Determinants of transcription factor regulatory range

GENet: A Graph-Based Model Leveraging Histone Marks and Transcription Factors for Enhanced Gene Expression Prediction

Understanding the regulatory mechanisms of gene expression is a crucial objective in genomics. Although the DNA sequence near the transcription start site (TSS) offers valuable insights, recent methods suggest that analyzing only the surrounding DNA may not suffice to accurately predict gene expression levels. We developed GENet (Gene Expression Network from Histone and Transcription Factor Integration), a novel approach that integrates essential regulatory signals from transcription factors and histone modifications into a graph-based model. GENet extends beyond simple DNA sequence analysis by incorporating additional layers of genetic control, which are vital for determining gene expression. Our method markedly enhances the prediction of mRNA levels compared to previous models that depend solely on DNA sequence data. The results underscore the significance of including comprehensive regulatory information in gene expression studies. GENet emerges as a promising tool for researchers, with potential applications extending from fundamental biological research to the development of medical therapies.

Causal gene regulatory analysis with RNA velocity reveals an interplay between slow and fast transcription factors

Single-cell expression dynamics, from differentiation trajectories or RNA velocity, have the potential to reveal causal links between transcription factors (TFs) and their target genes in gene regulatory networks (GRNs). However, existing methods either overlook these expression dynamics or necessitate that cells be ordered along a linear pseudotemporal axis, which is incompatible with branching trajectories. We introduce Velorama, an approach to causal GRN inference that represents single-cell differentiation dynamics as a directed acyclic graph of cells, constructed from pseudotime or RNA velocity measurements. Additionally, Velorama enables the estimation of the speed at which TFs influence target genes. Applying Velorama, we uncover evidence that the speed of a TF's interactions is tied to its regulatory function. For human corticogenesis, we find that slow TFs are linked to gliomas, while fast TFs are associated with neuropsychiatric diseases. We expect Velorama to become a critical part of the RNA velocity toolkit for investigating the causal drivers of differentiation and disease.

Meta-analysis of differential gene expression in lower motor neurons isolated by laser capture microdissection from post-mortem ALS spinal cords

Introduction

ALS is a fatal neurodegenerative disease for which underlying mechanisms are incompletely understood. The motor neuron is a central player in ALS pathogenesis but different transcriptome signatures have been derived from bulk analysis of post-mortem tissue and iPSC-derived motor neurons (iPSC-MNs).

Methods

This study performed a meta-analysis of six gene expression studies (microarray and RNA-seq) in which laser capture microdissection (LCM) was used to isolate lower motor neurons from post-mortem spinal cords of ALS and control (CTL) subjects. Differentially expressed genes (DEGs) with consistent ALS versus CTL expression differences across studies were identified.

Results

The analysis identified 222 ALS-increased DEGs (FDR <0.10, SMD >0.80) and 278 ALS-decreased DEGs (FDR <0.10, SMD < -0.80). ALS-increased DEGs were linked to PI3K-AKT signaling, innate immunity, inflammation, motor neuron differentiation and extracellular matrix. ALS-decreased DEGs were associated with the ubiquitin-proteosome system, microtubules, axon growth, RNA-binding proteins and synaptic membrane. ALS-decreased DEG mRNAs frequently interacted with RNA-binding proteins (e.g., FUS, HuR). The complete set of DEGs (increased and decreased) overlapped significantly with genes near ALS-associated SNP loci (p < 0.01). Transcription factor target motifs with increased proximity to ALS-increased DEGs were identified, most notably DNA elements predicted to interact with forkhead transcription factors (e.g., FOXP1) and motor neuron and pancreas homeobox 1 (MNX1). Some of these DNA elements overlie ALS-associated SNPs within known enhancers and are predicted to have genotype-dependent MNX1 interactions. DEGs were compared to those identified from SOD1-G93A mice and bulk spinal cord segments or iPSC-MNs from ALS patients. There was good correspondence with transcriptome changes from SOD1-G93A mice (r ≤ 0.408) but most DEGs were not differentially expressed in bulk spinal cords or iPSC-MNs and transcriptome-wide effect size correlations were weak (bulk tissue: r ≤ 0.207, iPSC-MN: r ≤ 0.037).

Conclusion

This study defines a robust transcriptome signature from LCM-based motor neuron studies of post-mortem tissue from ALS and CTL subjects. This signature differs from those obtained from analysis of bulk spinal cord segments and iPSC-MNs. Results provide insight into mechanisms underlying gene dysregulation in ALS and highlight connections between these mechanisms, ALS genetics, and motor neuron biology.

Affinity-tagged SMAD1 and SMAD5 mouse lines reveal transcriptional reprogramming mechanisms during early pregnancy

Endometrial decidualization, a prerequisite for successful pregnancies, relies on transcriptional reprogramming driven by progesterone receptor (PR) and bone morphogenetic protein (BMP)-SMAD1/SMAD5 signaling pathways. Despite their critical roles in early pregnancy, how these pathways intersect in reprogramming the endometrium into a receptive state remains unclear. To define how SMAD1 and/or SMAD5 integrate BMP signaling in the uterus during early pregnancy, we generated two novel transgenic mouse lines with affinity tags inserted into the endogenous SMAD1 and SMAD5 loci (Smad1HA/HA and Smad5PA/PA). By profiling the genome-wide distribution of SMAD1, SMAD5, and PR in the mouse uterus, we demonstrated the unique and shared roles of SMAD1 and SMAD5 during the window of implantation. We also showed the presence of a conserved SMAD1, SMAD5, and PR genomic binding signature in the uterus during early pregnancy. To functionally characterize the translational aspects of our findings, we demonstrated that SMAD1/5 knockdown in human endometrial stromal cells suppressed expressions of canonical decidual markers (IGFBP1, PRL, FOXO1) and PR-responsive genes (RORB, KLF15). Here, our studies provide novel tools to study BMP signaling pathways and highlight the fundamental roles of SMAD1/5 in mediating both BMP signaling pathways and the transcriptional response to progesterone (P4) during early pregnancy.

Affinity-tagged SMAD1 and SMAD5 mouse lines reveal transcriptional reprogramming mechanisms during early pregnancy

Endometrial decidualization, a prerequisite for successful pregnancies, relies on transcriptional reprogramming driven by progesterone receptor (PR) and bone morphogenetic protein (BMP)-SMAD1/SMAD5 signaling pathways. Despite their critical roles in early pregnancy, how these pathways intersect in reprogramming the endometrium into a receptive state remains unclear. To define how SMAD1 and/or SMAD5 integrate BMP signaling in the uterus during early pregnancy, we generated two novel transgenic mouse lines with affinity tags inserted into the endogenous SMAD1 and SMAD5 loci (Smad1HA/HA and Smad5PA/PA). By profiling the genome-wide distribution of SMAD1, SMAD5, and PR in the mouse uterus, we demonstrated the unique and shared roles of SMAD1 and SMAD5 during the window of implantation. We also showed the presence of a conserved SMAD1, SMAD5, and PR genomic binding signature in the uterus during early pregnancy. To functionally characterize the translational aspects of our findings, we demonstrated that SMAD1/5 knockdown in human endometrial stromal cells suppressed expressions of canonical decidual markers (IGFBP1, PRL, FOXO1) and PR-responsive genes (RORB, KLF15). Here, our studies provide novel tools to study BMP signaling pathways and highlight the fundamental roles of SMAD1/5 in mediating both BMP signaling pathways and the transcriptional response to progesterone (P4) during early pregnancy.

Multimodal learning of noncoding variant effects using genome sequence and chromatin structure

MOTIVATION

A growing amount of noncoding genetic variants, including single-nucleotide polymorphisms, are found to be associated with complex human traits and diseases. Their mechanistic interpretation is relatively limited and can use the help from computational prediction of their effects on epigenetic profiles. However, current models often focus on local, 1D genome sequence determinants and disregard global, 3D chromatin structure that critically affects epigenetic events.

RESULTS

We find that noncoding variants of unexpected high similarity in epigenetic profiles, with regards to their relatively low similarity in local sequences, can be largely attributed to their proximity in chromatin structure. Accordingly, we have developed a multimodal deep learning scheme that incorporates both data of 1D genome sequence and 3D chromatin structure for predicting noncoding variant effects. Specifically, we have integrated convolutional and recurrent neural networks for sequence embedding and graph neural networks for structure embedding despite the resolution gap between the two types of data, while utilizing recent DNA language models. Numerical results show that our models outperform competing sequence-only models in predicting epigenetic profiles and their use of long-range interactions complement sequence-only models in extracting regulatory motifs. They prove to be excellent predictors for noncoding variant effects in gene expression and pathogenicity, whether in unsupervised "zero-shot" learning or supervised "few-shot" learning.

AVAILABILITY AND IMPLEMENTATION

Codes and data can be accessed at https://github.com/Shen-Lab/ncVarPred-1D3D and https://zenodo.org/record/7975777.

OGRE: calculate, visualize, and analyze overlap between genomic input regions and public annotations

BACKGROUND

Modern genome sequencing leads to an ever-growing collection of genomic annotations. Combining these elements with a set of input regions (e.g. genes) would yield new insights in genomic associations, such as those involved in gene regulation. The required data are scattered across different databases making a manual approach tiresome, unpractical, and prone to error. Semi-automatic approaches require programming skills in data parsing, processing, overlap calculation, and visualization, which most biomedical researchers lack. Our aim was to develop an automated tool providing all necessary algorithms, benefiting both bioinformaticians and researchers without bioinformatic training.

RESULTS

We developed overlapping annotated genomic regions (OGRE) as a comprehensive tool to associate and visualize input regions with genomic annotations. It does so by parsing regions of interest, mining publicly available annotations, and calculating possible overlaps between them. The user can thus identify location, type, and number of associated regulatory elements. Results are presented as easy to understand visualizations and result tables. We applied OGRE to recent studies and could show high reproducibility and potential new insights. To demonstrate OGRE's performance in terms of running time and output, we have conducted a benchmark and compared its features with similar tools.

CONCLUSIONS

OGRE's functions and built-in annotations can be applied as a downstream overlap association step, which is compatible with most genomic sequencing outputs, and can thus enrich pre-existing analyses pipelines. Compared to similar tools, OGRE shows competitive performance, offers additional features, and has been successfully applied to two recent studies. Overall, OGRE addresses the lack of tools for automatic analysis, local genomic overlap calculation, and visualization by providing an easy to use, end-to-end solution for both biologists and computational scientists.

Transfer learning enables predictions in network biology

Mapping gene networks requires large amounts of transcriptomic data to learn the connections between genes, which impedes discoveries in settings with limited data, including rare diseases and diseases affecting clinically inaccessible tissues. Recently, transfer learning has revolutionized fields such as natural language understanding1,2 and computer vision3 by leveraging deep learning models pretrained on large-scale general datasets that can then be fine-tuned towards a vast array of downstream tasks with limited task-specific data. Here, we developed a context-aware, attention-based deep learning model, Geneformer, pretrained on a large-scale corpus of about 30 million single-cell transcriptomes to enable context-specific predictions in settings with limited data in network biology. During pretraining, Geneformer gained a fundamental understanding of network dynamics, encoding network hierarchy in the attention weights of the model in a completely self-supervised manner. Fine-tuning towards a diverse panel of downstream tasks relevant to chromatin and network dynamics using limited task-specific data demonstrated that Geneformer consistently boosted predictive accuracy. Applied to disease modelling with limited patient data, Geneformer identified candidate therapeutic targets for cardiomyopathy. Overall, Geneformer represents a pretrained deep learning model from which fine-tuning towards a broad range of downstream applications can be pursued to accelerate discovery of key network regulators and candidate therapeutic targets.

Characterizing the targets of transcription regulators by aggregating ChIP-seq and perturbation expression data sets

Mapping the gene targets of chromatin-associated transcription regulators (TRs) is a major goal of genomics research. ChIP-seq of TRs and experiments that perturb a TR and measure the differential abundance of gene transcripts are a primary means by which direct relationships are tested on a genomic scale. It has been reported that there is a poor overlap in the evidence across gene regulation strategies, emphasizing the need for integrating results from multiple experiments. Although research consortia interested in gene regulation have produced a valuable trove of high-quality data, there is an even greater volume of TR-specific data throughout the literature. In this study, we show a workflow for the identification, uniform processing, and aggregation of ChIP-seq and TR perturbation experiments for the ultimate purpose of ranking human and mouse TR-target interactions. Focusing on an initial set of eight regulators (ASCL1, HES1, MECP2, MEF2C, NEUROD1, PAX6, RUNX1, and TCF4), we identified 497 experiments suitable for analysis. We used this corpus to examine data concordance, to identify systematic patterns of the two data types, and to identify putative orthologous interactions between human and mouse. We build upon commonly used strategies to forward a procedure for aggregating and combining these two genomic methodologies, assessing these rankings against independent literature-curated evidence. Beyond a framework extensible to other TRs, our work also provides empirically ranked TR-target listings, as well as transparent experiment-level gene summaries for community use.

MIRA: joint regulatory modeling of multimodal expression and chromatin accessibility in single cells

Rigorously comparing gene expression and chromatin accessibility in the same single cells could illuminate the logic of how coupling or decoupling of these mechanisms regulates fate commitment. Here we present MIRA, probabilistic multimodal models for integrated regulatory analysis, a comprehensive methodology that systematically contrasts transcription and accessibility to infer the regulatory circuitry driving cells along cell state trajectories. MIRA leverages topic modeling of cell states and regulatory potential modeling of individual gene loci. MIRA thereby represents cell states in an efficient and interpretable latent space, infers high-fidelity cell state trees, determines key regulators of fate decisions at branch points and exposes the variable influence of local accessibility on transcription at distinct loci. Applied to epidermal differentiation and embryonic brain development from two different multimodal platforms, MIRA revealed that early developmental genes were tightly regulated by local chromatin landscape whereas terminal fate genes were titrated without requiring extensive chromatin remodeling.

Tunable transcription factor library for robust quantification of regulatory properties in Escherichia coli

Predicting the quantitative regulatory function of transcription factors (TFs) based on factors such as binding sequence, binding location, and promoter type is not possible. The interconnected nature of gene networks and the difficulty in tuning individual TF concentrations make the isolated study of TF function challenging. Here, we present a library of Escherichia coli strains designed to allow for precise control of the concentration of individual TFs enabling the study of the role of TF concentration on physiology and regulation. We demonstrate the usefulness of this resource by measuring the regulatory function of the zinc‐responsive TF, ZntR, and the paralogous TF pair, GalR/GalS. For ZntR, we find that zinc alters ZntR regulatory function in a way that enables activation of the regulated gene to be robust with respect to ZntR concentration. For GalR and GalS, we are able to demonstrate that these paralogous TFs have fundamentally distinct regulatory roles beyond differences in binding affinity.

Spatial Features and Functional Implications of Plant 3D Genome Organization

The advent of high-throughput sequencing-based methods for chromatin conformation, accessibility, and immunoprecipitation assays has been a turning point in 3D genomics. Altogether, these new tools have been pushing upward the interpretation of pioneer cytogenetic evidence for a higher order in chromatin packing. Here, we review the latest development in our understanding of plant spatial genome structures and different levels of organization and discuss their functional implications. Then, we spotlight the complexity of organellar (i.e., mitochondria and plastids) genomes and discuss their 3D packing into nucleoids. Finally, we propose unaddressed research axes to investigate functional links between chromatin-like dynamics and transcriptional regulation within organellar nucleoids.

Human Cytomegalovirus IE2 Both Activates and Represses Initiation and Modulates Elongation in a Context-Dependent Manner

Human cytomegalovirus (HCMV) immediate-early 2 (IE2) protein is a multifunctional transcription factor that is essential for lytic HCMV infection. IE2 functions as an activator of viral early genes, negatively regulates its own promoter, and is required for viral replication. The mechanisms by which IE2 executes these distinct functions are incompletely understood. Using PRO-Seq, which profiles nascent transcripts, and a recently developed DFF-chromatin immunoprecipitation (DFF-ChIP; employs chromatin digestion by the endonuclease DNA fragmentation factor prior to IP) approach that resolves occupancy and local chromatin environment, we show that IE2 controls viral gene transcription in three distinct capacities during late HCMV infection and reveal mechanisms that involve direct binding of IE2 to viral DNA. IE2 represses a subset of viral promoters by binding within their core promoter regions and blocking the assembly of preinitiation complexes (PICs). Remarkably, IE2 forms a repressive complex at the major immediate-early promoter region involving direct association of IE2 with nucleosomes and TBP. IE2 stimulates transcription by binding nearby, but not within, core promoter regions. In addition, IE2 functions as a direct roadblock to transcription elongation. At one locus, this function of IE2 appears to be important for the synthesis of a spliced viral RNA. Consistent with the minimal observed effects of IE2 depletion on host gene transcription, IE2 does not functionally engage the host genome. Our results reveal mechanisms of transcriptional control by IE2, uncover a previously unknown function of IE2 as a Pol II elongation modulator, and demonstrate that DFF-ChIP is a useful tool for probing transcription factor occupancy and interactions between transcription factors and nucleosomes at high resolution.

Annotating regulatory elements by heterogeneous network embedding

MOTIVATION

Regulatory elements (REs), such as enhancers and promoters, are known as regulatory sequences functional in a heterogeneous regulatory network to control gene expression by recruiting transcription regulators and carrying genetic variants in a context specific way. Annotating those REs relies on costly and labor-intensive next-generation sequencing and RNA-guided editing technologies in many cellular contexts.

RESULTS

We propose a systematic Gene Ontology Annotation method for Regulatory Elements (RE-GOA) by leveraging the powerful word embedding in natural language processing. We first assemble a heterogeneous network by integrating context specific regulations, protein-protein interactions and gene ontology (GO) terms. Then we perform network embedding and associate regulatory elements with GO terms by assessing their similarity in a low dimensional vector space. With three applications, we show that RE-GOA outperforms existing methods in annotating TFs' binding sites from ChIP-seq data, in functional enrichment analysis of differentially accessible peaks from ATAC-seq data, and in revealing genetic correlation among phenotypes from their GWAS summary statistics data.

AVAILABILITY AND IMPLEMENTATION

The source code and the systematic RE annotation for human and mouse are available at https://github.com/AMSSwanglab/RE-GOA.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Known allosteric proteins have central roles in genetic disease

Allostery is a form of protein regulation, where ligands that bind sites located apart from the active site can modify the activity of the protein. The molecular mechanisms of allostery have been extensively studied, because allosteric sites are less conserved than active sites, and drugs targeting them are more specific than drugs binding the active sites. Here we quantify the importance of allostery in genetic disease. We show that 1) known allosteric proteins are central in disease networks, contribute to genetic disease and comorbidities much more than non-allosteric proteins, and there is an association between being allosteric and involvement in disease; 2) they are enriched in many major disease types like hematopoietic diseases, cardiovascular diseases, cancers, diabetes, or diseases of the central nervous system; 3) variants from cancer genome-wide association studies are enriched near allosteric proteins, indicating their importance to polygenic traits; and 4) the importance of allosteric proteins in disease is due, at least partly, to their central positions in protein-protein interaction networks, and less due to their dynamical properties.

Progesterone Inhibits the Establishment of Activation-Associated Chromatin During TH1 Differentiation

TH1-mediated diseases such as multiple sclerosis (MS) and rheumatoid arthritis (RA) improve during pregnancy, coinciding with increasing levels of the pregnancy hormone progesterone (P4), highlighting P4 as a potential mediator of this immunomodulation. Here, we performed detailed characterization of how P4 affects the chromatin and transcriptomic landscape during early human TH1 differentiation, utilizing both ATAC-seq and RNA-seq. Time series analysis of the earlier events (0.5-24 hrs) during TH1 differentiation revealed that P4 counteracted many of the changes induced during normal differentiation, mainly by downregulating key regulatory genes and their upstream transcription factors (TFs) involved in the initial T-cell activation. Members of the AP-1 complex such as FOSL1, FOSL2, JUN and JUNB were particularly affected, in both in promoters and in distal regulatory elements. Moreover, the changes induced by P4 were significantly enriched for disease-associated changes related to both MS and RA, revealing several shared upstream TFs, where again JUN was highlighted to be of central importance. Our findings support an immune regulatory role for P4 during pregnancy by impeding T-cell activation, a crucial checkpoint during pregnancy and in T-cell mediated diseases, and a central event prior to T-cell lineage commitment. Indeed, P4 is emerging as a likely candidate involved in disease modulation during pregnancy and further studies evaluating P4 as a potential treatment option are needed.

Molecular competition can shape enhancer activity in the Drosophila embryo

Transgenic reporters allow the measurement of regulatory DNA activity in vivo and consequently have long been useful tools for studying enhancers. Despite their utility, few studies have investigated the effects these reporters may have on the expression of other genes. Understanding these effects is required to accurately interpret reporter data and characterize gene regulatory mechanisms. By measuring the expression of Kruppel (Kr) enhancer reporters in live Drosophila embryos, we find reporters inhibit one another's expression and that of a nearby endogenous gene. Using synthetic transcription factor (TF) binding site arrays, we present evidence that competition for TFs is partially responsible for the observed transcriptional inhibition. We develop a simple thermodynamic model that predicts competition of the measured magnitude specifically when TF binding is restricted to distinct nuclear subregions. Our findings underline an unexpected role of the non-homogenous nature of the nucleus in regulating gene expression.

Nucleus Type-Specific DNA Methylomics Reveals Epigenetic "Memory" of Prior Adaptation in Skeletal Muscle

Using a mouse model of conditional and inducible in vivo fluorescent myonuclear labeling (HSA-GFP), sorting purification of nuclei, low-input reduced representation bisulfite sequencing (RRBS), and a translatable and reversible model of exercise (progressive weighted wheel running, PoWeR), we provide the first nucleus type-specific epigenetic information on skeletal muscle adaptation and detraining. Adult (>4 mo) HSA-GFP mice performed PoWeR for 8 wk then detrained for 12 wk; age-matched untrained mice were used to control for the long duration of the study. Myonuclei and interstitial nuclei from plantaris muscles were isolated for RRBS. Relative to untrained, PoWeR caused similar myonuclear CpG hypo- and hyper-methylation of promoter regions and substantial hypomethylation in interstitial nuclear promoters. Over-representation analysis of promoters revealed a larger number of hyper- versus hypo-methylated pathways in both nuclear populations after training and evidence for reciprocal regulation of methylation between nucleus types, with hypomethylation of promoter regions in Wnt signaling-related genes in myonuclei and hypermethylation in interstitial nuclei. After 12 wk of detraining, promoter CpGs in documented muscle remodeling-associated genes and pathways that were differentially methylated immediately after PoWeR were persistently differentially methylated in myonuclei, along with long-term promoter hypomethylation in interstitial nuclei. No enduring gene expression changes in muscle tissue were observed using RNA-sequencing. Upon 4 wk of retraining, mice that trained previously grew more at the whole muscle and fiber type-specific cellular level than training naïve mice, with no difference in myonuclear number. Muscle nuclei have a methylation epi-memory of prior training that may augment muscle adaptability to retraining.

Filtering of Data-Driven Gene Regulatory Networks Using Drosophila melanogaster as a Case Study

Gene Regulatory Networks (GRNs) allow the study of regulation of gene expression of whole genomes. Among the most relevant advantages of using networks to depict this key process, there is the visual representation of large amounts of information and the application of graph theory to generate new knowledge. Nonetheless, despite the many uses of GRNs, it is still difficult and expensive to assign Transcription Factors (TFs) to the regulation of specific genes. ChIP-Seq allows the determination of TF Binding Sites (TFBSs) over whole genomes, but it is still an expensive technique that can only be applied one TF at a time and requires replicates to reduce its noise. Once TFBSs are determined, the assignment of each TF and its binding sites to the regulation of specific genes is not trivial, and it is often performed by carrying out site-specific experiments that are unfeasible to perform in all possible binding sites. Here, we addressed these relevant issues with a two-step methodology using Drosophila melanogaster as a case study. First, our protocol starts by gathering all transcription factor binding sites (TFBSs) determined with ChIP-Seq experiments available at ENCODE and FlyBase. Then each TFBS is used to assign TFs to the regulation of likely target genes based on the TFBS proximity to the transcription start site of all genes. In the final step, to try to select the most likely regulatory TF from those previously assigned to each gene, we employ GENIE3, a random forest-based method, and more than 9,000 RNA-seq experiments from D. melanogaster. Following, we employed known TF protein-protein interactions to estimate the feasibility of regulatory events in our filtered networks. Finally, we show how known interactions between co-regulatory TFs of each gene increase after the second step of our approach, and thus, the consistency of the TF-gene assignment. Also, we employed our methodology to create a network centered on the Drosophila melanogaster gene Hr96 to demonstrate the role of this transcription factor on mitochondrial gene regulation.

UniBind: maps of high-confidence direct TF-DNA interactions across nine species

BACKGROUND

Transcription factors (TFs) bind specifically to TF binding sites (TFBSs) at cis-regulatory regions to control transcription. It is critical to locate these TF-DNA interactions to understand transcriptional regulation. Efforts to predict bona fide TFBSs benefit from the availability of experimental data mapping DNA binding regions of TFs (chromatin immunoprecipitation followed by sequencing - ChIP-seq).

RESULTS

In this study, we processed ~ 10,000 public ChIP-seq datasets from nine species to provide high-quality TFBS predictions. After quality control, it culminated with the prediction of ~ 56 million TFBSs with experimental and computational support for direct TF-DNA interactions for 644 TFs in > 1000 cell lines and tissues. These TFBSs were used to predict > 197,000 cis-regulatory modules representing clusters of binding events in the corresponding genomes. The high-quality of the TFBSs was reinforced by their evolutionary conservation, enrichment at active cis-regulatory regions, and capacity to predict combinatorial binding of TFs. Further, we confirmed that the cell type and tissue specificity of enhancer activity was correlated with the number of TFs with binding sites predicted in these regions. All the data is provided to the community through the UniBind database that can be accessed through its web-interface ( https://unibind.uio.no/ ), a dedicated RESTful API, and as genomic tracks. Finally, we provide an enrichment tool, available as a web-service and an R package, for users to find TFs with enriched TFBSs in a set of provided genomic regions.

CONCLUSIONS

UniBind is the first resource of its kind, providing the largest collection of high-confidence direct TF-DNA interactions in nine species.

From bedside to bench: regulation of host factors in SARS-CoV-2 infection

The zoonotic coronavirus SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2), which causes COVID-19 (coronavirus disease-2019), has resulted in a pandemic. This has led to an urgent need to understand the molecular determinants of SARS-CoV-2 infection, factors associated with COVID-19 heterogeneity and severity, and therapeutic options for these patients. In this review, we discuss the role of host factors in SARS-CoV-2 infection and describe variations in host factor expression as mechanisms underlying the symptoms and severity of COVID-19. We focus on two host factors, angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2), implicated in SARS-CoV-2 infection. We also discuss genetic variants associated with COVID-19 severity revealed in selected patients and based on genome-wide association studies (GWASs). Furthermore, we highlight important advances in cell and chromatin biology, such as single-cell RNA and chromatin sequencing and chromosomal conformation assays, as methods that may aid in the discovery of viral-host interactions in COVID-19. Understanding how regulation of host factor genes varies in physiological and pathological states might explain the heterogeneity observed in SARS-CoV-2 infection, help identify pathways for therapeutic development, and identify patients most likely to progress to severe COVID-19.

Evidence of widespread, independent sequence signature for transcription factor cobinding

Transcription factors (TFs) are the vocabulary that genomes use to regulate gene expression and phenotypes. The interactions among TFs enrich this vocabulary and orchestrate diverse biological processes. Although simple models identify open chromatin and the presence of TF motifs as the two major contributors to TF binding patterns, it remains elusive what contributes to the in vivo TF cobinding landscape. In this study, we developed a machine learning algorithm to explore the contributors of the cobinding patterns. The algorithm substantially outperforms the state-of-the-field models for TF cobinding prediction. Game theory-based feature importance analysis reveals that, for most of the TF pairs we studied, independent motif sequences contribute one or more of the two TFs under investigation to their cobinding patterns. Such independent motif sequences include, but are not limited to, transcription initiation-related proteins and known TF complexes. We found the motif sequence signatures and the TFs are rarely mutual, corroborating a hierarchical and directional organization of the regulatory network and refuting the possibility of artifacts caused by shared sequence similarity with the TFs under investigation. We modeled such regulatory language with directed graphs, which reveal shared, global factors that are related to many binding and cobinding patterns.

Tumor suppressor p53: from engaging DNA to target gene regulation

The p53 transcription factor confers its potent tumor suppressor functions primarily through the regulation of a large network of target genes. The recent explosion of next generation sequencing protocols has enabled the study of the p53 gene regulatory network (GRN) and underlying mechanisms at an unprecedented depth and scale, helping us to understand precisely how p53 controls gene regulation. Here, we discuss our current understanding of where and how p53 binds to DNA and chromatin, its pioneer-like role, and how this affects gene regulation. We provide an overview of the p53 GRN and the direct and indirect mechanisms through which p53 affects gene regulation. In particular, we focus on delineating the ubiquitous and cell type-specific network of regulatory elements that p53 engages; reviewing our understanding of how, where, and when p53 binds to DNA and the mechanisms through which these events regulate transcription. Finally, we discuss the evolution of the p53 GRN and how recent work has revealed remarkable differences between vertebrates, which are of particular importance to cancer researchers using mouse models.

Integrative analyses of single-cell transcriptome and regulome using MAESTRO

We present Model-based AnalysEs of Transcriptome and RegulOme (MAESTRO), a comprehensive open-source computational workflow ( http://github.com/liulab-dfci/MAESTRO ) for the integrative analyses of single-cell RNA-seq (scRNA-seq) and ATAC-seq (scATAC-seq) data from multiple platforms. MAESTRO provides functions for pre-processing, alignment, quality control, expression and chromatin accessibility quantification, clustering, differential analysis, and annotation. By modeling gene regulatory potential from chromatin accessibilities at the single-cell level, MAESTRO outperforms the existing methods for integrating the cell clusters between scRNA-seq and scATAC-seq. Furthermore, MAESTRO supports automatic cell-type annotation using predefined cell type marker genes and identifies driver regulators from differential scRNA-seq genes and scATAC-seq peaks.