Inferring active regulatory networks from gene expression data using a combination of prior knowledge and enrichment analysis

Inferring upstream regulatory genes of FOXP3 in human regulatory T cells from time-series transcriptomic data

The discovery of upstream regulatory genes of a gene of interest still remains challenging. Here we applied a scalable computational method to unbiasedly predict candidate regulatory genes of critical transcription factors by searching the whole genome. We illustrated our approach with a case study on the master regulator FOXP3 of human primary regulatory T cells (Tregs). While target genes of FOXP3 have been identified, its upstream regulatory machinery still remains elusive. Our methodology selected five top-ranked candidates that were tested via proof-of-concept experiments. Following knockdown, three out of five candidates showed significant effects on the mRNA expression of FOXP3 across multiple donors. This provides insights into the regulatory mechanisms modulating FOXP3 transcriptional expression in Tregs. Overall, at the genome level this represents a high level of accuracy in predicting upstream regulatory genes of key genes of interest.

Exploration of Target Spaces in the Human Genome for Protein and Peptide Drugs

After decades of development, protein and peptide drugs have now grown into a major drug class in the marketplace. Target identification and validation are crucial for the discovery of protein and peptide drugs, and bioinformatics prediction of targets based on the characteristics of known target proteins will help improve the efficiency and success rate of target selection. However, owing to the developmental history in the pharmaceutical industry, previous systematic exploration of the target spaces has mainly focused on traditional small-molecule drugs, while studies related to protein and peptide drugs are lacking. Here, we systematically explore the target spaces in the human genome specifically for protein and peptide drugs. Compared with other proteins, both successful protein and peptide drug targets have many special characteristics, and are also significantly different from those of small-molecule drugs in many aspects. Based on these features, we develop separate effective genome-wide target prediction models for protein and peptide drugs. Finally, a user-friendly web server, Predictor Of Protein and PeptIde drugs' therapeutic Targets (POPPIT) (http://poppit.ncpsb.org.cn/), is established, which provides not only target prediction specifically for protein and peptide drugs but also abundant annotations for predicted targets.

Generation of Realistic Gene Regulatory Networks by Enriching for Feed-Forward Loops

The regulatory relationships between genes and proteins in a cell form a gene regulatory network (GRN) that controls the cellular response to changes in the environment. A number of inference methods to reverse engineer the original GRN from large-scale expression data have recently been developed. However, the absence of ground-truth GRNs when evaluating the performance makes realistic simulations of GRNs necessary. One aspect of this is that local network motif analysis of real GRNs indicates that the feed-forward loop (FFL) is significantly enriched. To simulate this properly, we developed a novel motif-based preferential attachment algorithm, FFLatt, which outperformed the popular GeneNetWeaver network generation tool in reproducing the FFL motif occurrence observed in literature-based biological GRNs. It also preserves important topological properties such as scale-free topology, sparsity, and average in/out-degree per node. We conclude that FFLatt is well-suited as a network generation module for a benchmarking framework with the aim to provide fair and robust performance evaluation of GRN inference methods.

Interplay of transcriptional signaling by progesterone, cyclic AMP, and inflammation in myometrial cells: implications for the control of human parturition

Parturition involves cellular signaling changes driven by the complex interplay between progesterone (P4), inflammation, and the cyclic adenosine monophosphate (cAMP) pathway. To characterize this interplay, we performed comprehensive transcriptomic studies utilizing eight treatment combinations on myometrial cell lines and tissue samples from pregnant women. We performed genome-wide RNA-sequencing on the hTERT-HM${}^{A/B}$ cell line treated with all combinations of P4, forskolin (FSK) (induces cAMP), and interleukin-1$\beta$ (IL-1$\beta$). We then performed gene set enrichment and regulatory network analyses to identify pathways commonly, differentially, or synergistically regulated by these treatments. Finally, we used tissue similarity index (TSI) to characterize the correspondence between cell lines and tissue phenotypes. We observed that in addition to their individual anti-inflammatory effects, P4 and cAMP synergistically blocked specific inflammatory pathways/regulators including STAT3/6, CEBPA/B, and OCT1/7, but not NF$\kappa$B. TSI analysis indicated that FSK + P4- and IL-1$\beta$-treated cells exhibit transcriptional signatures highly similar to non-laboring and laboring term myometrium, respectively. Our results identify potential therapeutic targets to prevent preterm birth and show that the hTERT-HM${}^{A/B}$ cell line provides an accurate transcriptional model for term myometrial tissue.

A Year in the Life of the EU-CardioRNA COST Action: CA17129 Catalysing Transcriptomics Research in Cardiovascular Disease

The EU-CardioRNA Cooperation in Science and Technology (COST) Action is a European-wide consortium established in 2018 with 31 European country members and four associate member countries to build bridges between translational researchers from academia and industry who conduct research on non-coding RNAs, cardiovascular diseases and similar research areas. EU-CardioRNA comprises four core working groups (WG1-4). In the first year since its launch, EU-CardioRNA met biannually to exchange and discuss recent findings in related fields of scientific research, with scientific sessions broadly divided up according to WG. These meetings are also an opportunity to establish interdisciplinary discussion groups, brainstorm ideas and make plans to apply for joint research grants and conduct other scientific activities, including knowledge transfer. Following its launch in Brussels in 2018, three WG meetings have taken place. The first of these in Lisbon, Portugal, the second in Istanbul, Turkey, and the most recent in Maastricht, The Netherlands. Each meeting includes a scientific session from each WG. This meeting report briefly describes the highlights and key take-home messages from each WG session in this first successful year of the EU-CardioRNA COST Action.

A probabilistic graphical model for system-wide analysis of gene regulatory networks

Motivation

The inference of gene regulatory networks (GRNs) from DNA microarray measurements forms a core element of systems biology-based phenotyping. In the recent past, numerous computational methodologies have been formalized to enable the deduction of reliable and testable predictions in today’s biology. However, little focus has been aimed at quantifying how well existing state-of-the-art GRNs correspond to measured gene-expression profiles.

Results

Here, we present a computational framework that combines the formulation of probabilistic graphical modeling, standard statistical estimation, and integration of high-throughput biological data to explore the global behavior of biological systems and the global consistency between experimentally verified GRNs and corresponding large microarray compendium data. The model is represented as a probabilistic bipartite graph, which can handle highly complex network systems and accommodates partial measurements of diverse biological entities, e.g. messengerRNAs, proteins, metabolites and various stimulators participating in regulatory networks. This method was tested on microarray expression data from the M3D database, corresponding to sub-networks on one of the best researched model organisms, Escherichia coli. Results show a surprisingly high correlation between the observed states and the inferred system’s behavior under various experimental conditions.

Availability and implementation

Processed data and software implementation using Matlab are freely available at https://github.com/kotiang54/PgmGRNs. Full dataset available from the M3D database.

Transcriptome reprogramming and myeloid skewing in haematopoietic stem and progenitor cells in systemic lupus erythematosus

OBJECTIVES

Haematopoietic stem and progenitor cells (HSPCs) are multipotent cells giving rise to both myeloid and lymphoid cell lineages. We reasoned that the aberrancies of immune cells in systemic lupus erythematosus (SLE) could be traced back to HSPCs.

METHODS

A global gene expression map of bone marrow (BM)-derived HSPCs was completed by RNA sequencing followed by pathway and enrichment analysis. The cell cycle status and apoptosis status of HSPCs were assessed by flow cytometry, while DNA damage was assessed via immunofluorescence.

RESULTS

Transcriptomic analysis of Lin-Sca-1+c-Kit+ haematopoietic progenitors from diseased lupus mice demonstrated a strong myeloid signature with expanded frequencies of common myeloid progenitors (CMPs)-but not of common lymphoid progenitors-reminiscent of a 'trained immunity' signature. CMP profiling revealed an intense transcriptome reprogramming with suppression of granulocytic regulators indicative of a differentiation arrest with downregulation trend of major regulators such as Cebpe, Cebpd and Csf3r, and disturbed myelopoiesis. Despite the differentiation arrest, frequencies of BM neutrophils were markedly increased in diseased mice, suggesting an alternative granulopoiesis pathway. In patients with SLE with severe disease, haematopoietic progenitor cells (CD34+) demonstrated enhanced proliferation, cell differentiation and transcriptional activation of cytokines and chemokines that drive differentiation towards myelopoiesis, thus mirroring the murine data.

CONCLUSIONS

Aberrancies of immune cells in SLE can be traced back to the BM HSPCs. Priming of HSPCs and aberrant regulation of myelopoiesis may contribute to inflammation and risk of flare.

TRIAL REGISTRATION NUMBER

4948/19-07-2016.

BINDER: computationally inferring a gene regulatory network for Mycobacterium abscessus

BACKGROUND

Although many of the genic features in Mycobacterium abscessus have been fully validated, a comprehensive understanding of the regulatory elements remains lacking. Moreover, there is little understanding of how the organism regulates its transcriptomic profile, enabling cells to survive in hostile environments. Here, to computationally infer the gene regulatory network for Mycobacterium abscessus we propose a novel statistical computational modelling approach: BayesIan gene regulatory Networks inferreD via gene coExpression and compaRative genomics (BINDER). In tandem with derived experimental coexpression data, the property of genomic conservation is exploited to probabilistically infer a gene regulatory network in Mycobacterium abscessus.Inference on regulatory interactions is conducted by combining 'primary' and 'auxiliary' data strata. The data forming the primary and auxiliary strata are derived from RNA-seq experiments and sequence information in the primary organism Mycobacterium abscessus as well as ChIP-seq data extracted from a related proxy organism Mycobacterium tuberculosis. The primary and auxiliary data are combined in a hierarchical Bayesian framework, informing the apposite bivariate likelihood function and prior distributions respectively. The inferred relationships provide insight to regulon groupings in Mycobacterium abscessus.

RESULTS

We implement BINDER on data relating to a collection of 167,280 regulator-target pairs resulting in the identification of 54 regulator-target pairs, across 5 transcription factors, for which there is strong probability of regulatory interaction.

CONCLUSIONS

The inferred regulatory interactions provide insight to, and a valuable resource for further studies of, transcriptional control in Mycobacterium abscessus, and in the family of Mycobacteriaceae more generally. Further, the developed BINDER framework has broad applicability, useable in settings where computational inference of a gene regulatory network requires integration of data sources derived from both the primary organism of interest and from related proxy organisms.

Promyelocytic leukemia protein (PML) controls breast cancer cell proliferation by modulating Forkhead transcription factors

The multitasking promyelocytic leukemia (PML) protein was originally recognized as a tumor‐suppressive factor, but more recent evidence has implicated PML in tumor cell prosurvival actions and poor patient prognosis in specific cancer settings. Here, we report that inducible PMLIV expression inhibits cell proliferation as well as self‐renewal and impairs cell cycle progression of breast cancer cell lines in a reversible manner. Transcriptomic profiling identified a large number of PML‐deregulated genes associated with various cell processes. Among them, cell cycle‐ and division‐related genes and their cognitive regulators are highly ranked. In this study, we focused on previously unknown PML targets, namely the Forkhead transcription factors. PML suppresses the Forkhead box subclass M1 (FOXM1) transcription factor at both the RNA and protein levels, along with many of its gene targets. We show that FOXM1 interacts with PMLIV primarily via its DNA‐binding domain and dynamically colocalizes in PML nuclear bodies. In parallel, PML modulates the activity of Forkhead box O3 (FOXO3), a factor opposing certain FOXM1 activities, to promote cell survival and stress resistance. Thus, PMLIV affects the balance of FOXO3 and FOXM1 transcriptional programs by acting on discrete gene subsets to favor both growth inhibition and survival. Interestingly, PMLIV‐specific knockdown mimicked ectopic expression vis‐à‐vis loss of proliferative ability and self‐renewal, but also led to loss of survival ability as shown by increased apoptosis. We propose that divergent or similar effects on cell physiology may be elicited by high or low PMLIV levels dictated by other concurrent genetic or epigenetic cancer cell states that may additionally account for its disparate effects in various cancer types.

An integrative transcriptome analysis framework for drug efficacy and similarity reveals drug-specific signatures of anti-TNF treatment in a mouse model of inflammatory polyarthritis

Anti-TNF agents have been in the first line of treatment of various inflammatory diseases such as Rheumatoid Arthritis and Crohn’s Disease, with a number of different biologics being currently in use. A detailed analysis of their effect at transcriptome level has nevertheless been lacking. We herein present a concise analysis of an extended transcriptomics profiling of four different anti-TNF biologics upon treatment of the established hTNFTg (Tg197) mouse model of spontaneous inflammatory polyarthritis. We implement a series of computational analyses that include clustering of differentially expressed genes, functional analysis and random forest classification. Taking advantage of our detailed sample structure, we devise metrics of treatment efficiency that take into account changes in gene expression compared to both the healthy and the diseased state. Our results suggest considerable variability in the capacity of different biologics to modulate gene expression that can be attributed to treatment-specific functional pathways and differential preferences to restore over- or under-expressed genes. Early intervention appears to manage inflammation in a more efficient way but is accompanied by increased effects on a number of genes that are seemingly unrelated to the disease. Administration at an early stage is also lacking in capacity to restore healthy expression levels of under-expressed genes. We record quantifiable differences among anti-TNF biologics in their efficiency to modulate over-expressed genes related to immune and inflammatory pathways. More importantly, we find a subset of the tested substances to have quantitative advantages in addressing deregulation of under-expressed genes involved in pathways related to known RA comorbidities. Our study shows the potential of transcriptomic analyses to identify comprehensive and distinct treatment-specific gene signatures combining disease-related and unrelated genes and proposes a generalized framework for the assessment of drug efficacy, the search of biosimilars and the evaluation of the efficacy of TNF small molecule inhibitors.

A number of anti-TNF drugs are being used in the treatment of inflammatory autoimmune diseases, such as Rheumatoid Arthritis and Crohn’s Disease. Despite their wide use there has been, to date, no detailed analysis of their effect on the affected tissues at a transcriptome level. In this work we applied four different anti-TNF drugs on an established mouse model of inflammatory polyarthritis and collected a large number of independent biological replicates from the synovial tissue of healthy, diseased and treated animals. We then applied a series of bioinformatics analyses in order to define the sets of genes, biological pathways and functions that are affected in the diseased animals and modulated by each of the different treatments. Our dataset allowed us to focus on previously overlooked aspects of gene regulation. We found that the majority of differentially expressed genes in disease are under-expressed and that they are also associated with functions related to Rheumatoid Arthritis comorbidities such as cardiovascular disease. We were also able to define gene and pathway subsets that are not changed in the disease but are, nonetheless, altered under various treatments and to use these subsets in drug classification and assessment. Through the application of machine learning approaches we created quantitative efficiency profiles for the tested drugs, which showed some to be more efficiently addressing changes in the inflammatory pathways, while others being quantitatively superior in restoring gene expression changes associated to disease comorbidities. We thus, propose a concise computational pipeline that may be used in the assessment of drug efficacy and biosimilarity and which may form the basis of evaluation protocols for small molecule TNF inhibitors.

Myometrial Transcriptional Signatures of Human Parturition

The process of parturition involves the transformation of the quiescent myometrium (uterine smooth muscle) to the highly contractile laboring state. This is thought to be driven by changes in gene expression in myometrial cells. Despite the existence of multiple myometrial gene expression studies, the transcriptional programs that initiate labor are not known. Here, we integrated three transcriptome datasets, one novel (NCBI Gene Expression Ominibus: GSE80172) and two existing, to characterize the gene expression changes in myometrium associated with the onset of labor at term. Computational analyses including classification, singular value decomposition, pathway enrichment, and network inference were applied to individual and combined datasets. Outcomes across studies were integrated with multiple protein and pathway databases to build a myometrial parturition signaling network. A high-confidence (significant across all studies) set of 126 labor genes were identified and machine learning models exhibited high reproducibility between studies. Labor signatures included both known (interleukins, cytokines) and unknown (apoptosis, MYC, cell proliferation/differentiation) pathways while cyclic AMP signaling and muscle relaxation were associated with non-labor. These signatures accurately classified and characterized the stages of labor. The data-derived parturition signaling networks provide new genes/signaling interactions to understand phenotype-specific processes and aid in future studies of parturition.

Wnt1 silences chemokine genes in dendritic cells and induces adaptive immune resistance in lung adenocarcinoma

Lung adenocarcinoma (LUAD)-derived Wnts increase cancer cell proliferative/stemness potential, but whether they impact the immune microenvironment is unknown. Here we show that LUAD cells use paracrine Wnt1 signaling to induce immune resistance. In TCGA, Wnt1 correlates strongly with tolerogenic genes. In another LUAD cohort, Wnt1 inversely associates with T cell abundance. Altering Wnt1 expression profoundly affects growth of murine lung adenocarcinomas and this is dependent on conventional dendritic cells (cDCs) and T cells. Mechanistically, Wnt1 leads to transcriptional silencing of CC/CXC chemokines in cDCs, T cell exclusion and cross-tolerance. Wnt-target genes are up-regulated in human intratumoral cDCs and decrease upon silencing Wnt1, accompanied by enhanced T cell cytotoxicity. siWnt1-nanoparticles given as single therapy or part of combinatorial immunotherapies act at both arms of the cancer-immune ecosystem to halt tumor growth. Collectively, our studies show that Wnt1 induces immunologically cold tumors through cDCs and highlight its immunotherapeutic targeting.

E. coli gene regulatory networks are inconsistent with gene expression data

Gene regulatory networks (GRNs) and gene expression data form a core element of systems biology-based phenotyping. Changes in the expression of transcription factors are commonly believed to have a causal effect on the expression of their targets. Here we evaluated in the best researched model organism, Escherichia coli, the consistency between a GRN and a large gene expression compendium. Surprisingly, a modest correlation was observed between the expression of transcription factors and their targets and, most noteworthy, both activating and repressing interactions were associated with positive correlation. When evaluated using a sign consistency model we found the regulatory network was not more consistent with measured expression than random network models. We conclude that, at least in E. coli, one cannot expect a causal relationship between the expression of transcription and factors their targets, and that the current static GRN does not adequately explain transcriptional regulation. The implications of this are profound as they question what we consider established knowledge of the systemic biology of cells and point to methodological limitations with respect to single omics analysis, static networks and temporality.

A common molecular signature of patients with sickle cell disease revealed by microarray meta-analysis and a genome-wide association study

A chronic inflammatory state to a large extent explains sickle cell disease (SCD) pathophysiology. Nonetheless, the principal dysregulated factors affecting this major pathway and their mechanisms of action still have to be fully identified and elucidated. Integrating gene expression and genome-wide association study (GWAS) data analysis represents a novel approach to refining the identification of key mediators and functions in complex diseases. Here, we performed gene expression meta-analysis of five independent publicly available microarray datasets related to homozygous SS patients with SCD to identify a consensus SCD transcriptomic profile. The meta-analysis conducted using the MetaDE R package based on combining p values (maxP approach) identified 335 differentially expressed genes (DEGs; 224 upregulated and 111 downregulated). Functional gene set enrichment revealed the importance of several metabolic pathways, of innate immune responses, erythrocyte development, and hemostasis pathways. Advanced analyses of GWAS data generated within the framework of this study by means of the atSNP R package and SIFT tool identified 60 regulatory single-nucleotide polymorphisms (rSNPs) occurring in the promoter of 20 DEGs and a deleterious SNP, affecting CAMKK2 protein function. This novel database of candidate genes, transcription factors, and rSNPs associated with SCD provides new markers that may help to identify new therapeutic targets.

Comorbid TNF-mediated heart valve disease and chronic polyarthritis share common mesenchymal cell-mediated aetiopathogenesis

OBJECTIVES

Patients with rheumatoid arthritis and spondyloarthritisshow higher mortality rates, mainly caused by cardiac comorbidities. The TghuTNF (Tg197) arthritis model develops tumour necrosis factor (TNF)-driven and mesenchymalsynovial fibroblast (SF)-dependent polyarthritis. Here, we investigate whether this model develops, similarly to human patients, comorbid heart pathology and explore cellular and molecular mechanisms linking arthritis to cardiac comorbidities.

METHODS

Histopathological analysis and echocardiographic evaluation of cardiac function were performed in the Tg197 model. Valve interstitial cells (VICs) were targeted by mice carrying the ColVI-Cretransgene. Tg197 ColVI-Cre Tnfr1fl/fl and Tg197 ColVI-Cre Tnfr1cneo/cneo mutant mice were used to explore the role of mesenchymal TNF signalling in the development of heart valve disease. Pathogenic VICs and SFs were further analysed by comparative RNA-sequencing analysis.

RESULTS

Tg197 mice develop left-sided heart valve disease, characterised by valvular fibrosis with minimal signs of inflammation. Thickened valve areas consist almost entirely of hyperproliferative ColVI-expressing mesenchymal VICs. Development of pathology results in valve stenosis and left ventricular dysfunction, accompanied by arrhythmic episodes and, occasionally, valvular regurgitation. TNF dependency of the pathology was indicated by disease modulation following pharmacological inhibition or mesenchymal-specific genetic ablation or activation of TNF/TNFR1 signalling. Tg197-derived VICs exhibited an activated phenotype ex vivo, reminiscent of the activated pathogenic phenotype of Tg197-derived SFs. Significant functional similarities between SFs and VICs were revealed by RNA-seq analysis, demonstrating common cellular mechanisms underlying TNF-mediated arthritides and cardiac comorbidities.

CONCLUSIONS

Comorbidheart valve disease and chronic polyarthritis are efficiently modelled in the Tg197 arthritis model and share common TNF/TNFR1-mediated, mesenchymal cell-specific aetiopathogenic mechanisms.

Mesenchymal TNFR2 promotes the development of polyarthritis and comorbid heart valve stenosis

Mesenchymal TNF signaling is etiopathogenic for inflammatory diseases such as rheumatoid arthritis and spondyloarthritis (SpA). The role of Tnfr1 in arthritis has been documented; however, Tnfr2 functions are unknown. Here, we investigate the mesenchymal-specific role of Tnfr2 in the TnfΔARE mouse model of SpA in arthritis and heart valve stenosis comorbidity by cell-specific, Col6a1-cre-driven gene targeting. We find that TNF/Tnfr2 signaling in resident synovial fibroblasts (SFs) and valvular interstitial cells (VICs) is detrimental for both pathologies, pointing to common cellular mechanisms. In contrast, systemic Tnfr2 provides protective signaling, since its complete deletion leads to severe deterioration of both pathologies. SFs and VICs lacking Tnfr2 fail to acquire pathogenic activated phenotypes and display increased expression of antiinflammatory cytokines associated with decreased Akt signaling. Comparative RNA sequencing experiments showed that the majority of the deregulated pathways in TnfΔARE mesenchymal-origin SFs and VICs, including proliferation, inflammation, migration, and disease-specific genes, are regulated by Tnfr2; thus, in its absence, they are maintained in a quiescent nonpathogenic state. Our data indicate a pleiotropy of Tnfr2 functions, with mesenchymal Tnfr2 driving cell activation and arthritis/valve stenosis pathogenesis only in the presence of systemic Tnfr2, whereas nonmesenchymal Tnfr2 overcomes this function, providing protective signals and, thus, containing both pathologies.

Structure and Chromosomal Organization of Yeast Genes Regulated by Topoisomerase II

Cellular DNA topoisomerases (topo I and topo II) are highly conserved enzymes that regulate the topology of DNA during normal genome transactions, such as DNA transcription and replication. In budding yeast, topo I is dispensable whereas topo II is essential, suggesting fundamental and exclusive roles for topo II, which might include the functions of the topo IIa and topo IIb isoforms found in mammalian cells. In this review, we discuss major findings of the structure and chromosomal organization of genes regulated by topo II in budding yeast. Experimental data was derived from short (10 min) and long term (120 min) responses to topo II inactivation in top-2 ts mutants. First, we discuss how short term responses reveal a subset of yeast genes that are regulated by topo II depending on their promoter architecture. These short term responses also uncovered topo II regulation of transcription across multi-gene clusters, plausibly by common DNA topology management. Finally, we examine the effects of deactivated topo II on the elongation of RNA transcripts. Each study provides an insight into the particular chromatin structure that interacts with the activity of topo II. These findings are of notable clinical interest as numerous anti-cancer therapies interfere with topo II activity.

Genome urbanization: clusters of topologically co-regulated genes delineate functional compartments in the genome of Saccharomyces cerevisiae

The eukaryotic genome evolves under the dual constraint of maintaining coordinated gene transcription and performing effective DNA replication and cell division, the coupling of which brings about inevitable DNA topological tension. DNA supercoiling is resolved and, in some cases, even harnessed by the genome through the function of DNA topoisomerases, as has been shown in the concurrent transcriptional activation and suppression of genes upon transient deactivation of topoisomerase II (topoII). By analyzing a genome-wide transcription run-on experiment upon thermal inactivation of topoII in Saccharomyces cerevisiae we were able to define 116 gene clusters of consistent response (either positive or negative) to topological stress. A comprehensive analysis of these topologically co-regulated gene clusters reveals pronounced preferences regarding their functional, regulatory and structural attributes. Genes that negatively respond to topological stress, are positioned in gene-dense pericentromeric regions, are more conserved and associated to essential functions, while upregulated gene clusters are preferentially located in the gene-sparse nuclear periphery, associated with secondary functions and under complex regulatory control. We propose that genome architecture evolves with a core of essential genes occupying a compact genomic ‘old town’, whereas more recently acquired, condition-specific genes tend to be located in a more spacious ‘suburban’ genomic periphery.

Genomic Responses of Mouse Synovial Fibroblasts During Tumor Necrosis Factor-Driven Arthritogenesis Greatly Mimic Those in Human Rheumatoid Arthritis

OBJECTIVE

Aberrant activation of synovial fibroblasts is a key determinant in the pathogenesis of rheumatoid arthritis (RA). The aims of this study were to produce a map of gene expression and epigenetic changes occurring in this cell type during disease progression in the human tumor necrosis factor (TNF)-transgenic model of arthritis and to identify commonalities with human synovial fibroblasts.

METHODS

We used deep sequencing to probe the transcriptome, the methylome, and the chromatin landscape of cultured mouse arthritogenic synovial fibroblasts at 3 stages of disease, as well as synovial fibroblasts stimulated with human TNF. We performed bioinformatics analyses at the gene, pathway, and network levels, compared mouse and human data, and validated selected genes in both species.

RESULTS

We found that synovial fibroblast arthritogenicity was reflected in distinct dynamic patterns of transcriptional dysregulation, which was especially enriched in pathways of the innate immune response and mesenchymal differentiation. A functionally representative subset of these changes was associated with methylation, mostly in gene bodies. The arthritogenic state involved highly active promoters, which were marked by histone H3K4 trimethylation. There was significant overlap between the mouse and human data at the level of dysregulated genes and to an even greater extent at the level of pathways.

CONCLUSION

This study is the first systematic examination of the pathogenic changes that occur in mouse synovial fibroblasts during progressive TNF-driven arthritogenesis. Significant correlations with the respective human RA synovial fibroblast data further validate the human TNF-transgenic mouse as a reliable model of the human disease. The resource of data generated in this work may serve as a framework for the discovery of novel pathogenic mechanisms and disease biomarkers.

Promyelocytic Leukemia Protein Is an Essential Regulator of Stem Cell Pluripotency and Somatic Cell Reprogramming

Promyelocytic leukemia protein (PML), the main constituent of PML nuclear bodies, regulates various physiological processes in different cell types. However, little is known about its functions in embryonic stem cells (ESC). Here, we report that PML contributes to ESC self-renewal maintenance by controlling cell-cycle progression and sustaining the expression of crucial pluripotency factors. Transcriptomic analysis and gain- or loss-of-function approaches showed that PML-deficient ESC exhibit morphological, metabolic, and growth properties distinct to naive and closer to the primed pluripotent state. During differentiation of embryoid bodies, PML influences cell-fate decisions between mesoderm and endoderm by controlling the expression of Tbx3. PML loss compromises the reprogramming ability of embryonic fibroblasts to induced pluripotent stem cells by inhibiting the transforming growth factor β pathway at the very early stages. Collectively, these results designate PML as a member of the regulatory network for ESC naive pluripotency and somatic cell reprogramming.

•

PML is essential for the maintenance of naive pluripotent cells

•

PML prevents the naive to primed pluripotency transition

•

PML influences cell-fate commitment through Tbx3 regulation

•

PML is required for iPSCs formation via regulation of TGF signaling pathway

Incorporating prior biological knowledge for network-based differential gene expression analysis using differentially weighted graphical LASSO

Background

Conventional differential gene expression analysis by methods such as student’s t-test, SAM, and Empirical Bayes often searches for statistically significant genes without considering the interactions among them. Network-based approaches provide a natural way to study these interactions and to investigate the rewiring interactions in disease versus control groups. In this paper, we apply weighted graphical LASSO (wgLASSO) algorithm to integrate a data-driven network model with prior biological knowledge (i.e., protein-protein interactions) for biological network inference. We propose a novel differentially weighted graphical LASSO (dwgLASSO) algorithm that builds group-specific networks and perform network-based differential gene expression analysis to select biomarker candidates by considering their topological differences between the groups.

Results

Through simulation, we showed that wgLASSO can achieve better performance in building biologically relevant networks than purely data-driven models (e.g., neighbor selection, graphical LASSO), even when only a moderate level of information is available as prior biological knowledge. We evaluated the performance of dwgLASSO for survival time prediction using two microarray breast cancer datasets previously reported by Bild et al. and van de Vijver et al. Compared with the top 10 significant genes selected by conventional differential gene expression analysis method, the top 10 significant genes selected by dwgLASSO in the dataset from Bild et al. led to a significantly improved survival time prediction in the independent dataset from van de Vijver et al. Among the 10 genes selected by dwgLASSO, UBE2S, SALL2, XBP1 and KIAA0922 have been confirmed by literature survey to be highly relevant in breast cancer biomarker discovery study. Additionally, we tested dwgLASSO on TCGA RNA-seq data acquired from patients with hepatocellular carcinoma (HCC) on tumors samples and their corresponding non-tumorous liver tissues. Improved sensitivity, specificity and area under curve (AUC) were observed when comparing dwgLASSO with conventional differential gene expression analysis method.

Conclusions

The proposed network-based differential gene expression analysis algorithm dwgLASSO can achieve better performance than conventional differential gene expression analysis methods by integrating information at both gene expression and network topology levels. The incorporation of prior biological knowledge can lead to the identification of biologically meaningful genes in cancer biomarker studies.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-017-1515-1) contains supplementary material, which is available to authorized users.

BGRMI: A method for inferring gene regulatory networks from time-course gene expression data and its application in breast cancer research

Reconstructing gene regulatory networks (GRNs) from gene expression data is a challenging problem. Existing GRN reconstruction algorithms can be broadly divided into model-free and model–based methods. Typically, model-free methods have high accuracy but are computation intensive whereas model-based methods are fast but less accurate. We propose Bayesian Gene Regulation Model Inference (BGRMI), a model-based method for inferring GRNs from time-course gene expression data. BGRMI uses a Bayesian framework to calculate the probability of different models of GRNs and a heuristic search strategy to scan the model space efficiently. Using benchmark datasets, we show that BGRMI has higher/comparable accuracy at a fraction of the computational cost of competing algorithms. Additionally, it can incorporate prior knowledge of potential gene regulation mechanisms and TF hetero-dimerization processes in the GRN reconstruction process. We incorporated existing ChIP-seq data and known protein interactions between TFs in BGRMI as sources of prior knowledge to reconstruct transcription regulatory networks of proliferating and differentiating breast cancer (BC) cells from time-course gene expression data. The reconstructed networks revealed key driver genes of proliferation and differentiation in BC cells. Some of these genes were not previously studied in the context of BC, but may have clinical relevance in BC treatment.