with transcription factors and co-regulators. The human body requires this epigenomic apparatus to be precisely installed throughout development and then adequately maintained during the lifespan. The causal role of the epigenome in human pathology, beyond imprinting disorders and specific tumour suppressor genes, was further brought into the spotlight by large-scale sequencing projects identifying that mutations in epigenomic machinery genes could be critical drivers in both cancer and developmental disorders. Abrogation of this cellular mechanism is providing new molecular insights into pathogenesis. However, deciphering the full breadth and implications of these epigenomic changes remains challenging. Knowledge is accruing regarding disease mechanisms and clinical biomarkers, through pathogenically relevant and surrogate tissue analyses, respectively. Advances include consortia generated cell-type specific reference epigenomes, high-throughput DNA methylome association studies, as well as insights into ageing-related diseases from biological 'clocks' constructed by machine learning algorithms. Also, 3rd-generation sequencing is beginning to disentangle the complexity of genetic and DNA modification haplotypes. Cell-free DNA methylation as a cancer biomarker has clear clinical utility and further potential to assess organ damage across many disorders. Finally, molecular understanding of disease aetiology brings with it the opportunity for exact therapeutic alteration of the epigenome through CRISPR-activation or inhibition.

Dissection of core promoter syntax through single nucleotide resolution modeling of transcription initiation

Our understanding of how the DNA sequences of cis-regulatory elements encode transcription initiation patterns remains limited. Here we introduce CLIPNET, a deep learning model trained on population-scale PRO-cap data that accurately predicts the position and quantity of transcription initiation with single nucleotide resolution from DNA sequence. Interpretation of CLIPNET revealed a complex regulatory syntax consisting of DNA-protein interactions in five major positions between -200 and +50 bp relative to the transcription start site, as well as more subtle positional preferences among different transcriptional activators. Transcriptional activator and core promoter motifs occupy different positions and play distinct roles in regulating initiation, with the former driving initiation quantity and the latter initiation position. We identified core promoter motifs that explain initiation patterns in the majority of promoters and enhancers, including DPR motifs and AT-rich TBP binding sequences in TATA-less promoters. Our results provide insights into the sequence architecture governing transcription initiation.

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.

BERT-TFBS: a novel BERT-based model for predicting transcription factor binding sites by transfer learning

Transcription factors (TFs) are proteins essential for regulating genetic transcriptions by binding to transcription factor binding sites (TFBSs) in DNA sequences. Accurate predictions of TFBSs can contribute to the design and construction of metabolic regulatory systems based on TFs. Although various deep-learning algorithms have been developed for predicting TFBSs, the prediction performance needs to be improved. This paper proposes a bidirectional encoder representations from transformers (BERT)-based model, called BERT-TFBS, to predict TFBSs solely based on DNA sequences. The model consists of a pre-trained BERT module (DNABERT-2), a convolutional neural network (CNN) module, a convolutional block attention module (CBAM) and an output module. The BERT-TFBS model utilizes the pre-trained DNABERT-2 module to acquire the complex long-term dependencies in DNA sequences through a transfer learning approach, and applies the CNN module and the CBAM to extract high-order local features. The proposed model is trained and tested based on 165 ENCODE ChIP-seq datasets. We conducted experiments with model variants, cross-cell-line validations and comparisons with other models. The experimental results demonstrate the effectiveness and generalization capability of BERT-TFBS in predicting TFBSs, and they show that the proposed model outperforms other deep-learning models. The source code for BERT-TFBS is available at https://github.com/ZX1998-12/BERT-TFBS.

(Alternative) transcription start sites as regulators of RNA processing

Alternative transcription start site usage (ATSS) is a widespread regulatory strategy that enables genes to choose between multiple genomic loci for initiating transcription. This mechanism is tightly controlled during development and is often altered in disease states. In this review, we examine the growing evidence highlighting a role for transcription start sites (TSSs) in the regulation of mRNA isoform selection during and after transcription. We discuss how the choice of transcription initiation sites influences RNA processing and the importance of this crosstalk for cell identity and organism function. We also speculate on possible mechanisms underlying the integration of transcriptional and post-transcriptional processes.

Gastrulation-stage gene expression in Nipbl+/- mouse embryos foreshadows the development of syndromic birth defects

In animal models, Nipbl deficiency phenocopies gene expression changes and birth defects seen in Cornelia de Lange syndrome, the most common cause of which is Nipbl haploinsufficiency. Previous studies in Nipbl+/- mice suggested that heart development is abnormal as soon as cardiogenic tissue is formed. To investigate this, we performed single-cell RNA sequencing on wild-type and Nipbl+/- mouse embryos at gastrulation and early cardiac crescent stages. Nipbl+/- embryos had fewer mesoderm cells than wild-type and altered proportions of mesodermal cell subpopulations. These findings were associated with underexpression of genes implicated in driving specific mesodermal lineages. In addition, Nanog was found to be overexpressed in all germ layers, and many gene expression changes observed in Nipbl+/- embryos could be attributed to Nanog overexpression. These findings establish a link between Nipbl deficiency, Nanog overexpression, and gene expression dysregulation/lineage misallocation, which ultimately manifest as birth defects in Nipbl+/- animals and Cornelia de Lange syndrome.

A deep profile of gene expression across 18 human cancers

Clinically and biologically valuable information may reside untapped in large cancer gene expression data sets. Deep unsupervised learning has the potential to extract this information with unprecedented efficacy but has thus far been hampered by a lack of biological interpretability and robustness. Here, we present DeepProfile, a comprehensive framework that addresses current challenges in applying unsupervised deep learning to gene expression profiles. We use DeepProfile to learn low-dimensional latent spaces for 18 human cancers from 50,211 transcriptomes. DeepProfile outperforms existing dimensionality reduction methods with respect to biological interpretability. Using DeepProfile interpretability methods, we show that genes that are universally important in defining the latent spaces across all cancer types control immune cell activation, while cancer type-specific genes and pathways define molecular disease subtypes. By linking DeepProfile latent variables to secondary tumor characteristics, we discover that tumor mutation burden is closely associated with the expression of cell cycle-related genes. DNA mismatch repair and MHC class II antigen presentation pathway expression, on the other hand, are consistently associated with patient survival. We validate these results through Kaplan-Meier analyses and nominate tumor-associated macrophages as an important source of survival-correlated MHC class II transcripts. Our results illustrate the power of unsupervised deep learning for discovery of novel cancer biology from existing gene expression data.

Genomic context sensitizes regulatory elements to genetic disruption

Enhancer function is frequently investigated piecemeal using truncated reporter assays or single deletion analysis. Thus it remains unclear to what extent enhancer function at native loci relies on surrounding genomic context. Using the Big-IN technology for targeted integration of large DNAs, we analyzed the regulatory architecture of the murine Igf2/H19 locus, a paradigmatic model of enhancer selectivity. We assembled payloads containing a 157-kb functional Igf2/H19 locus and engineered mutations to genetically direct CTCF occupancy at the imprinting control region (ICR) that switches the target gene of the H19 enhancer cluster. Contrasting activity of payloads delivered at the endogenous Igf2/H19 locus or ectopically at Hprt revealed that the Igf2/H19 locus includes additional, previously unknown long-range regulatory elements. Exchanging components of the Igf2/H19 locus with the well-studied Sox2 locus showed that the H19 enhancer cluster functioned poorly out of context, and required its native surroundings to activate Sox2 expression. Conversely, the Sox2 locus control region (LCR) could activate both Igf2 and H19 outside its native context, but its activity was only partially modulated by CTCF occupancy at the ICR. Analysis of regulatory DNA actuation across different cell types revealed that, while the H19 enhancers are tightly coordinated within their native locus, the Sox2 LCR acts more independently. We show that these enhancer clusters typify broader classes of loci genome-wide. Our results show that unexpected dependencies may influence even the most studied functional elements, and our synthetic regulatory genomics approach permits large-scale manipulation of complete loci to investigate the relationship between locus architecture and function.

GUANinE v1.0: Benchmark Datasets for Genomic AI Sequence-to-Function Models

Computational genomics increasingly relies on machine learning methods for genome interpretation, and the recent adoption of neural sequence-to-function models highlights the need for rigorous model specification and controlled evaluation, problems familiar to other fields of AI. Research strategies that have greatly benefited other fields - including benchmarking, auditing, and algorithmic fairness - are also needed to advance the field of genomic AI and to facilitate model development. Here we propose a genomic AI benchmark, GUANinE, for evaluating model generalization across a number of distinct genomic tasks. Compared to existing task formulations in computational genomics, GUANinE is large-scale, de-noised, and suitable for evaluating pretrained models. GUANinE v1.0 primarily focuses on functional genomics tasks such as functional element annotation and gene expression prediction, and it also draws upon connections to evolutionary biology through sequence conservation tasks. The current GUANinE tasks provide insight into the performance of existing genomic AI models and non-neural baselines, with opportunities to be refined, revisited, and broadened as the field matures. Finally, the GUANinE benchmark allows us to evaluate new self-supervised T5 models and explore the tradeoffs between tokenization and model performance, while showcasing the potential for self-supervision to complement existing pretraining procedures.

Overlapping binding sites underlie TF genomic occupancy

Sequence-specific DNA binding by transcription factors (TFs) is a crucial step in gene regulation. However, current high-throughput in vitro approaches cannot reliably detect lower affinity TF-DNA interactions, which play key roles in gene regulation. Here, we developed PADIT-seq ( p rotein a ffinity to D NA by in vitro transcription and RNA seq uencing) to assay TF binding preferences to all 10-bp DNA sequences at far greater sensitivity than prior approaches. The expanded catalogs of low affinity DNA binding sites for the human TFs HOXD13 and EGR1 revealed that nucleotides flanking high affinity DNA binding sites create overlapping lower affinity sites that together modulate TF genomic occupancy in vivo . Formation of such extended recognition sequences stems from an inherent property of TF binding sites to interweave each other and expands the genomic sequence space for identifying noncoding variants that directly alter TF binding.

One-Sentence Summary

Overlapping DNA binding sites underlie TF genomic occupancy through their inherent propensity to interweave each other.

Predicting cell population-specific gene expression from genomic sequence

Most regulatory elements, especially enhancer sequences, are cell population-specific. One could even argue that a distinct set of regulatory elements is what defines a cell population. However, discovering which non-coding regions of the DNA are essential in which context, and as a result, which genes are expressed, is a difficult task. Some computational models tackle this problem by predicting gene expression directly from the genomic sequence. These models are currently limited to predicting bulk measurements and mainly make tissue-specific predictions. Here, we present a model that leverages single-cell RNA-sequencing data to predict gene expression. We show that cell population-specific models outperform tissue-specific models, especially when the expression profile of a cell population and the corresponding tissue are dissimilar. Further, we show that our model can prioritize GWAS variants and learn motifs of transcription factor binding sites. We envision that our model can be useful for delineating cell population-specific regulatory elements.

EvoAug-TF: extending evolution-inspired data augmentations for genomic deep learning to TensorFlow

SUMMARY

Deep neural networks (DNNs) have been widely applied to predict the molecular functions of the non-coding genome. DNNs are data hungry and thus require many training examples to fit data well. However, functional genomics experiments typically generate limited amounts of data, constrained by the activity levels of the molecular function under study inside the cell. Recently, EvoAug was introduced to train a genomic DNN with evolution-inspired augmentations. EvoAug-trained DNNs have demonstrated improved generalization and interpretability with attribution analysis. However, EvoAug only supports PyTorch-based models, which limits its applications to a broad class of genomic DNNs based in TensorFlow. Here, we extend EvoAug's functionality to TensorFlow in a new package, we call EvoAug-TF. Through a systematic benchmark, we find that EvoAug-TF yields comparable performance with the original EvoAug package.

AVAILABILITY AND IMPLEMENTATION

EvoAug-TF is freely available for users and is distributed under an open-source MIT license. Researchers can access the open-source code on GitHub (https://github.com/p-koo/evoaug-tf). The pre-compiled package is provided via PyPI (https://pypi.org/project/evoaug-tf) with in-depth documentation on ReadTheDocs (https://evoaug-tf.readthedocs.io). The scripts for reproducing the results are available at (https://github.com/p-koo/evoaug-tf_analysis).

Evaluating the representational power of pre-trained DNA language models for regulatory genomics

The emergence of genomic language models (gLMs) offers an unsupervised approach to learn a wide diversity of cis-regulatory patterns in the non-coding genome without requiring labels of functional activity generated by wet-lab experiments. Previous evaluations have shown pre-trained gLMs can be leveraged to improve prediction performance across a broad range of regulatory genomics tasks, albeit using relatively simple benchmark datasets and baseline models. Since the gLMs in these studies were tested upon fine-tuning their weights for each downstream task, determining whether gLM representations embody a foundational understanding of cis-regulatory biology remains an open question. Here we evaluate the representational power of pre-trained gLMs to predict and interpret cell-type-specific functional genomics data that span DNA and RNA regulation. Our findings suggest that current gLMs do not offer substantial advantages over conventional machine learning approaches that use one-hot encoded sequences. This work highlights a major limitation with current gLMs, raising potential issues in conventional pre-training strategies for the non-coding genome.

Interpreting cis-regulatory mechanisms from genomic deep neural networks using surrogate models

Deep neural networks (DNNs) have greatly advanced the ability to predict genome function from sequence. Interpreting genomic DNNs in terms of biological mechanisms, however, remains difficult. Here we introduce SQUID, a genomic DNN interpretability framework based on surrogate modeling. SQUID approximates genomic DNNs in user-specified regions of sequence space using surrogate models, i.e., simpler models that are mechanistically interpretable. Importantly, SQUID removes the confounding effects that nonlinearities and heteroscedastic noise in functional genomics data can have on model interpretation. Benchmarking analysis on multiple genomic DNNs shows that SQUID, when compared to established interpretability methods, identifies motifs that are more consistent across genomic loci and yields improved single-nucleotide variant-effect predictions. SQUID also supports surrogate models that quantify epistatic interactions within and between cis-regulatory elements. SQUID thus advances the ability to mechanistically interpret genomic DNNs.

Uncovering co-regulatory modules and gene regulatory networks in the heart through machine learning-based analysis of large-scale epigenomic data

The availability of large-scale epigenomic data from various cell types and conditions has yielded valuable insights for evaluating and learning features predicting the co-binding of transcription factors (TF). However, prior attempts to develop models predicting motif co-occurrence lacked scalability for globally analyzing any motif combination or making cross-species predictions. Moreover, mapping co-regulatory modules (CRM) to gene regulatory networks (GRN) is crucial for understanding underlying function. Currently, no comprehensive pipeline exists for large-scale, rapid, and accurate CRM and GRN identification. In this study, we analyzed and evaluated different TF binding characteristics facilitating biologically significant co-binding to identify all potential clusters of co-binding TFs. We curated the UniBind database, containing ChIP-Seq data from over 1983 samples and 232 TFs, and implemented two machine learning models to predict CRMs and the potential regulatory networks they operate on. Two machine learning models, Convolution Neural Networks (CNN) and Random Forest Classifier(RFC), used to predict co-binding between TFs, were compared using precision-recall Receiver Operating Characteristic (ROC) curves. CNN outperformed RFC (AUC 0.94 vs. 0.88) and achieved higher F1 scores (0.938 vs. 0.872). The CRMs generated by the clustering algorithm were validated against ChipAtlas and MCOT, revealing additional motifs forming CRMs. We predicted 200k CRMs for 50k+ human genes, validated against recent CRM prediction methods with 100% overlap. Further, we narrowed our focus to study heart-related regulatory motifs, filtering the generated CRMs to report 1784 Cardiac CRMs containing at least four cardiac TFs. Identified cardiac CRMs revealed potential novel regulators like ARID3A and RXRB for SCAD, including known TFs like PPARG for F11R. Our findings highlight the importance of the NKX family of transcription factors in cardiac development and provide potential targets for further investigation in cardiac disease.

Evaluation and optimization of sequence-based gene regulatory deep learning models

Neural networks have emerged as immensely powerful tools in predicting functional genomic regions, notably evidenced by recent successes in deciphering gene regulatory logic. However, a systematic evaluation of how model architectures and training strategies impact genomics model performance is lacking. To address this gap, we held a DREAM Challenge where competitors trained models on a dataset of millions of random promoter DNA sequences and corresponding expression levels, experimentally determined in yeast, to best capture the relationship between regulatory DNA and gene expression. For a robust evaluation of the models, we designed a comprehensive suite of benchmarks encompassing various sequence types. While some benchmarks produced similar results across the top-performing models, others differed substantially. All top-performing models used neural networks, but diverged in architectures and novel training strategies, tailored to genomics sequence data. To dissect how architectural and training choices impact performance, we developed the Prix Fixe framework to divide any given model into logically equivalent building blocks. We tested all possible combinations for the top three models and observed performance improvements for each. The DREAM Challenge models not only achieved state-of-the-art results on our comprehensive yeast dataset but also consistently surpassed existing benchmarks on Drosophila and human genomic datasets. Overall, we demonstrate that high-quality gold-standard genomics datasets can drive significant progress in model development.

DNA-Diffusion: Leveraging Generative Models for Controlling Chromatin Accessibility and Gene Expression via Synthetic Regulatory Elements

The challenge of systematically modifying and optimizing regulatory elements for precise gene expression control is central to modern genomics and synthetic biology. Advancements in generative AI have paved the way for designing synthetic sequences with the aim of safely and accurately modulating gene expression. We leverage diffusion models to design context-specific DNA regulatory sequences, which hold significant potential toward enabling novel therapeutic applications requiring precise modulation of gene expression. Our framework uses a cell type-specific diffusion model to generate synthetic 200 bp regulatory elements based on chromatin accessibility across different cell types. We evaluate the generated sequences based on key metrics to ensure they retain properties of endogenous sequences: transcription factor binding site composition, potential for cell type-specific chromatin accessibility, and capacity for sequences generated by DNA diffusion to activate gene expression in different cell contexts using state-of-the-art prediction models. Our results demonstrate the ability to robustly generate DNA sequences with cell type-specific regulatory potential. DNA-Diffusion paves the way for revolutionizing a regulatory modulation approach to mammalian synthetic biology and precision gene therapy.

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.

Detection of new pioneer transcription factors as cell-type-specific nucleosome binders

Wrapping of DNA into nucleosomes restricts accessibility to DNA and may affect the recognition of binding motifs by transcription factors. A certain class of transcription factors, the pioneer transcription factors, can specifically recognize their DNA binding sites on nucleosomes, initiate local chromatin opening, and facilitate the binding of co-factors in a cell-type-specific manner. For the majority of human pioneer transcription factors, the locations of their binding sites, mechanisms of binding, and regulation remain unknown. We have developed a computational method to predict the cell-type-specific ability of transcription factors to bind nucleosomes by integrating ChIP-seq, MNase-seq, and DNase-seq data with details of nucleosome structure. We have demonstrated the ability of our approach in discriminating pioneer from canonical transcription factors and predicted new potential pioneer transcription factors in H1, K562, HepG2, and HeLa-S3 cell lines. Last, we systematically analyzed the interaction modes between various pioneer transcription factors and detected several clusters of distinctive binding sites on nucleosomal DNA.

DARDN: A Deep-Learning Approach for CTCF Binding Sequence Classification and Oncogenic Regulatory Feature Discovery

Characterization of gene regulatory mechanisms in cancer is a key task in cancer genomics. CCCTC-binding factor (CTCF), a DNA binding protein, exhibits specific binding patterns in the genome of cancer cells and has a non-canonical function to facilitate oncogenic transcription programs by cooperating with transcription factors bound at flanking distal regions. Identification of DNA sequence features from a broad genomic region that distinguish cancer-specific CTCF binding sites from regular CTCF binding sites can help find oncogenic transcription factors in a cancer type. However, the presence of long DNA sequences without localization information makes it difficult to perform conventional motif analysis. Here, we present DNAResDualNet (DARDN), a computational method that utilizes convolutional neural networks (CNNs) for predicting cancer-specific CTCF binding sites from long DNA sequences and employs DeepLIFT, a method for interpretability of deep learning models that explains the model's output in terms of the contributions of its input features. The method is used for identifying DNA sequence features associated with cancer-specific CTCF binding. Evaluation on DNA sequences associated with CTCF binding sites in T-cell acute lymphoblastic leukemia (T-ALL) and other cancer types demonstrates DARDN's ability in classifying DNA sequences surrounding cancer-specific CTCF binding from control constitutive CTCF binding and identifying sequence motifs for transcription factors potentially active in each specific cancer type. We identify potential oncogenic transcription factors in T-ALL, acute myeloid leukemia (AML), breast cancer (BRCA), colorectal cancer (CRC), lung adenocarcinoma (LUAD), and prostate cancer (PRAD). Our work demonstrates the power of advanced machine learning and feature discovery approach in finding biologically meaningful information from complex high-throughput sequencing data.

Comparative analysis of models in predicting the effects of SNPs on TF-DNA binding using large-scale in vitro and in vivo data

Non-coding variants associated with complex traits can alter the motifs of transcription factor (TF)-deoxyribonucleic acid binding. Although many computational models have been developed to predict the effects of non-coding variants on TF binding, their predictive power lacks systematic evaluation. Here we have evaluated 14 different models built on position weight matrices (PWMs), support vector machines, ordinary least squares and deep neural networks (DNNs), using large-scale in vitro (i.e. SNP-SELEX) and in vivo (i.e. allele-specific binding, ASB) TF binding data. Our results show that the accuracy of each model in predicting SNP effects in vitro significantly exceeds that achieved in vivo. For in vitro variant impact prediction, kmer/gkm-based machine learning methods (deltaSVM_HT-SELEX, QBiC-Pred) trained on in vitro datasets exhibit the best performance. For in vivo ASB variant prediction, DNN-based multitask models (DeepSEA, Sei, Enformer) trained on the ChIP-seq dataset exhibit relatively superior performance. Among the PWM-based methods, tRap demonstrates better performance in both in vitro and in vivo evaluations. In addition, we find that TF classes such as basic leucine zipper factors could be predicted more accurately, whereas those such as C2H2 zinc finger factors are predicted less accurately, aligning with the evolutionary conservation of these TF classes. We also underscore the significance of non-sequence factors such as cis-regulatory element type, TF expression, interactions and post-translational modifications in influencing the in vivo predictive performance of TFs. Our research provides valuable insights into selecting prioritization methods for non-coding variants and further optimizing such models.

EvoAug-TF: Extending evolution-inspired data augmentations for genomic deep learning to TensorFlow

Deep neural networks (DNNs) have been widely applied to predict the molecular functions of regulatory regions in the non-coding genome. DNNs are data hungry and thus require many training examples to fit data well. However, functional genomics experiments typically generate limited amounts of data, constrained by the activity levels of the molecular function under study inside the cell. Recently, EvoAug was introduced to train a genomic DNN with evolution-inspired augmentations. EvoAug-trained DNNs have demonstrated improved generalization and interpretability with attribution analysis. However, EvoAug only supports PyTorch-based models, which limits its applications to a broad class of genomic DNNs based in TensorFlow. Here, we extend EvoAug's functionality to TensorFlow in a new package we call EvoAug-TF. Through a systematic benchmark, we find that EvoAug-TF yields comparable performance with the original EvoAug package.

Availability

EvoAug-TF is freely available for users and is distributed under an open-source MIT license. Researchers can access the open-source code on GitHub (https://github.com/p-koo/evoaug-tf). The pre-compiled package is provided via PyPI (https://pypi.org/project/evoaug-tf) with in-depth documentation on ReadTheDocs (https://evoaug-tf.readthedocs.io). The scripts for reproducing the results are available at (https://github.com/p-koo/evoaug-tf_analysis).

Integrating Multi-Omics Data With EHR for Precision Medicine Using Advanced Artificial Intelligence

With the recent advancement of novel biomedical technologies such as high-throughput sequencing and wearable devices, multi-modal biomedical data ranging from multi-omics molecular data to real-time continuous bio-signals are generated at an unprecedented speed and scale every day. For the first time, these multi-modal biomedical data are able to make precision medicine close to a reality. However, due to data volume and the complexity, making good use of these multi-modal biomedical data requires major effort. Researchers and clinicians are actively developing artificial intelligence (AI) approaches for data-driven knowledge discovery and causal inference using a variety of biomedical data modalities. These AI-based approaches have demonstrated promising results in various biomedical and healthcare applications. In this review paper, we summarize the state-of-the-art AI models for integrating multi-omics data and electronic health records (EHRs) for precision medicine. We discuss the challenges and opportunities in integrating multi-omics data with EHRs and future directions. We hope this review can inspire future research and developing in integrating multi-omics data with EHRs for precision medicine.

JASPAR 2024: 20th anniversary of the open-access database of transcription factor binding profiles

JASPAR (https://jaspar.elixir.no/) is a widely-used open-access database presenting manually curated high-quality and non-redundant DNA-binding profiles for transcription factors (TFs) across taxa. In this 10th release and 20th-anniversary update, the CORE collection has expanded with 329 new profiles. We updated three existing profiles and provided orthogonal support for 72 profiles from the previous release's UNVALIDATED collection. Altogether, the JASPAR 2024 update provides a 20% increase in CORE profiles from the previous release. A trimming algorithm enhanced profiles by removing low information content flanking base pairs, which were likely uninformative (within the capacity of the PFM models) for TFBS predictions and modelling TF-DNA interactions. This release includes enhanced metadata, featuring a refined classification for plant TFs' structural DNA-binding domains. The new JASPAR collections prompt updates to the genomic tracks of predicted TF binding sites (TFBSs) in 8 organisms, with human and mouse tracks available as native tracks in the UCSC Genome browser. All data are available through the JASPAR web interface and programmatically through its API and the updated Bioconductor and pyJASPAR packages. Finally, a new TFBS extraction tool enables users to retrieve predicted JASPAR TFBSs intersecting their genomic regions of interest.