Predicting gene expression in T cell differentiation from histone modifications and transcription factor binding affinities by linear mixture models

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.

Differential Gene Expression Prediction by Ensemble Deep Networks on Histone Modification Data

Predicting differential gene expression (DGE) from Histone modifications (HM) signal is crucial to understand how HM controls cell functional heterogeneity through influencing differential gene regulation. Most existing prediction methods use fixed-length bins to represent HM signals and transmit these bins into a single machine learning model to predict differential expression genes of single cell type or cell type pair. However, the inappropriate bin length may cause the splitting of the important HM segment and lead to information loss. Furthermore, the bias of single learning model may limit the prediction accuracy. Considering these problems, in this paper, we proposes an Ensemble deep neural networks framework for predicting Differential Gene Expression (EnDGE). EnDGE employs different feature extractors on input HM signal data with different bin lengths and fuses the feature vectors for DGE prediction. Ensemble multiple learning models with different HM signal cutting strategies helps to keep the integrity and consistency of genetic information in each signal segment, and offset the bias of individual models. Besides the popular feature extractors, we also propose a new Residual Network based model with higher prediction accuracy to increase the diversity of feature extractors. Experiments on the real datasets from the Roadmap Epigenome Project (REMC) show that for all cell type pairs, EnDGE significantly outperforms the state-of-the-art baselines for differential gene expression prediction.

Accurate and highly interpretable prediction of gene expression from histone modifications

Background

Histone Mark Modifications (HMs) are crucial actors in gene regulation, as they actively remodel chromatin to modulate transcriptional activity: aberrant combinatorial patterns of HMs have been connected with several diseases, including cancer. HMs are, however, reversible modifications: understanding their role in disease would allow the design of ‘epigenetic drugs’ for specific, non-invasive treatments. Standard statistical techniques were not entirely successful in extracting representative features from raw HM signals over gene locations. On the other hand, deep learning approaches allow for effective automatic feature extraction, but at the expense of model interpretation.

Results

Here, we propose ShallowChrome, a novel computational pipeline to model transcriptional regulation via HMs in both an accurate and interpretable way. We attain state-of-the-art results on the binary classification of gene transcriptional states over 56 cell-types from the REMC database, largely outperforming recent deep learning approaches. We interpret our models by extracting insightful gene-specific regulative patterns, and we analyse them for the specific case of the PAX5 gene over three differentiated blood cell lines. Finally, we compare the patterns we obtained with the characteristic emission patterns of ChromHMM, and show that ShallowChrome is able to coherently rank groups of chromatin states w.r.t. their transcriptional activity.

Conclusions

In this work we demonstrate that it is possible to model HM-modulated gene expression regulation in a highly accurate, yet interpretable way. Our feature extraction algorithm leverages on data downstream the identification of enriched regions to retrieve gene-wise, statistically significant and dynamically located features for each HM. These features are highly predictive of gene transcriptional state, and allow for accurate modeling by computationally efficient logistic regression models. These models allow a direct inspection and a rigorous interpretation, helping to formulate quantifiable hypotheses.

A deep learning model to identify gene expression level using cobinding transcription factor signals

Gene expression is directly controlled by transcription factors (TFs) in a complex combination manner. It remains a challenging task to systematically infer how the cooperative binding of TFs drives gene activity. Here, we quantitatively analyzed the correlation between TFs and surveyed the TF interaction networks associated with gene expression in GM12878 and K562 cell lines. We identified six TF modules associated with gene expression in each cell line. Furthermore, according to the enrichment characteristics of TFs in these TF modules around a target gene, a convolutional neural network model, called TFCNN, was constructed to identify gene expression level. Results showed that the TFCNN model achieved a good prediction performance for gene expression. The average of the area under receiver operating characteristics curve (AUC) can reach up to 0.975 and 0.976, respectively in GM12878 and K562 cell lines. By comparison, we found that the TFCNN model outperformed the prediction models based on SVM and LDA. This is due to the TFCNN model could better extract the combinatorial interaction among TFs. Further analysis indicated that the abundant binding of regulatory TFs dominates expression of target genes, while the cooperative interaction between TFs has a subtle regulatory effects. And gene expression could be regulated by different TF combinations in a nonlinear way. These results are helpful for deciphering the mechanism of TF combination regulating gene expression.

Machine learning for deciphering cell heterogeneity and gene regulation

Epigenetics studies inheritable and reversible modifications of DNA that allow cells to control gene expression throughout their development and in response to environmental conditions. In computational epigenomics, machine learning is applied to study various epigenetic mechanisms genome wide. Its aim is to expand our understanding of cell differentiation, that is their specialization, in health and disease. Thus far, most efforts focus on understanding the functional encoding of the genome and on unraveling cell-type heterogeneity. Here, we provide an overview of state-of-the-art computational methods and their underlying statistical concepts, which range from matrix factorization and regularized linear regression to deep learning methods. We further show how the rise of single-cell technology leads to new computational challenges and creates opportunities to further our understanding of epigenetic regulation.

Functional effects of variation in transcription factor binding highlight long-range gene regulation by epromoters

Identifying DNA cis-regulatory modules (CRMs) that control the expression of specific genes is crucial for deciphering the logic of transcriptional control. Natural genetic variation can point to the possible gene regulatory function of specific sequences through their allelic associations with gene expression. However, comprehensive identification of causal regulatory sequences in brute-force association testing without incorporating prior knowledge is challenging due to limited statistical power and effects of linkage disequilibrium. Sequence variants affecting transcription factor (TF) binding at CRMs have a strong potential to influence gene regulatory function, which provides a motivation for prioritizing such variants in association testing. Here, we generate an atlas of CRMs showing predicted allelic variation in TF binding affinity in human lymphoblastoid cell lines and test their association with the expression of their putative target genes inferred from Promoter Capture Hi-C and immediate linear proximity. We reveal >1300 CRM TF-binding variants associated with target gene expression, the majority of them undetected with standard association testing. A large proportion of CRMs showing associations with the expression of genes they contact in 3D localize to the promoter regions of other genes, supporting the notion of 'epromoters': dual-action CRMs with promoter and distal enhancer activity.

A comprehensive review of computational prediction of genome-wide features

There are significant correlations among different types of genetic, genomic and epigenomic features within the genome. These correlations make the in silico feature prediction possible through statistical or machine learning models. With the accumulation of a vast amount of high-throughput data, feature prediction has gained significant interest lately, and a plethora of papers have been published in the past few years. Here we provide a comprehensive review on these published works, categorized by the prediction targets, including protein binding site, enhancer, DNA methylation, chromatin structure and gene expression. We also provide discussions on some important points and possible future directions.

Bayesian mixture regression analysis for regulation of Pluripotency in ES cells

BACKGROUND

Observed levels of gene expression strongly depend on both activity of DNA binding transcription factors (TFs) and chromatin state through different histone modifications (HMs). In order to recover the functional relationship between local chromatin state, TF binding and observed levels of gene expression, regression methods have proven to be useful tools. They have been successfully applied to predict mRNA levels from genome-wide experimental data and they provide insight into context-dependent gene regulatory mechanisms. However, heterogeneity arising from gene-set specific regulatory interactions is often overlooked.

RESULTS

We show that regression models that predict gene expression by using experimentally derived ChIP-seq profiles of TFs can be significantly improved by mixture modelling. In order to find biologically relevant gene clusters, we employ a Bayesian allocation procedure which allows us to integrate additional biological information such as three-dimensional nuclear organization of chromosomes and gene function. The data integration procedure involves transforming the additional data into gene similarity values. We propose a generic similarity measure that is especially suitable for situations where the additional data are of both continuous and discrete type, and compare its performance with similar measures in the context of mixture modelling.

CONCLUSIONS

We applied the proposed method on a data from mouse embryonic stem cells (ESC). We find that including additional data results in mixture components that exhibit biologically meaningful gene clusters, and provides valuable insight into the heterogeneity of the regulatory interactions.

On the problem of confounders in modeling gene expression

Motivation

Modeling of Transcription Factor (TF) binding from both ChIP-seq and chromatin accessibility data has become prevalent in computational biology. Several models have been proposed to generate new hypotheses on transcriptional regulation. However, there is no distinct approach to derive TF binding scores from ChIP-seq and open chromatin experiments. Here, we review biases of various scoring approaches and their effects on the interpretation and reliability of predictive gene expression models.

Results

We generated predictive models for gene expression using ChIP-seq and DNase1-seq data from DEEP and ENCODE. Via randomization experiments, we identified confounders in TF gene scores derived from both ChIP-seq and DNase1-seq data. We reviewed correction approaches for both data types, which reduced the influence of identified confounders without harm to model performance. Also, our analyses highlighted further quality control measures, in addition to model performance, that may help to assure model reliability and to avoid misinterpretation in future studies.

Availability and implementation

The software used in this study is available online at https://github.com/SchulzLab/TEPIC.

Supplementary information

Supplementary data are available at Bioinformatics online.

Chromatin Signature and Transcription Factor Binding Provide a Predictive Basis for Understanding Plant Gene Expression

Chromatin accessibility and post-transcriptional histone modifications play important roles in gene expression regulation. However, little is known about the joint effect of multiple chromatin modifications on the gene expression level in plants, despite that the regulatory roles of individual histone marks such as H3K4me3 in gene expression have been well-documented. By using machine-learning methods, we systematically performed gene expression level prediction based on multiple chromatin modifications data in Arabidopsis and rice. We found that as few as four histone modifications were sufficient to yield good prediction performance, and H3K4me3 and H3K36me3 being the top two predictors with known functions related to transcriptional initiation and elongation, respectively. We demonstrated that the predictive powers differed between protein-coding and non-coding genes as well as between CpG-enriched and CpG-depleted genes. We also showed that the predictive model trained in one tissue or species could be applied to another tissue or species, suggesting shared underlying mechanisms. More interestingly, the gene expression levels of conserved orthologs are easier to predict than the species-specific genes. In addition, chromatin state of distal enhancers was moderately correlated to gene expression but was dispensable if given the chromatin features of the proximal regions of genes. We further extended the analysis to transcription factor (TF) binding data. Strikingly, the combinatorial effects of only a few TFs were roughly fit to gene expression levels in Arabidopsis. Overall, by using quantitative modeling, we provide a comprehensive and unbiased perspective on the epigenetic and TF-mediated regulation of gene expression in plants.

Prediction of Gene Expression Patterns With Generalized Linear Regression Model

Cell reprogramming has played important roles in medical science, such as tissue repair, organ reconstruction, disease treatment, new drug development, and new species breeding. Oct4, a core pluripotency factor, has especially played a key role in somatic cell reprogramming through transcriptional control and affects the expression level of genes by its combination intensity. However, the quantitative relationship between Oct4 combination intensity and target gene expression is still not clear. Therefore, firstly, a generalized linear regression method was constructed to predict gene expression values in promoter regions affected by Oct4 combination intensity. Training data, including Oct4 combination intensity and target gene expression, were from promoter regions of genes with different cell development stages. Additionally, the quantitative relationship between gene expression and Oct4 combination intensity was analyzed with the proposed model. Then, the quantitative relationship between gene expression and Oct4 combination intensity at each stage of cell development was classified into high and low levels. Experimental analysis showed that the combination height of Oct4-inhibited gene expression decremented by a temporal exponential value, whereas the combination width of Oct4-promoted gene expression incremented by a temporal logarithmic value. Experimental results showed that the proposed method can achieve goodness of fit with high confidence.

Revealing transcription factor and histone modification co-localization and dynamics across cell lines by integrating ChIP-seq and RNA-seq data

Background

Interactions among transcription factors (TFs) and histone modifications (HMs) play an important role in the precise regulation of gene expression. The context specificity of those interactions and further its dynamics in normal and disease remains largely unknown. Recent development in genomics technology enables transcription profiling by RNA-seq and protein’s binding profiling by ChIP-seq. Integrative analysis of the two types of data allows us to investigate TFs and HMs interactions both from the genome co-localization and downstream target gene expression.

Results

We propose a integrative pipeline to explore the co-localization of 55 TFs and 11 HMs and its dynamics in human GM12878 and K562 by matched ChIP-seq and RNA-seq data from ENCODE. We classify TFs and HMs into three types based on their binding enrichment around transcription start site (TSS). Then a set of statistical indexes are proposed to characterize the TF-TF and TF-HM co-localizations. We found that Rad21, SMC3, and CTCF co-localized across five cell lines. High resolution Hi-C data in GM12878 shows that they associate most of the Hi-C peak loci with a specific CTCF-motif “anchor” and supports that CTCF, SMC3, and RAD2 co-localization serves important role in 3D chromatin structure. Meanwhile, 17 TF-TF pairs are highly dynamic between GM12878 and K562. We then build SVM models to correlate high and low expression level of target genes with TF binding and HM strength. We found that H3k9ac, H3k27ac, and three TFs (ELF1, TAF1, and POL2) are predictive with the accuracy about 85~92%.

Conclusion

We propose a pipeline to analyze the co-localization of TF and HM and their dynamics across cell lines from ChIP-seq, and investigate their regulatory potency by RNA-seq. The integrative analysis of two level data reveals new insight for the cooperation of TFs and HMs and is helpful in understanding cell line specificity of TF/HM interactions.

Electronic supplementary material

The online version of this article (10.1186/s12864-018-5278-5) contains supplementary material, which is available to authorized users.

The HTPmod Shiny application enables modeling and visualization of large-scale biological data

The wave of high-throughput technologies in genomics and phenomics are enabling data to be generated on an unprecedented scale and at a reasonable cost. Exploring the large-scale data sets generated by these technologies to derive biological insights requires efficient bioinformatic tools. Here we introduce an interactive, open-source web application (HTPmod) for high-throughput biological data modeling and visualization. HTPmod is implemented with the Shiny framework by integrating the computational power and professional visualization of R and including various machine-learning approaches. We demonstrate that HTPmod can be used for modeling and visualizing large-scale, high-dimensional data sets (such as multiple omics data) under a broad context. By reinvestigating example data sets from recent studies, we find not only that HTPmod can reproduce results from the original studies in a straightforward fashion and within a reasonable time, but also that novel insights may be gained from fast reinvestigation of existing data by HTPmod.

Dijun Chen et al. present HTPmod, a Shiny web application for modeling and visualization of large-scale genomic and phenomic datasets. The authors show that HTPmod can quickly reproduce analyses of high-throughput biological datasets and produce publication-quality figures.

DeepChrome: deep-learning for predicting gene expression from histone modifications

MOTIVATION

Histone modifications are among the most important factors that control gene regulation. Computational methods that predict gene expression from histone modification signals are highly desirable for understanding their combinatorial effects in gene regulation. This knowledge can help in developing 'epigenetic drugs' for diseases like cancer. Previous studies for quantifying the relationship between histone modifications and gene expression levels either failed to capture combinatorial effects or relied on multiple methods that separate predictions and combinatorial analysis. This paper develops a unified discriminative framework using a deep convolutional neural network to classify gene expression using histone modification data as input. Our system, called DeepChrome, allows automatic extraction of complex interactions among important features. To simultaneously visualize the combinatorial interactions among histone modifications, we propose a novel optimization-based technique that generates feature pattern maps from the learnt deep model. This provides an intuitive description of underlying epigenetic mechanisms that regulate genes.

RESULTS

We show that DeepChrome outperforms state-of-the-art models like Support Vector Machines and Random Forests for gene expression classification task on 56 different cell-types from REMC database. The output of our visualization technique not only validates the previous observations but also allows novel insights about combinatorial interactions among histone modification marks, some of which have recently been observed by experimental studies.

AVAILABILITY AND IMPLEMENTATION

Codes and results are available at www.deepchrome.org

CONTACT

yanjun@virginia.edu

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

RegulatorTrail: a web service for the identification of key transcriptional regulators

Transcriptional regulators such as transcription factors and chromatin modifiers play a central role in most biological processes. Alterations in their activities have been observed in many diseases, e.g. cancer. Hence, it is of utmost importance to evaluate and assess the effects of transcriptional regulators on natural and pathogenic processes. Here, we present RegulatorTrail, a web service that provides rich functionality for the identification and prioritization of key transcriptional regulators that have a strong impact on, e.g. pathological processes. RegulatorTrail offers eight methods that use regulator binding information in combination with transcriptomic or epigenomic data to infer the most influential regulators. Our web service not only provides an intuitive web interface, but also a well-documented RESTful API that allows for a straightforward integration into third-party workflows. The presented case studies highlight the capabilities of our web service and demonstrate its potential for the identification of influential regulators: we successfully identified regulators that might explain the increased malignancy in metastatic melanoma compared to primary tumors, as well as important regulators in macrophages. RegulatorTrail is freely accessible at: https://regulatortrail.bioinf.uni-sb.de/.

Combining transcription factor binding affinities with open-chromatin data for accurate gene expression prediction

The binding and contribution of transcription factors (TF) to cell specific gene expression is often deduced from open-chromatin measurements to avoid costly TF ChIP-seq assays. Thus, it is important to develop computational methods for accurate TF binding prediction in open-chromatin regions (OCRs). Here, we report a novel segmentation-based method, TEPIC, to predict TF binding by combining sets of OCRs with position weight matrices. TEPIC can be applied to various open-chromatin data, e.g. DNaseI-seq and NOMe-seq. Additionally, Histone-Marks (HMs) can be used to identify candidate TF binding sites. TEPIC computes TF affinities and uses open-chromatin/HM signal intensity as quantitative measures of TF binding strength. Using machine learning, we find low affinity binding sites to improve our ability to explain gene expression variability compared to the standard presence/absence classification of binding sites. Further, we show that both footprints and peaks capture essential TF binding events and lead to a good prediction performance. In our application, gene-based scores computed by TEPIC with one open-chromatin assay nearly reach the quality of several TF ChIP-seq data sets. Finally, these scores correctly predict known transcriptional regulators as illustrated by the application to novel DNaseI-seq and NOMe-seq data for primary human hepatocytes and CD4+ T-cells, respectively.

Differential peak calling of ChIP-seq signals with replicates with THOR

The study of changes in protein-DNA interactions measured by ChIP-seq on dynamic systems, such as cell differentiation, response to treatments or the comparison of healthy and diseased individuals, is still an open challenge. There are few computational methods comparing changes in ChIP-seq signals with replicates. Moreover, none of these previous approaches addresses ChIP-seq specific experimental artefacts arising from studies with biological replicates. We propose THOR, a Hidden Markov Model based approach, to detect differential peaks between pairs of biological conditions with replicates. THOR provides all pre- and post-processing steps required in ChIP-seq analyses. Moreover, we propose a novel normalization approach based on housekeeping genes to deal with cases where replicates have distinct signal-to-noise ratios. To evaluate differential peak calling methods, we delineate a methodology using both biological and simulated data. This includes an evaluation procedure that associates differential peaks with changes in gene expression as well as histone modifications close to these peaks. We evaluate THOR and seven competing methods on data sets with distinct characteristics from in vitro studies with technical replicates to clinical studies of cancer patients. Our evaluation analysis comprises of 13 comparisons between pairs of biological conditions. We show that THOR performs best in all scenarios.

Human Transcriptome and Chromatin Modifications: An ENCODE Perspective

A decade-long project, led by several international research groups, called the Encyclopedia of DNA Elements (ENCODE), recently released an unprecedented amount of data. The ambitious project covers transcriptome, cistrome, epigenome, and interactome data from more than 1,600 sets of experiments in human. To make use of this valuable resource, it is important to understand the information it represents and the techniques that were used to generate these data. In this review, we introduce the data that ENCODE generated, summarize the observations from the data analysis, and revisit a computational approach that ENCODE used to predict gene expression, with a focus on the human transcriptome and its association with chromatin modifications.

Taking into account nucleosomes for predicting gene expression

The eukaryotic genome is organized in a chain of nucleosomes that consist of 145-147 bp of DNA wrapped around a histone octamer protein core. Binding of transcription factors (TF) to nucleosomal DNA is frequently impeded, which makes it a challenging task to calculate TF occupancy at a given regulatory genomic site for predicting gene expression. Here, we review methods to calculate TF binding to DNA in the presence of nucleosomes. The main theoretical problems are (i) the computation speed that is becoming a bottleneck when partial unwrapping of DNA from the nucleosome is considered, (ii) the perturbation of the binding equilibrium by the activity of ATP-dependent chromatin remodelers, which translocate nucleosomes along the DNA, and (iii) the model parameterization from high-throughput sequencing data and fluorescence microscopy experiments in living cells. We discuss strategies that address these issues to efficiently compute transcription factor binding in chromatin.

Inferring epigenetic and transcriptional regulation during blood cell development with a mixture of sparse linear models

MOTIVATION

Blood cell development is thought to be controlled by a circuit of transcription factors (TFs) and chromatin modifications that determine the cell fate through activating cell type-specific expression programs. To shed light on the interplay between histone marks and TFs during blood cell development, we model gene expression from regulatory signals by means of combinations of sparse linear regression models.

RESULTS

The mixture of sparse linear regression models was able to improve the gene expression prediction in relation to the use of a single linear model. Moreover, it performed an efficient selection of regulatory signals even when analyzing all TFs with known motifs (>600). The method identified interesting roles for histone modifications and a selection of TFs related to blood development and chromatin remodelling.

AVAILABILITY

The method and datasets are available from http://www.cin.ufpe.br/~igcf/SparseMix.

CONTACT

igcf@cin.ufpe.br

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Modeling gene expression using chromatin features in various cellular contexts

BACKGROUND

Previous work has demonstrated that chromatin feature levels correlate with gene expression. The ENCODE project enables us to further explore this relationship using an unprecedented volume of data. Expression levels from more than 100,000 promoters were measured using a variety of high-throughput techniques applied to RNA extracted by different protocols from different cellular compartments of several human cell lines. ENCODE also generated the genome-wide mapping of eleven histone marks, one histone variant, and DNase I hypersensitivity sites in seven cell lines.

RESULTS

We built a novel quantitative model to study the relationship between chromatin features and expression levels. Our study not only confirms that the general relationships found in previous studies hold across various cell lines, but also makes new suggestions about the relationship between chromatin features and gene expression levels. We found that expression status and expression levels can be predicted by different groups of chromatin features, both with high accuracy. We also found that expression levels measured by CAGE are better predicted than by RNA-PET or RNA-Seq, and different categories of chromatin features are the most predictive of expression for different RNA measurement methods. Additionally, PolyA+ RNA is overall more predictable than PolyA- RNA among different cell compartments, and PolyA+ cytosolic RNA measured with RNA-Seq is more predictable than PolyA+ nuclear RNA, while the opposite is true for PolyA- RNA.

CONCLUSIONS

Our study provides new insights into transcriptional regulation by analyzing chromatin features in different cellular contexts.