A protein activity assay to measure global transcription factor activity reveals determinants of chromatin accessibility

TIANA: transcription factors cooperativity inference analysis with neural attention

BACKGROUND

Growing evidence suggests that distal regulatory elements are essential for cellular function and states. The sequences within these distal elements, especially motifs for transcription factor binding, provide critical information about the underlying regulatory programs. However, cooperativities between transcription factors that recognize these motifs are nonlinear and multiplexed, rendering traditional modeling methods insufficient to capture the underlying mechanisms. Recent development of attention mechanism, which exhibit superior performance in capturing dependencies across input sequences, makes them well-suited to uncover and decipher intricate dependencies between regulatory elements.

RESULT

We present Transcription factors cooperativity Inference Analysis with Neural Attention (TIANA), a deep learning framework that focuses on interpretability. In this study, we demonstrated that TIANA could discover biologically relevant insights into co-occurring pairs of transcription factor motifs. Compared with existing tools, TIANA showed superior interpretability and robust performance in identifying putative transcription factor cooperativities from co-occurring motifs.

CONCLUSION

Our results suggest that TIANA can be an effective tool to decipher transcription factor cooperativities from distal sequence data. TIANA can be accessed through: https://github.com/rzzli/TIANA .

Multiomic profiling of transcription factor binding and function in human brain

Transcription factors (TFs) orchestrate gene expression programs crucial for brain function, but we lack detailed information about TF binding in human brain tissue. We generated a multiomic resource (ChIP-seq, ATAC-seq, RNA-seq, DNA methylation) on bulk tissues and sorted nuclei from several postmortem brain regions, including binding maps for more than 100 TFs. We demonstrate improved measurements of TF activity, including motif recognition and gene expression modeling, upon identification and removal of high TF occupancy regions. Further, predictive TF binding models demonstrate a bias for these high-occupancy sites. Neuronal TFs SATB2 and TBR1 bind unique regions depleted for such sites and promote neuronal gene expression. Binding sites for TFs, including TBR1 and PKNOX1, are enriched for risk variants associated with neuropsychiatric disorders, predominantly in neurons. This work, titled BrainTF, is a powerful resource for future studies seeking to understand the roles of specific TFs in regulating gene expression in the human brain.

A universal system for boosting gene expression in eukaryotic cell-lines

We demonstrate a transcriptional regulatory design algorithm that can boost expression in yeast and mammalian cell lines. The system consists of a simplified transcriptional architecture composed of a minimal core promoter and a synthetic upstream regulatory region (sURS) composed of up to three motifs selected from a list of 41 motifs conserved in the eukaryotic lineage. The sURS system was first characterized using an oligo-library containing 189,990 variants. We validate the resultant expression model using a set of 43 unseen sURS designs. The validation sURS experiments indicate that a generic set of grammar rules for boosting and attenuation may exist in yeast cells. Finally, we demonstrate that this generic set of grammar rules functions similarly in mammalian CHO-K1 and HeLa cells. Consequently, our work provides a design algorithm for boosting the expression of promoters used for expressing industrially relevant proteins in yeast and mammalian cell lines.

Hold out the genome: a roadmap to solving the cis-regulatory code

Gene expression is regulated by transcription factors that work together to read cis-regulatory DNA sequences. The 'cis-regulatory code' - how cells interpret DNA sequences to determine when, where and how much genes should be expressed - has proven to be exceedingly complex. Recently, advances in the scale and resolution of functional genomics assays and machine learning have enabled substantial progress towards deciphering this code. However, the cis-regulatory code will probably never be solved if models are trained only on genomic sequences; regions of homology can easily lead to overestimation of predictive performance, and our genome is too short and has insufficient sequence diversity to learn all relevant parameters. Fortunately, randomly synthesized DNA sequences enable testing a far larger sequence space than exists in our genomes, and designed DNA sequences enable targeted queries to maximally improve the models. As the same biochemical principles are used to interpret DNA regardless of its source, models trained on these synthetic data can predict genomic activity, often better than genome-trained models. Here we provide an outlook on the field, and propose a roadmap towards solving the cis-regulatory code by a combination of machine learning and massively parallel assays using synthetic DNA.

Sea-ATI unravels novel vocabularies of plant active cistrome

The cistrome consists of all cis-acting regulatory elements recognized by transcription factors (TFs). However, only a portion of the cistrome is active for TF binding in a specific tissue. Resolving the active cistrome in plants remains challenging. In this study, we report the assay sequential extraction assisted-active TF identification (sea-ATI), a low-input method that profiles the DNA sequences recognized by TFs in a target tissue. We applied sea-ATI to seven plant tissues to survey their active cistrome and generated 41 motif models, including 15 new models that represent previously unidentified cis-regulatory vocabularies. ATAC-seq and RNA-seq analyses confirmed the functionality of the cis-elements from the new models, in that they are actively bound in vivo, located near the transcription start site, and influence chromatin accessibility and transcription. Furthermore, comparing dimeric WRKY CREs between sea-ATI and DAP-seq libraries revealed that thermodynamics and genetic drifts cooperatively shaped their evolution. Notably, sea-ATI can identify not only positive but also negative regulatory cis-elements, thereby providing unique insights into the functional non-coding genome of plants.

ExplaiNN: interpretable and transparent neural networks for genomics

Deep learning models such as convolutional neural networks (CNNs) excel in genomic tasks but lack interpretability. We introduce ExplaiNN, which combines the expressiveness of CNNs with the interpretability of linear models. ExplaiNN can predict TF binding, chromatin accessibility, and de novo motifs, achieving performance comparable to state-of-the-art methods. Its predictions are transparent, providing global (cell state level) as well as local (individual sequence level) biological insights into the data. ExplaiNN can serve as a plug-and-play platform for pretrained models and annotated position weight matrices. ExplaiNN aims to accelerate the adoption of deep learning in genomic sequence analysis by domain experts.

PIM1 promotes hepatic conversion by suppressing reprogramming-induced ferroptosis and cell cycle arrest

Protein kinase-mediated phosphorylation plays a critical role in many biological processes. However, the identification of key regulatory kinases is still a great challenge. Here, we develop a trans-omics-based method, central kinase inference, to predict potentially key kinases by integrating quantitative transcriptomic and phosphoproteomic data. Using known kinases associated with anti-cancer drug resistance, the accuracy of our method denoted by the area under the curve is 5.2% to 29.5% higher than Kinase-Substrate Enrichment Analysis. We further use this method to analyze trans-omic data in hepatocyte maturation and hepatic reprogramming of human dermal fibroblasts, uncovering 5 kinases as regulators in the two processes. Further experiments reveal that a serine/threonine kinase, PIM1, promotes hepatic conversion and protects human dermal fibroblasts from reprogramming-induced ferroptosis and cell cycle arrest. This study not only reveals new regulatory kinases, but also provides a helpful method that might be extended to predict central kinases involved in other biological processes.

Protein kinase-mediated phosphorylation plays a critical role in many biological processes. Here the authors develop a trans-omics-based algorithm called Central Kinase Inference to integrate quantitative transcriptomic and phosphoproteomic data, finding that PIM1 promotes hepatic conversion by suppressing reprogramming-induced ferroptosis and cell cycle arrest.

Characterization of sequence determinants of enhancer function using natural genetic variation

Sequence variation in enhancers that control cell-type-specific gene transcription contributes significantly to phenotypic variation within human populations. However, it remains difficult to predict precisely the effect of any given sequence variant on enhancer function due to the complexity of DNA sequence motifs that determine transcription factor (TF) binding to enhancers in their native genomic context. Using F₁-hybrid cells derived from crosses between distantly related inbred strains of mice, we identified thousands of enhancers with allele-specific TF binding and/or activity. We find that genetic variants located within the central region of enhancers are most likely to alter TF binding and enhancer activity. We observe that the AP-1 family of TFs (Fos/Jun) are frequently required for binding of TEAD TFs and for enhancer function. However, many sequence variants outside of core motifs for AP-1 and TEAD also impact enhancer function, including sequences flanking core TF motifs and AP-1 half sites. Taken together, these data represent one of the most comprehensive assessments of allele-specific TF binding and enhancer function to date and reveal how sequence changes at enhancers alter their function across evolutionary timescales.

WhichTF is functionally important in your open chromatin data?

We present WhichTF, a computational method to identify functionally important transcription factors (TFs) from chromatin accessibility measurements. To rank TFs, WhichTF applies an ontology-guided functional approach to compute novel enrichment by integrating accessibility measurements, high-confidence pre-computed conservation-aware TF binding sites, and putative gene-regulatory models. Comparison with prior sheer abundance-based methods reveals the unique ability of WhichTF to identify context-specific TFs with functional relevance, including NF-κB family members in lymphocytes and GATA factors in cardiac cells. To distinguish the transcriptional regulatory landscape in closely related samples, we apply differential analysis and demonstrate its utility in lymphocyte, mesoderm developmental, and disease cells. We find suggestive, under-characterized TFs, such as RUNX3 in mesoderm development and GLI1 in systemic lupus erythematosus. We also find TFs known for stress response, suggesting routine experimental caveats that warrant careful consideration. WhichTF yields biological insight into known and novel molecular mechanisms of TF-mediated transcriptional regulation in diverse contexts, including human and mouse cell types, cell fate trajectories, and disease-associated cells.

Transcription factors (TFs), a class of DNA binding proteins, regulate tissue- and cell-type-specific expression of genes. Identifying the critical TFs in a given cellular context leads to investigating molecular regulatory mechanisms in development, differentiation, and disease. Because there are more than 1,500 human TFs, experimental measurements of genome-wide occupancy across all TFs have been challenging. While computational approaches play pivotal roles, most existing methods rely on statistical enrichment, focusing either on sequence motif similarity recognized by TFs or the similarity of the genomic region of interest with the previously characterized TF occupancy profile. Here we propose WhichTF as an alternative, incorporating curated biomedical knowledge from ontology and integrating it with the high-confidence prediction of conserved TF binding sites in user-provided genomic regions of interest. We develop a new WhichTF score to rank TFs and demonstrate its applicability across human and mouse cell types, cellular differentiation trajectories, and disease-associated cells.

Sequence determinants of human gene regulatory elements

DNA can determine where and when genes are expressed, but the full set of sequence determinants that control gene expression is unknown. Here, we measured the transcriptional activity of DNA sequences that represent an ~100 times larger sequence space than the human genome using massively parallel reporter assays (MPRAs). Machine learning models revealed that transcription factors (TFs) generally act in an additive manner with weak grammar and that most enhancers increase expression from a promoter by a mechanism that does not appear to involve specific TF–TF interactions. The enhancers themselves can be classified into three types: classical, closed chromatin and chromatin dependent. We also show that few TFs are strongly active in a cell, with most activities being similar between cell types. Individual TFs can have multiple gene regulatory activities, including chromatin opening and enhancing, promoting and determining transcription start site (TSS) activity, consistent with the view that the TF binding motif is the key atomic unit of gene expression.

Analysis of massively parallel reporter assays measuring the transcriptional activity of DNA sequences indicates that most transcription factor (TF) activity is additive and does not rely on specific TF–TF interactions. Individual TFs can have different gene regulatory activities.

CASCADE: high-throughput characterization of regulatory complex binding altered by non-coding variants

Non-coding DNA variants (NCVs) impact gene expression by altering binding sites for regulatory complexes. New high-throughput methods are needed to characterize the impact of NCVs on regulatory complexes. We developed CASCADE (Customizable Approach to Survey Complex Assembly at DNA Elements), an array-based high-throughput method to profile cofactor (COF) recruitment. CASCADE identifies DNA-bound transcription factor-cofactor (TF-COF) complexes in nuclear extracts and quantifies the impact of NCVs on their binding. We demonstrate CASCADE sensitivity in characterizing condition-specific recruitment of COFs p300 and RBBP5 (MLL subunit) to the CXCL10 promoter in lipopolysaccharide (LPS)-stimulated human macrophages and quantify the impact of all possible NCVs. To demonstrate applicability to NCV screens, we profile TF-COF binding to ~1,700 single-nucleotide polymorphism quantitative trait loci (SNP-QTLs) in human macrophages and identify perturbed ETS domain-containing complexes. CASCADE will facilitate high-throughput testing of molecular mechanisms of NCVs for diverse biological applications.

Bray et al. develop CASCADE, a method to profile transcription factor (TF)-cofactor (COF) complexes binding to DNA. They demonstrate the approach by profiling complex binding across the CXCL10 cytokine promoter and to ~1,700 single-nucleotide polymorphisms (SNPs). They anticipate that CASCADE can be applied to diverse biological systems to examine regulatory complex binding to DNA.

Proteogenomics of non-small cell lung cancer reveals molecular subtypes associated with specific therapeutic targets and immune evasion mechanisms

Despite major advancements in lung cancer treatment, long-term survival is still rare, and a deeper understanding of molecular phenotypes would allow the identification of specific cancer dependencies and immune evasion mechanisms. Here we performed in-depth mass spectrometry (MS)-based proteogenomic analysis of 141 tumors representing all major histologies of non-small cell lung cancer (NSCLC). We identified six distinct proteome subtypes with striking differences in immune cell composition and subtype-specific expression of immune checkpoints. Unexpectedly, high neoantigen burden was linked to global hypomethylation and complex neoantigens mapped to genomic regions, such as endogenous retroviral elements and introns, in immune-cold subtypes. Further, we linked immune evasion with LAG3 via STK11 mutation-dependent HNF1A activation and FGL1 expression. Finally, we develop a data-independent acquisition MS-based NSCLC subtype classification method, validate it in an independent cohort of 208 NSCLC cases and demonstrate its clinical utility by analyzing an additional cohort of 84 late-stage NSCLC biopsy samples.

Structural insights into the interaction between transcription factors and the nucleosome

In eukaryotic cells, DNA interacts with two main types of binding proteins: transcription factors and histones. Histones form the core of nucleosomes and display weak sequence preference owing to differences in bendability of different DNA sequences. By contrast, the affinity of transcription factors is highly dependent on DNA sequence - all sequences are bound with moderate affinity, but only few specific sequences are bound more tightly via molecular recognition of the DNA bases. Transcription factors can interact with nucleosomes directly by recognizing nucleosome-associated DNA and also indirectly by recruiting histone-modifying enzymes and nucleosome remodelers. These interactions result in sequence-dependent formation of a pattern of open and closed chromatin, where specific positions are occupied by transcription factors, histone-modifying enzymes, and modified histones. These patterns are then recognized by large DNA-associated macromolecular complexes such as cohesin and RNA polymerase II, which are involved in regulation of higher-order chromatin structure and transcription, respectively. Here, we review recent work that has provided structural and mechanistic insight into the interactions between all these classes of DNA-associated proteins.

Improving representations of genomic sequence motifs in convolutional networks with exponential activations

Deep convolutional neural networks (CNNs) trained on regulatory genomic sequences tend to build representations in a distributed manner, making it a challenge to extract learned features that are biologically meaningful, such as sequence motifs. Here we perform a comprehensive analysis on synthetic sequences to investigate the role that CNN activations have on model interpretability. We show that employing an exponential activation to first layer filters consistently leads to interpretable and robust representations of motifs compared to other commonly used activations. Strikingly, we demonstrate that CNNs with better test performance do not necessarily imply more interpretable representations with attribution methods. We find that CNNs with exponential activations significantly improve the efficacy of recovering biologically meaningful representations with attribution methods. We demonstrate these results generalise to real DNA sequences across several in vivo datasets. Together, this work demonstrates how a small modification to existing CNNs, i.e. setting exponential activations in the first layer, can significantly improve the robustness and interpretabilty of learned representations directly in convolutional filters and indirectly with attribution methods.

Occupancy maps of 208 chromatin-associated proteins in one human cell type

Transcription factors are DNA-binding proteins that have key roles in gene regulation^1,2. Genome-wide occupancy maps of transcriptional regulators are important for understanding gene regulation and its effects on diverse biological processes^3–6. However, only a minority of the more than 1,600 transcription factors encoded in the human genome has been assayed. Here we present, as part of the ENCODE (Encyclopedia of DNA Elements) project, data and analyses from chromatin immunoprecipitation followed by high-throughput sequencing (ChIP–seq) experiments using the human HepG2 cell line for 208 chromatin-associated proteins (CAPs). These comprise 171 transcription factors and 37 transcriptional cofactors and chromatin regulator proteins, and represent nearly one-quarter of CAPs expressed in HepG2 cells. The binding profiles of these CAPs form major groups associated predominantly with promoters or enhancers, or with both. We confirm and expand the current catalogue of DNA sequence motifs for transcription factors, and describe motifs that correspond to other transcription factors that are co-enriched with the primary ChIP target. For example, FOX family motifs are enriched in ChIP–seq peaks of 37 other CAPs. We show that motif content and occupancy patterns can distinguish between promoters and enhancers. This catalogue reveals high-occupancy target regions at which many CAPs associate, although each contains motifs for only a minority of the numerous associated transcription factors. These analyses provide a more complete overview of the gene regulatory networks that define this cell type, and demonstrate the usefulness of the large-scale production efforts of the ENCODE Consortium.

ChIP–seq and CETCh–seq data are used to analyse binding maps for 208 transcription factors and other chromatin-associated proteins in a single human cell type, providing a comprehensive catalogue of the transcription factor landscape and gene regulatory networks in these cells.

Polycomb- and Methylation-Independent Roles of EZH2 as a Transcription Activator

Enhancer of Zeste 2 (EZH2) is the enzymatic subunit of Polycomb Repressive Complex 2 (PRC2), which catalyzes histone H3 lysine 27 trimethylation (H3K27me3) at target promoters for gene silencing. Here, we report that EZH2 activates androgen receptor (AR) gene transcription through direct occupancy at its promoter. Importantly, this activating role of EZH2 is independent of PRC2 and its methyltransferase activities. Genome-wide assays revealed extensive EZH2 occupancy at promoters marked by either H3K27ac or H3K27me3, leading to gene activation or repression, respectively. Last, we demonstrate enhanced efficacy of enzymatic EZH2 inhibitors when used in combination with AR antagonists in blocking the dual roles of EZH2 and suppressing prostate cancer progression in vitro and in vivo. Taken together, our study reports EZH2 as a transcriptional activator, a key target of which is AR, and suggests a drug-combinatory approach to treat advanced prostate cancer.

Kim et al. report EZH2 as a transcriptional activator that directly induces AR gene expression in a Polycomb- and methylation-independent manner, providing a mechanism to escape enzymatic EZH2 inhibitors. Combination of inhibitors with AR-targeted therapies showed a strong synergy in blocking the EZH2 downstream pathways and suppressing prostate cancer progression.

Contributions of regulated transcription and mRNA decay to the dynamics of gene expression

Organisms have acquired sophisticated regulatory networks that control gene expression in response to cellular perturbations. Understanding of the mechanisms underlying the coordinated changes in gene expression in response to external and internal stimuli is a fundamental issue in biology. Recent advances in high-throughput technologies have enabled the measurement of diverse biological information, including gene expression levels, kinetics of gene expression, and interactions among gene expression regulatory molecules. By coupling these technologies with quantitative modeling, we can now uncover the biological roles and mechanisms of gene regulation at the system level. This review consists of two parts. First, we focus on the methods using uridine analogs that measure synthesis and decay rates of RNAs, which demonstrate how cells dynamically change the regulation of gene expression in response to both internal and external cues. Second, we discuss the underlying mechanisms of these changes in kinetics, including the functions of transcription factors and RNA-binding proteins. Overall, this review will help to clarify a system-level view of gene expression programs in cells. This article is categorized under: Regulatory RNAs/RNAi/Riboswitches > Regulatory RNAs RNA Turnover and Surveillance > Regulation of RNA Stability RNA Methods > RNA Analyses in vitro and In Silico.