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

Massively parallel reporter assay confirms regulatory potential of hQTLs and reveals important variants in lupus and other autoimmune diseases

We designed a massively parallel reporter assay (MPRA) in an Epstein-Barr virus transformed B cell line to directly characterize the potential for histone post-translational modifications, i.e., histone quantitative trait loci (hQTLs), expression QTLs (eQTLs), and variants on systemic lupus erythematosus (SLE) and autoimmune (AI) disease risk haplotypes to modulate regulatory activity in an allele-dependent manner. Our study demonstrates that hQTLs, as a group, are more likely to modulate regulatory activity in an MPRA compared with other variant classes tested, including a set of eQTLs previously shown to interact with hQTLs and tested AI risk variants. In addition, we nominate 17 variants (including 11 previously unreported) as putative causal variants for SLE and another 14 for various other AI diseases, prioritizing these variants for future functional studies in primary and immortalized B cells. Thus, we uncover important insights into the mechanistic relationships among genotype, epigenetics, and gene expression in SLE and AI disease phenotypes.

PIBF1 regulates multiple gene expression via impeding long-range chromatin interaction to drive the malignant transformation of HPV16 integration epithelial cells

INTRODUCTION

Human papillomavirus (HPV) integration can induce gene expression dysregulation by destroying higher-order chromatin structure in cervical cancer.

OBJECTIVES

We established a 13q22 site-specific HPV16 gene knock-in cell model to interrogate the changes in chromatin structure at the initial stages of host cell malignant transformation.

METHODS

We designed a CRISPR-Cas9 system with sgRNA targeting 13q22 site and constructed the HPV16 gene donor. Cells were cotransfected, screened, and fluorescence sorted. The whole genome sequencing (WGS) was used to confirm the precise HPV16 gene integration site. Western blot and qRT-PCR were used to measure gene expression. In vitro and in vivo analysis were performed to estimate the tumorigenic potential of the HPV16 knock-in cell model. Combined Hi-C, chromatin immunoprecipitation and RNA sequencing analyses revealed correlations between chromatin structure and gene expression. We performed a coimmunoprecipitation assay with anti-PIBF1 antibody to identify endogenous interacting proteins. In vivo analysis was used to determine the role of PIBF1 in the tumor growth of cervical cancer cells.

RESULTS

We successfully established a 13q22 site-specific HPV16 gene knock-in cell model. We found that HPV integration promoted cell proliferation, invasion and stratified growth in vitro, and monoclonal proliferation in vivo. HPV integration divided the affected topologically associated domain (TAD) into two smaller domains, and the progesterone-induced blocking factor 1 (PIBF1) gene near the integration site was upregulated, although PIBF1 was not enriched at the domain boundary by CUT-Tag signal analysis. Moreover, PIBF1 was found to interact with the cohesin complex off chromatin to reduce contact domain formation by disrupting the cohesin ring-shaped structure, causing dysregulation of tumorigenesis-related genes. Xenograft experiments determined the role of PIBF1 in the proliferation in cervical cancer cells.

CONCLUSION

We highlight that PIBF1, a potential chromatin structure regulatory protein, is activated by HPV integration, which provides new insights into HPV integration-driven cervical carcinogenesis.

Massively Parallel Reporter Assay Confirms Regulatory Potential of hQTLs and Reveals Important Variants in Lupus and Other Autoimmune Diseases

Objective

To systematically characterize the potential for histone post-translational modifications, i.e., histone quantitative trait loci (hQTLs), expression QTLs (eQTLs), and variants on systemic lupus erythematosus (SLE) and autoimmune (AI) disease risk haplotypes to modulate gene expression in an allele dependent manner.

Methods

We designed a massively parallel reporter assay (MPRA) containing ~32K variants and transfected it into an Epstein-Barr virus transformed B cell line generated from an SLE case.

Results

Our study expands our understanding of hQTLs, illustrating that epigenetic QTLs are more likely to contribute to functional mechanisms than eQTLs and other variant types, and a large proportion of hQTLs overlap transcription start sites (TSS) of noncoding RNAs. In addition, we nominate 17 variants (including 11 novel) as putative causal variants for SLE and another 14 for various other AI diseases, prioritizing these variants for future functional studies primary and immortalized B cells.

Conclusion

We uncover important insights into the mechanistic relationships between genotype, epigenetics, gene expression, and SLE and AI disease phenotypes.

Integrated multiomics analysis of chromosome 19 miRNA cluster in bladder cancer

With 46 microRNAs (miRNAs) embedded tandemly over a distance of ~100 kb, chromosome 19 microRNA cluster (C19MC) is the largest miRNA cluster in the human genome. The C19MC is transcribed from a long noncoding genomic region and is usually expressed simultaneously at a higher level. Hence, we performed an integrative multiomics data analysis to examine C19MC regulation, expression patterns, and their impact on bladder cancer (BCa). We found that 43 members of C19MC were highly expressed in BCa. However, its co-localization with recurrent copy number variation (CNV) gain was not statistically significant to implicate its upregulation. It has been reported that C19MC expression is regulated by a well-established CpG island situated 17.6 kb upstream of the transcription start site, but we found that CpG probes at this island were hypomethylated, which was not statistically significant in the BCa cohort. In addition, the promoter region of C19MC is strongly regulated by a group of seven transcription factors (NR2F6, SREBF1, TBP, GATA3, GABPB1, ETV4, and ZNF444) and five chromatin modifiers (SMC3, KDMA1, EZH2, RAD21, and CHD7). Interestingly, these 12 genes were found to be overexpressed in BCa patients. Further, C19MC targeted 42 tumor suppressor (TS) genes that were downregulated, of which 15 were significantly correlated with patient survival. Our findings suggest that transcription factors and chromatin modifiers at the promoter region may regulate C19MC overexpression. The upregulated C19MC members, transcription regulators, and TS genes can be further exploited as potential diagnostic and prognostic indicators as well as for therapeutic management of BCa.

Quantitative trait locus (xQTL) approaches identify risk genes and drug targets from human non-coding genomes

Advances and reduction of costs in various sequencing technologies allow for a closer look at variations present in the non-coding regions of the human genome. Correlating non-coding variants with large-scale multi-omic data holds the promise not only of a better understanding of likely causal connections between non-coding DNA and expression of traits but also identifying potential disease-modifying medicines. Genome-phenome association studies have created large datasets of DNA variants that are associated with multiple traits or diseases, such as Alzheimer's disease; yet, the functional consequences of variants, in particular of non-coding variants, remain largely unknown. Recent advances in functional genomics and computational approaches have led to the identification of potential roles of DNA variants, such as various quantitative trait locus (xQTL) techniques. Multi-omic assays and analytic approaches toward xQTL have identified links between genetic loci and human transcriptomic, epigenomic, proteomic and metabolomic data. In this review, we first discuss the recent development of xQTL from multi-omic findings. We then highlight multimodal analysis of xQTL and genetic data for identification of risk genes and drug targets using Alzheimer's disease as an example. We finally discuss challenges and future research directions (e.g. artificial intelligence) for annotation of non-coding variants in complex diseases.

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.

The functional insight into the genetics of cardiovascular disease: results from the post-GWAS study

Cardiovascular diseases (CVDs), the leading cause of death worldwide, generally refer to a range of pathological conditions with the involvement of the heart and the blood vessels. A sizable fraction of the susceptibility loci is known, but the underlying mechanisms have been established only for a small proportion. Therefore, there is an increasing need to explore the functional relevance of trait-associated variants and, moreover, to search for novel risk genetic variation. We have reported the bioinformatic approach allowing effective identification of functional non-coding variants by integrated analysis of genome-wide data. Here, the analysis of 1361 previously identified regulatory SNPs (rSNPs) was performed to provide new insights into cardiovascular risk. We found 773,471 coding co-segregating markers for input rSNPs using the 1000 Genomes Project. The intersection of GWAS-derived SNPs with a relevance to cardiovascular traits with these markers was analyzed within a window of 10 Kbp. The effects on the transcription factor (TF) binding sites were explored by DeFine models. Functional pathway enrichment and protein– protein interaction (PPI) network analyses were performed on the targets and the extended genes by STRING and DAVID. Eighteen rSNPs were functionally linked to cardiovascular risk. A significant impact on binding sites of thirteen TFs including those involved in blood cells formation, hematopoiesis, macrophage function, inflammation, and vasoconstriction was found in K562 cells. 21 rSNP gene targets and 5 partners predicted by PPI were enriched for spliceosome and endocytosis KEGG pathways, endosome sorting complex and mRNA splicing REACTOME pathways. Related Gene Ontology terms included mRNA splicing and processing, endosome transport and protein catabolic processes. Together, the findings provide further insight into the biological basis of CVDs and highlight the importance of the precise regulation of splicing and alternative splicing.

Computational modeling of chromatin accessibility identified important epigenomic regulators

Chromatin accessibility is essential for transcriptional activation of genomic regions. It is well established that transcription factors (TFs) and histone modifications (HMs) play critical roles in chromatin accessibility regulation. However, there is a lack of studies that quantify these relationships. Here we constructed a two-layer model to predict chromatin accessibility by integrating DNA sequence, TF binding, and HM signals. By applying the model to two human cell lines (GM12878 and HepG2), we found that DNA sequences had limited power for accessibility prediction, while both TF binding and HM signals predicted chromatin accessibility with high accuracy. According to the HM model, HM features determined chromatin accessibility in a cell line shared manner, with the prediction power attributing to five core HM types. Results from the TF model indicated that chromatin accessibility was determined by a subset of informative TFs including both cell line-specific and generic TFs. The combined model of both TF and HM signals did not further improve the prediction accuracy, indicating that they provide redundant information in terms of chromatin accessibility prediction. The TFs and HM models can also distinguish the chromatin accessibility of proximal versus distal transcription start sites with high accuracy.

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.

Superenhancer-transcription factor regulatory network in malignant tumors

Objective

This study aims to identify superenhancer (SE)-transcriptional factor (TF) regulatory network related to eight common malignant tumors based on ChIP-seq data modified by histone H3K27ac in the enhancer region of the SRA database.

Methods

H3K27ac ChIP-seq data of eight common malignant tumor samples were downloaded from the SRA database and subjected to comparison with the human reference genome hg19. TFs regulated by SEs were screened with HOMER software. Core regulatory circuitry (CRC) in malignant tumor samples was defined through CRCmapper software and validated by RNA-seq data in TCGA. The findings were substantiated in bladder cancer cell experiments.

Results

Different malignant tumors could be distinguished through the H3K27ac signal. After SE identification in eight common malignant tumor samples, 35 SE-regulated genes were defined as malignant tumor-specific. SE-regulated specific TFs effectively distinguished the types of malignant tumors. Finally, we obtained 60 CRC TFs, and SMAD3 exhibited a strong H3K27ac signal in eight common malignant tumor samples. In vitro experimental data verified the presence of a SE-TF regulatory network in bladder cancer, and SE-TF regulatory network enhanced the malignant phenotype of bladder cancer cells.

Conclusion

The SE-TF regulatory network with SMAD3 as the core TF may participate in the carcinogenesis of malignant tumors.

The transcription factor code: a beacon for histone methyltransferase docking

Histone methylation is required for the establishment and maintenance of gene expression patterns that determine cellular identity, and its perturbation often leads to aberrant development and disease. Recruitment of histone methyltransferases (HMTs) to gene regulatory elements (GREs) of developmental genes is important for the correct activation and silencing of these genes, but the drivers of this recruitment are largely unknown. Here we propose that lineage-instructive transcription factors (Lin-TFs) act as general recruiters of HMT complexes to cell type-specific GREs through protein-protein interactions. We also postulate that the specificity of these interactions is dictated by Lin-TF post-translational modifications (PTMs), which act as a 'transcription factor code' that can determine the directionality of cell fate decisions during differentiation and development.

3Scover: Identifying Safeguard TF from Cell Type-TF Specificity Network by an Extended Minimum Set Cover Model

Transcription factors (TFs) define cellular identity either by activating target cell program or by silencing donor program as demonstrated by intensive cell reprogramming studies. Here, we propose an extended minimum set cover model with stable selection (3Scover) to systematically identify silencing TFs, named safeguard TFs, from omics data. First, a cell type-TF specificity network is constructed to systematically link cell types with their specifically expressed TFs. Then we search the minimum TF set to cover this network with “many but one specificity” characteristic and integrate many subsampling models for a stable solution. 3Scover identified 30 safeguard TFs in human and mouse. These safeguard TFs are significantly enriched in the experimentally discovered reprogramming panel with their protein-protein interactors. In addition, they tend to interact closely with chromatin regulators, negatively regulate transcription, and function earlier in development. Collectively, 3Scover allows us to probe master TFs and combinatorial regulation in controlling cell identity.

•

Cell type-TF specificity networks reveal the relationships among TF and cell identity

•

3SCover extracts safeguard TFs by “many but one specificity” and parsimony principle

•

Safeguard TFs are enriched in reprogramming panel and interact closely with CR

•

Safeguard TFs are conserved in mouse and human

Cell-specific exon methylation and CTCF binding in neurons regulate calcium ion channel splicing and function

Cell-specific alternative splicing modulates myriad cell functions and is disrupted in disease. The mechanisms governing alternative splicing are known for relatively few genes and typically focus on RNA splicing factors. In sensory neurons, cell-specific alternative splicing of the presynaptic Ca_V channel Cacna1b gene modulates opioid sensitivity. How this splicing is regulated is unknown. We find that cell and exon-specific DNA hypomethylation permits CTCF binding, the master regulator of mammalian chromatin structure, which, in turn, controls splicing in a DRG-derived cell line. In vivo, hypomethylation of an alternative exon specifically in nociceptors, likely permits CTCF binding and expression of Ca_V2.2 channel isoforms with increased opioid sensitivity in mice. Following nerve injury, exon methylation is increased, and splicing is disrupted. Our studies define the molecular mechanisms of cell-specific alternative splicing of a functionally validated exon in normal and disease states – and reveal a potential target for the treatment of chronic pain.

The spatial binding model of the pioneer factor Oct4 with its target genes during cell reprogramming

Understanding the target regulation between pioneer factor and its binding genes is crucial for improving the efficiency of TF-mediated reprogramming. Oct4 as the only one factor that cannot be substituted by other POU members, it is urgent need to develop a quantitative model for describing the spatial binding pattern with its target genes. The dynamic profiles of pioneer factor Oct4-binding showed that the major wave occurs at the intermediate stage of cell reprogramming (from day 7 to day 15), and the promoter is the preferred targeting regions. The Oct4-binding distributions perform significant chromosome bias. The overall enrichment on chromosome 1–11 is higher than that on the others. The dramatic event of TF-mediated reprogramming is mainly concentrated on autosomes. We also found that the spatial binding ability of Oct4 binding can be represented quantitatively by using three parameters of peaks (height, width and distance). The dynamic changes of Oct4-binding demonstrated that the width play more important roles in regulating expression of target genes. At last, a multivariate linear regression was introduced to establish the spatial binding model of the Oct4-binding. The evaluation results confirmed that the height and width is positively correlated with the gene expression. And the additive interaction terms of height and width can better optimize the model performance than the multiplicative terms. The best average coefficients of determination of improved model achieved to 81.38%. Our study will provide new insights into the cooperative regulation of spatial binding pattern of pioneer factors in cell reprogramming.