Cis-regulatory Codes Research Articles

Decoding biology with massively parallel reporter assays and machine learning.

Massively parallel reporter assays (MPRAs) are powerful tools for quantifying the impacts of sequence variation on gene expression. Reading out molecular phenotypes with sequencing enables interrogating the impact of sequence variation beyond genome scale. Machine learning models integrate and codify information learned from MPRAs and enable generalization by predicting sequences outside the training data set. Models can provide a quantitative understanding of cis-regulatory codes controlling gene expression, enable variant stratification, and guide the design of synthetic regulatory elements for applications from synthetic biology to mRNA and gene therapy. This review focuses on cis-regulatory MPRAs, particularly those that interrogate cotranscriptional and post-transcriptional processes: alternative splicing, cleavage and polyadenylation, translation, and mRNA decay.

Unveil cis-acting combinatorial mRNA motifs by interpreting deep neural network.

Cis-acting mRNA elements play a key role in the regulation of mRNA stability and translation efficiency. Revealing the interactions of these elements and their impact plays a crucial role in understanding the regulation of the mRNA translation process, which supports the development of mRNA-based medicine or vaccines. Deep neural networks (DNN) can learn complex cis-regulatory codes from RNA sequences. However, extracting these cis-regulatory codes efficiently from DNN remains a significant challenge. Here, we propose a method based on our toolkit NeuronMotif and motif mutagenesis, which not only enables the discovery of diverse and high-quality motifs but also efficiently reveals motif interactions. By interpreting deep-learning models, we have discovered several crucial motifs that impact mRNA translation efficiency and stability, as well as some unknown motifs or motif syntax, offering novel insights for biologists. Furthermore, we note that it is challenging to enrich motif syntax in datasets composed of randomly generated sequences, and they may not contain sufficient biological signals. The source code and data used to produce the results and analyses presented in this manuscript are available from GitHub (https://github.com/WangLabTHU/combmotif).

Modeling alternative translation initiation sites in plants reveals evolutionarily conserved cis-regulatory codes in eukaryotes.

mRNA translation relies on identifying translation initiation sites (TISs) in mRNAs. Alternative TISs are prevalent across plant transcriptomes, but the mechanisms for their recognition are unclear. Using ribosome profiling and machine learning, we developed models for predicting alternative TISs in the tomato (Solanum lycopersicum). Distinct feature sets were predictive of AUG and nonAUG TISs in 5' untranslated regions and coding sequences, including a novel CU-rich sequence that promoted plant TIS activity, a translational enhancer found across dicots and monocots, and humans and viruses. Our results elucidate the mechanistic and evolutionary basis of TIS recognition, whereby cis-regulatory RNA signatures affect start site selection. The TIS prediction model provides global estimates of TISs to discover neglected protein-coding genes across plant genomes. The prevalence of cis-regulatory signatures across plant species, humans, and viruses suggests their broad and critical roles in reprogramming the translational landscape.

NeuronMotif: Deciphering cis-regulatory codes by layer-wise demixing of deep neural networks

Discovering DNA regulatory sequence motifs and their relative positions is vital to understanding the mechanisms of gene expression regulation. Although deep convolutional neural networks (CNNs) have achieved great success in predicting cis-regulatory elements, the discovery of motifs and their combinatorial patterns from these CNN models has remained difficult. We show that the main difficulty is due to the problem of multifaceted neurons which respond to multiple types of sequence patterns. Since existing interpretation methods were mainly designed to visualize the class of sequences that can activate the neuron, the resulting visualization will correspond to a mixture of patterns. Such a mixture is usually difficult to interpret without resolving the mixed patterns. We propose the NeuronMotif algorithm to interpret such neurons. Given any convolutional neuron (CN) in the network, NeuronMotif first generates a large sample of sequences capable of activating the CN, which typically consists of a mixture of patterns. Then, the sequences are "demixed" in a layer-wise manner by backward clustering of the feature maps of the involved convolutional layers. NeuronMotif can output the sequence motifs, and the syntax rules governing their combinations are depicted by position weight matrices organized in tree structures. Compared to existing methods, the motifs found by NeuronMotif have more matches to known motifs in the JASPAR database. The higher-order patterns uncovered for deep CNs are supported by the literature and ATAC-seq footprinting. Overall, NeuronMotif enables the deciphering of cis-regulatory codes from deep CNs and enhances the utility of CNN in genome interpretation.

The cis-regulatory codes of response to combined heat and drought stress in Arabidopsis thaliana.

Plants respond to their environment by dynamically modulating gene expression. A powerful approach for understanding how these responses are regulated is to integrate information about cis-regulatory elements (CREs) into models called cis-regulatory codes. Transcriptional response to combined stress is typically not the sum of the responses to the individual stresses. However, cis-regulatory codes underlying combined stress response have not been established. Here we modeled transcriptional response to single and combined heat and drought stress in Arabidopsis thaliana. We grouped genes by their pattern of response (independent, antagonistic and synergistic) and trained machine learning models to predict their response using putative CREs (pCREs) as features (median F-measure = 0.64). We then developed a deep learning approach to integrate additional omics information (sequence conservation, chromatin accessibility and histone modification) into our models, improving performance by 6.2%. While pCREs important for predicting independent and antagonistic responses tended to resemble binding motifs of transcription factors associated with heat and/or drought stress, important synergistic pCREs resembled binding motifs of transcription factors not known to be associated with stress. These findings demonstrate how in silico approaches can improve our understanding of the complex codes regulating response to combined stress and help us identify prime targets for future characterization.

Cis-Regulatory Code for Predicting Plant Cell-Type Transcriptional Response to High Salinity.

Multicellular organisms have diverse cell types with distinct roles in development and responses to the environment. At the transcriptional level, the differences in the environmental response between cell types are due to differences in regulatory programs. In plants, although cell-type environmental responses have been examined, it is unclear how these responses are regulated. Here, we identify a set of putative cis-regulatory elements (pCREs) enriched in the promoters of genes responsive to high-salinity stress in six Arabidopsis (Arabidopsis thaliana) root cell types. We then use these pCREs to establish cis-regulatory codes (i.e. models predicting whether a gene is responsive to high salinity for each cell type with machine learning). These pCRE-based models outperform models using in vitro binding data of 758 Arabidopsis transcription factors. Surprisingly, organ pCREs identified based on the whole-root high-salinity response can predict cell-type responses as well as pCREs derived from cell-type data, because organ and cell-type pCREs predict complementary subsets of high-salinity response genes. Our findings not only advance our understanding of the regulatory mechanisms of the plant spatial transcriptional response through cis-regulatory codes but also suggest broad applicability of the approach to any species, particularly those with little or no trans-regulatory data.

Mapping Gene Regulatory Networks in Drosophila Eye Development by Large-Scale Transcriptome Perturbations and Motif Inference

Genome control is operated by transcription factors (TFs) controlling their target genes by binding to promoters and enhancers. Conceptually, the interactions between TFs, their binding sites, and their functional targets are represented by gene regulatory networks (GRNs). Deciphering in vivo GRNs underlying organ development in an unbiased genome-wide setting involves identifying both functional TF-gene interactions and physical TF-DNA interactions. To reverse engineer the GRNs of eye development in Drosophila, we performed RNA-seq across 72 genetic perturbations and sorted cell types and inferred a coexpression network. Next, we derived direct TF-DNA interactions using computational motif inference, ultimately connecting 241 TFs to 5,632 direct target genes through 24,926 enhancers. Using this network, we found network motifs, cis-regulatory codes, and regulators of eye development. We validate the predicted target regions of Grainyhead by ChIP-seq and identify this factor as a general cofactor in the eye network, being bound to thousands of nucleosome-free regions.

Bayesian hierarchical model of protein-binding microarrayk-mer data reduces noise and identifies transcription factor subclasses and preferredk-mers

Sequence-specific transcription factors (TFs) regulate the expression of their target genes through interactions with specific DNA-binding sites in the genome. Data on TF-DNA binding specificities are essential for understanding how regulatory specificity is achieved. Numerous studies have used universal protein-binding microarray (PBM) technology to determine the in vitro binding specificities of hundreds of TFs for all possible 8 bp sequences (8mers). We have developed a Bayesian analysis of variance (ANOVA) model that decomposes these 8mer data into background noise, TF familywise effects and effects due to the particular TF. Adjusting for background noise improves PBM data quality and concordance with in vivo TF binding data. Moreover, our model provides simultaneous identification of TF subclasses and their shared sequence preferences, and also of 8mers bound preferentially by individual members of TF subclasses. Such results may aid in deciphering cis-regulatory codes and determinants of protein-DNA binding specificity. Source code, compiled code and R and Python scripts are available from http://thebrain.bwh.harvard.edu/hierarchicalANOVA. Supplementary data are available at Bioinformatics online.

LOESS correction for length variation in gene set-based genomic sequence analysis

Sequence analysis algorithms are often applied to sets of DNA, RNA or protein sequences to identify common or distinguishing features. Controlling for sequence length variation is critical to properly score sequence features and identify true biological signals rather than length-dependent artifacts. Several cis-regulatory module discovery algorithms exhibit a substantial dependence between DNA sequence score and sequence length. Our newly developed LOESS method is flexible in capturing diverse score-length relationships and is more effective in correcting DNA sequence scores for length-dependent artifacts, compared with four other approaches. Application of this method to genes co-expressed during Drosophila melanogaster embryonic mesoderm development or neural development scored by the Lever motif analysis algorithm resulted in successful recovery of their biologically validated cis-regulatory codes. The LOESS length-correction method is broadly applicable, and may be useful not only for more accurate inference of cis-regulatory codes, but also for detection of other types of patterns in biological sequences. Source code and compiled code are available from http://thebrain.bwh.harvard.edu/LM_LOESS/

Gene Set-Based Module Discovery Decodes cis-Regulatory Codes Governing Diverse Gene Expression across Human Multiple Tissues

Decoding transcriptional programs governing transcriptomic diversity across human multiple tissues is a major challenge in bioinformatics. To address this problem, a number of computational methods have focused on cis-regulatory codes driving overexpression or underexpression in a single tissue as compared to others. On the other hand, we recently proposed a different approach to mine cis-regulatory codes: starting from gene sets sharing common cis-regulatory motifs, the method screens for expression modules based on expression coherence. However, both approaches seem to be insufficient to capture transcriptional programs that control gene expression in a subset of all samples. Especially, this limitation would be serious when analyzing multiple tissue data. To overcome this limitation, we developed a new module discovery method termed BEEM (Biclusering-based Extraction of Expression Modules) in order to discover expression modules that are functional in a subset of tissues. We showed that, when applied to expression profiles of human multiple tissues, BEEM finds expression modules missed by two existing approaches that are based on the coherent expression and the single tissue-specific differential expression. From the BEEM results, we obtained new insights into transcriptional programs controlling transcriptomic diversity across various types of tissues. This study introduces BEEM as a powerful tool for decoding regulatory programs from a compendium of gene expression profiles.

Genome-wide discovery of human heart enhancers

The various organogenic programs deployed during embryonic development rely on the precise expression of a multitude of genes in time and space. Identifying the cis-regulatory elements responsible for this tightly orchestrated regulation of gene expression is an essential step in understanding the genetic pathways involved in development. We describe a strategy to systematically identify tissue-specific cis-regulatory elements that share combinations of sequence motifs. Using heart development as an experimental framework, we employed a combination of Gibbs sampling and linear regression to build a classifier that identifies heart enhancers based on the presence and/or absence of various sequence features, including known and putative transcription factor (TF) binding specificities. In distinguishing heart enhancers from a large pool of random noncoding sequences, the performance of our classifier is vastly superior to four commonly used methods, with an accuracy reaching 92% in cross-validation. Furthermore, most of the binding specificities learned by our method resemble the specificities of TFs widely recognized as key players in heart development and differentiation, such as SRF, MEF2, ETS1, SMAD, and GATA. Using our classifier as a predictor, a genome-wide scan identified over 40,000 novel human heart enhancers. Although the classifier used no gene expression information, these novel enhancers are strongly associated with genes expressed in the heart. Finally, in vivo tests of our predictions in mouse and zebrafish achieved a validation rate of 62%, significantly higher than what is expected by chance. These results support the existence of underlying cis-regulatory codes dictating tissue-specific transcription in mammalian genomes and validate our enhancer classifier strategy as a method to uncover these regulatory codes.

Uncovering cis Regulatory Codes Using Synthetic Promoter Shuffling

Revealing the spectrum of combinatorial regulation of transcription at individual promoters is essential for understanding the complex structure of biological networks. However, the computations represented by the integration of various molecular signals at complex promoters are difficult to decipher in the absence of simple cis regulatory codes. Here we synthetically shuffle the regulatory architecture — operator sequences binding activators and repressors — of a canonical bacterial promoter. The resulting library of complex promoters allows for rapid exploration of promoter encoded logic regulation. Among all possible logic functions, NOR and ANDN promoter encoded logics predominate. A simple transcriptional cis regulatory code determines both logics, establishing a straightforward map between promoter structure and logic phenotype. The regulatory code is determined solely by the type of transcriptional regulation combinations: two repressors generate a NOR: NOT (a OR b) whereas a repressor and an activator generate an ANDN: a AND NOT b. Three-input versions of both logics, having an additional repressor as an input, are also present in the library. The resulting complex promoters cover a wide dynamic range of transcriptional strengths. Synthetic promoter shuffling represents a fast and efficient method for exploring the spectrum of complex regulatory functions that can be encoded by complex promoters. From an engineering point of view, synthetic promoter shuffling enables the experimental testing of the functional properties of complex promoters that cannot necessarily be inferred ab initio from the known properties of the individual genetic components. Synthetic promoter shuffling may provide a useful experimental tool for studying naturally occurring promoter shuffling.

Expression-Guided In Silico Evaluation of Candidate Cis Regulatory Codes for Drosophila Muscle Founder Cells

While combinatorial models of transcriptional regulation can be inferred for metazoan systems from a priori biological knowledge, validation requires extensive and time-consuming experimental work. Thus, there is a need for computational methods that can evaluate hypothesized cis regulatory codes before the difficult task of experimental verification is undertaken. We have developed a novel computational framework (termed “CodeFinder”) that integrates transcription factor binding site and gene expression information to evaluate whether a hypothesized transcriptional regulatory model (TRM; i.e., a set of co-regulating transcription factors) is likely to target a given set of co-expressed genes. Our basic approach is to simultaneously predict cis regulatory modules (CRMs) associated with a given gene set and quantify the enrichment for combinatorial subsets of transcription factor binding site motifs comprising the hypothesized TRM within these predicted CRMs. As a model system, we have examined a TRM experimentally demonstrated to drive the expression of two genes in a sub-population of cells in the developing Drosophila mesoderm, the somatic muscle founder cells. This TRM was previously hypothesized to be a general mode of regulation for genes expressed in this cell population. In contrast, the present analyses suggest that a modified form of this cis regulatory code applies to only a subset of founder cell genes, those whose gene expression responds to specific genetic perturbations in a similar manner to the gene on which the original model was based. We have confirmed this hypothesis by experimentally discovering six (out of 12 tested) new CRMs driving expression in the embryonic mesoderm, four of which drive expression in founder cells.