Markov State Models of gene regulatory networks

Background

Gene regulatory networks with dynamics characterized by multiple stable states underlie cell fate-decisions. Quantitative models that can link molecular-level knowledge of gene regulation to a global understanding of network dynamics have the potential to guide cell-reprogramming strategies. Networks are often modeled by the stochastic Chemical Master Equation, but methods for systematic identification of key properties of the global dynamics are currently lacking.

Results

The method identifies the number, phenotypes, and lifetimes of long-lived states for a set of common gene regulatory network models. Application of transition path theory to the constructed Markov State Model decomposes global dynamics into a set of dominant transition paths and associated relative probabilities for stochastic state-switching.

Conclusions

In this proof-of-concept study, we found that the Markov State Model provides a general framework for analyzing and visualizing stochastic multistability and state-transitions in gene networks. Our results suggest that this framework—adopted from the field of atomistic Molecular Dynamics—can be a useful tool for quantitative Systems Biology at the network scale.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-017-0394-4) contains supplementary material, which is available to authorized users.

Computational Approaches to Study Gene Regulatory Networks

The goal of the gene regulatory network (GRN) inference is to determine the interactions between genes given heterogeneous data capturing spatiotemporal gene expression. Since transcription underlines all cellular processes, the inference of GRN is the first step in deciphering the determinants of the dynamics of biological systems. Here, we first describe the generic steps of the inference approaches that rely on similarity measures and group the similarity measures based on the computational methodology used. For each group of similarity measures, we not only review the existing approaches but also describe specifically the detailed steps of the existing state-of-the-art algorithms.

Boolean Dynamic Modeling Approaches to Study Plant Gene Regulatory Networks: Integration, Validation, and Prediction

Mathematical models based on dynamical systems theory are well-suited tools for the integration of available molecular experimental data into coherent frameworks in order to propose hypotheses about the cooperative regulatory mechanisms driving developmental processes. Computational analysis of the proposed models using well-established methods enables testing the hypotheses by contrasting predictions with observations. Within such framework, Boolean gene regulatory network dynamical models have been extensively used in modeling plant development. Boolean models are simple and intuitively appealing, ideal tools for collaborative efforts between theorists and experimentalists. In this chapter we present protocols used in our group for the study of diverse plant developmental processes. We focus on conceptual clarity and practical implementation, providing directions to the corresponding technical literature.

Variable neighborhood search for reverse engineering of gene regulatory networks

A new search heuristic, Divided Neighborhood Exploration Search, designed to be used with inference algorithms such as Bayesian networks to improve on the reverse engineering of gene regulatory networks is presented. The approach systematically moves through the search space to find topologies representative of gene regulatory networks that are more likely to explain microarray data. In empirical testing it is demonstrated that the novel method is superior to the widely employed greedy search techniques in both the quality of the inferred networks and computational time.

Meis/UNC-62 isoform dependent regulation of CoupTF-II/UNC-55 and GABAergic motor neuron subtype differentiation

Gene regulatory networks orchestrate the assembly of functionally related cells within a cellular network. Subtle differences often exist among functionally related cells within such networks. How differences are created among cells with similar functions has been difficult to determine due to the complexity of both the gene and the cellular networks. In Caenorhabditis elegans, the DD and VD motor neurons compose a cross-inhibitory, GABAergic network that coordinates dorsal and ventral muscle contractions during locomotion. The Pitx2 homologue, UNC-30, acts as a terminal selector gene to create similarities and the Coup-TFII homologue, UNC-55, is necessary for creating differences between the two motor neuron classes. What is the organizing gene regulatory network responsible for initiating the expression of UNC-55 and thus creating differences between the DD and VD motor neurons? We show that the unc-55 promoter has modules that contain Meis/UNC-62 binding sites. These sites can be subdivided into regions that are capable of activating or repressing UNC-55 expression in different motor neurons. Interestingly, different isoforms of UNC-62 are responsible for the activation and the stabilization of unc-55 transcription. Furthermore, specific isoforms of UNC-62 are required for proper synaptic patterning of the VD motor neurons. Isoform specific regulation of differentiating neurons is a relatively unexplored area of research and presents a mechanism for creating differences among functionally related cells within a network.

The Local Edge Machine: inference of dynamic models of gene regulation

We present a novel approach, the Local Edge Machine, for the inference of regulatory interactions directly from time-series gene expression data. We demonstrate its performance, robustness, and scalability on in silico datasets with varying behaviors, sizes, and degrees of complexity. Moreover, we demonstrate its ability to incorporate biological prior information and make informative predictions on a well-characterized in vivo system using data from budding yeast that have been synchronized in the cell cycle. Finally, we use an atlas of transcription data in a mammalian circadian system to illustrate how the method can be used for discovery in the context of large complex networks.

The Impact of Gene Silencing on Horizontal Gene Transfer and Bacterial Evolution

The H-NS family of DNA-binding proteins is the subject of intense study due to its important roles in the regulation of horizontally acquired genes critical for virulence, antibiotic resistance, and metabolism. Xenogeneic silencing proteins, typified by the H-NS protein of Escherichia coli, specifically target and downregulate expression from AT-rich genes by selectively recognizing specific structural features unique to the AT-rich minor groove. In doing so, these proteins facilitate bacterial evolution; enabling these cells to engage in horizontal gene transfer while buffering potential any detrimental fitness consequences that may result from it. Xenogeneic silencing and counter-silencing explain how bacterial cells can evolve effective gene regulatory strategies in the face of rampant gene gain and loss and it has extended our understanding of bacterial gene regulation beyond the classic operon model. Here we review the structures and mechanisms of xenogeneic silencers as well as their impact on bacterial evolution. Several H-NS-like proteins appear to play a role in facilitating gene transfer by other mechanisms including by regulating transposition, conjugation, and participating in the activation of virulence loci like the locus of enterocyte effacement pathogenicity island of pathogenic strains of E. coli. Evidence suggests that the critical determinants that dictate whether an H-NS-like protein will be a silencer or will perform a different function do not lie in the DNA-binding domain but, rather, in the domains that control oligomerization. This suggests that H-NS-like proteins are transcription factors that both recognize and alter the shape of DNA to exert specific effects that include but are not limited to gene silencing.

Extending the Multi-level Method for the Simulation of Stochastic Biological Systems

The multi-level method for discrete-state systems, first introduced by Anderson and Higham (SIAM Multiscale Model Simul 10(1):146-179, 2012), is a highly efficient simulation technique that can be used to elucidate statistical characteristics of biochemical reaction networks. A single point estimator is produced in a cost-effective manner by combining a number of estimators of differing accuracy in a telescoping sum, and, as such, the method has the potential to revolutionise the field of stochastic simulation. In this paper, we present several refinements of the multi-level method which render it easier to understand and implement, and also more efficient. Given the substantial and complex nature of the multi-level method, the first part of this work reviews existing literature, with the aim of providing a practical guide to the use of the multi-level method. The second part provides the means for a deft implementation of the technique and concludes with a discussion of a number of open problems.

Direct lineage reprogramming via pioneer factors; a detour through developmental gene regulatory networks

Although many approaches have been employed to generate defined fate in vitro, the resultant cells often appear developmentally immature or incompletely specified, limiting their utility. Growing evidence suggests that current methods of direct lineage conversion may rely on the transition through a developmental intermediate. Here, I hypothesize that complete conversion between cell fates is more probable and feasible via reversion to a developmentally immature state. I posit that this is due to the role of pioneer transcription factors in engaging silent, unmarked chromatin and activating hierarchical gene regulatory networks responsible for embryonic patterning. Understanding these developmental contexts will be essential for the precise engineering of cell identity.

A catalog of Xenopus tropicalis transcription factors and their regional expression in the early gastrula stage embryo

Gene regulatory networks (GRNs) involve highly combinatorial interactions between transcription factors and short sequence motifs in cis-regulatory modules of target genes to control cellular phenotypes. The GRNs specifying most cell types are largely unknown and are the subject of wide interest. A catalog of transcription factors is a valuable tool toward obtaining a deeper understanding of the role of these critical effectors in any biological setting. Here we present a comprehensive catalog of the transcription factors for the diploid frog Xenopus tropicalis. We identify 1235 genes encoding DNA-binding transcription factors, comparable to the numbers found in typical mammalian species. In detail, the repertoire of X. tropicalis transcription factor genes is nearly identical to human and mouse, with the exception of zinc finger family members, and a small number of species/lineage-specific gene duplications and losses relative to the mammalian repertoires. We applied this resource to the identification of transcription factors differentially expressed in the early gastrula stage embryo. We find transcription factor enrichment in Spemann's organizer, the ventral mesoderm, ectoderm and endoderm, and report 218 TFs that show regionalized expression patterns at this stage. Many of these have not been previously reported as expressed in the early embryo, suggesting thus far unappreciated roles for many transcription factors in the GRNs regulating early development. We expect our transcription factor catalog will facilitate myriad studies using Xenopus as a model system to understand basic biology and human disease.

Redeployment of a conserved gene regulatory network during Aedes aegypti development

Changes in gene regulatory networks (GRNs) underlie the evolution of morphological novelty and developmental system drift. The fruitfly Drosophila melanogaster and the dengue and Zika vector mosquito Aedes aegypti have substantially similar nervous system morphology. Nevertheless, they show significant divergence in a set of genes co-expressed in the midline of the Drosophila central nervous system, including the master regulator single minded and downstream genes including short gastrulation, Star, and NetrinA. In contrast to Drosophila, we find that midline expression of these genes is either absent or severely diminished in A. aegypti. Instead, they are co-expressed in the lateral nervous system. This suggests that in A. aegypti this "midline GRN" has been redeployed to a new location while lost from its previous site of activity. In order to characterize the relevant GRNs, we employed the SCRMshaw method we previously developed to identify transcriptional cis-regulatory modules in both species. Analysis of these regulatory sequences in transgenic Drosophila suggests that the altered gene expression observed in A. aegypti is the result of trans-dependent redeployment of the GRN, potentially stemming from cis-mediated changes in the expression of sim and other as-yet unidentified regulators. Our results illustrate a novel "repeal, replace, and redeploy" mode of evolution in which a conserved GRN acquires a different function at a new site while its original function is co-opted by a different GRN. This represents a striking example of developmental system drift in which the dramatic shift in gene expression does not result in gross morphological changes, but in more subtle differences in development and function of the late embryonic nervous system.

Insights into Biological Complexity from Simple Foundations

We discuss an overtly "simple approach" to complex biological systems borrowing selectively from theoretical physics. The approach is framed by three maxims, and we show examples of its success in two different applications: investigating cellular robustness at the level of gene regulatory networks and quantifying rare events of DNA replication errors.

An anterior signaling center patterns and sizes the anterior neuroectoderm of the sea urchin embryo

Anterior signaling centers help specify and pattern the early anterior neuroectoderm (ANE) in many deuterostomes. In sea urchin the ANE is restricted to the anterior of the late blastula stage embryo, where it forms a simple neural territory comprising several types of neurons as well as the apical tuft. Here, we show that during early development, the sea urchin ANE territory separates into inner and outer regulatory domains that express the cardinal ANE transcriptional regulators FoxQ2 and Six3, respectively. FoxQ2 drives this patterning process, which is required to eliminate six3 expression from the inner domain and activate the expression of Dkk3 and sFRP1/5, two secreted Wnt modulators. Dkk3 and low expression levels of sFRP1/5 act additively to potentiate the Wnt/JNK signaling pathway governing the positioning of the ANE territory around the anterior pole, whereas high expression levels of sFRP1/5 antagonize Wnt/JNK signaling. sFRP1/5 and Dkk3 levels are rigidly maintained via autorepressive and cross-repressive interactions with Wnt signaling components and additional ANE transcription factors. Together, these data support a model in which FoxQ2 initiates an anterior patterning center that implements correct size and positions of ANE structures. Comparisons of functional and expression studies in sea urchin, hemichordate and chordate embryos reveal striking similarities among deuterostome ANE regulatory networks and the molecular mechanism that positions and defines ANE borders. These data strongly support the idea that the sea urchin embryo uses an ancient anterior patterning system that was present in the common ambulacrarian/chordate ancestor.

Biological computational approaches: new hopes to improve (re)programming robustness, regenerative medicine and cancer therapeutics

Hundreds of transcription factors (TFs) are expressed and work in each cell type, but the identity of the cells is defined and maintained through the activity of a small number of core TFs. Existing reprogramming strategies predominantly focus on the ectopic expression of core TFs of an intended fate in a given cell type regardless of the state of native/somatic gene regulatory networks (GRNs) of the starting cells. Interestingly, an important point is that how much products of the reprogramming, transdifferentiation and differentiation (programming) are identical to their in vivo counterparts. There is evidence that shows that direct fate conversions of somatic cells are not complete, with target cell identity not fully achieved. Manipulation of core TFs provides a powerful tool for engineering cell fate in terms of extinguishment of native GRNs, the establishment of a new GRN, and preventing installation of aberrant GRNs. Conventionally, core TFs are selected to convert one cell type into another mostly based on literature and the experimental identification of genes that are differentially expressed in one cell type compared to the specific cell types. Currently, there is not a universal standard strategy for identifying candidate core TFs. Remarkably, several biological computational platforms are developed, which are capable of evaluating the fidelity of reprogramming methods and refining existing protocols. The current review discusses some deficiencies of reprogramming technologies in the production of a pure population of authentic target cells. Furthermore, it reviews the role of computational approaches (e.g. CellNet, KeyGenes, Mogrify, etc.) in improving (re)programming methods and consequently in regenerative medicine and cancer therapeutics.

Systems biology of lens development: A paradigm for disease gene discovery in the eye

Over the past several decades, the biology of the developing lens has been investigated using molecular genetics-based approaches in various vertebrate model systems. These efforts, involving target gene knockouts or knockdowns, have led to major advances in our understanding of lens morphogenesis and the pathological basis of cataracts, as well as of other lens related eye defects. In particular, we now have a functional understanding of regulators such as Pax6, Six3, Sox2, Oct1 (Pou2f1), Meis1, Pnox1, Zeb2 (Sip1), Mab21l1, Foxe3, Tfap2a (Ap2-alpha), Pitx3, Sox11, Prox1, Sox1, c-Maf, Mafg, Mafk, Hsf4, Fgfrs, Bmp7, and Tdrd7 in this tissue. However, whether these individual regulators interact or their targets overlap, and the significance of such interactions during lens morphogenesis, is not well defined. The arrival of high-throughput approaches for gene expression profiling (microarrays, RNA-sequencing (RNA-seq), etc.), which can be coupled with chromatin immunoprecipitation (ChIP) or RNA immunoprecipitation (RIP) assays, along with improved computational resources and publically available datasets (e.g. those containing comprehensive protein-protein, protein-DNA information), presents new opportunities to advance our understanding of the lens tissue on a global systems level. Such systems-level knowledge will lead to the derivation of the underlying lens gene regulatory network (GRN), defined as a circuit map of the regulator-target interactions functional in lens development, which can be applied to expedite cataract gene discovery. In this review, we cover the various systems-level approaches such as microarrays, RNA-seq, and ChIP that are already being applied to lens studies and discuss strategies for assembling and interpreting these vast amounts of high-throughput information for effective dispersion to the scientific community. In particular, we discuss strategies for effective interpretation of this new information in the context of the rich knowledge obtained through the application of traditional single-gene focused experiments on the lens. Finally, we discuss our vision for integrating these diverse high-throughput datasets in a single web-based user-friendly tool iSyTE (integrated Systems Tool for Eye gene discovery) - a resource that is already proving effective in the identification and characterization of genes linked to lens development and cataract. We anticipate that application of a similar approach to other ocular tissues such as the retina and the cornea, and even other organ systems, will significantly impact disease gene discovery.

PeTTSy: a computational tool for perturbation analysis of complex systems biology models

Background

Over the last decade sensitivity analysis techniques have been shown to be very useful to analyse complex and high dimensional Systems Biology models. However, many of the currently available toolboxes have either used parameter sampling, been focused on a restricted set of model observables of interest, studied optimisation of a objective function, or have not dealt with multiple simultaneous model parameter changes where the changes can be permanent or temporary.

Results

Here we introduce our new, freely downloadable toolbox, PeTTSy (Perturbation Theory Toolbox for Systems). PeTTSy is a package for MATLAB which implements a wide array of techniques for the perturbation theory and sensitivity analysis of large and complex ordinary differential equation (ODE) based models. PeTTSy is a comprehensive modelling framework that introduces a number of new approaches and that fully addresses analysis of oscillatory systems. It examines sensitivity analysis of the models to perturbations of parameters, where the perturbation timing, strength, length and overall shape can be controlled by the user. This can be done in a system-global setting, namely, the user can determine how many parameters to perturb, by how much and for how long. PeTTSy also offers the user the ability to explore the effect of the parameter perturbations on many different types of outputs: period, phase (timing of peak) and model solutions. PeTTSy can be employed on a wide range of mathematical models including free-running and forced oscillators and signalling systems.

To enable experimental optimisation using the Fisher Information Matrix it efficiently allows one to combine multiple variants of a model (i.e. a model with multiple experimental conditions) in order to determine the value of new experiments. It is especially useful in the analysis of large and complex models involving many variables and parameters.

Conclusions

PeTTSy is a comprehensive tool for analysing large and complex models of regulatory and signalling systems. It allows for simulation and analysis of models under a variety of environmental conditions and for experimental optimisation of complex combined experiments. With its unique set of tools it makes a valuable addition to the current library of sensitivity analysis toolboxes. We believe that this software will be of great use to the wider biological, systems biology and modelling communities.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-016-0972-2) contains supplementary material, which is available to authorized users.

Implications of Developmental Gene Regulatory Networks Inside and Outside Developmental Biology

The insight that the genomic control of developmental process is encoded in the form of gene regulatory networks has profound impacts on many areas of modern bioscience. Most importantly, it affects developmental biology itself, as it means that a causal understanding of development requires knowledge of the architecture of regulatory network interactions. Furthermore, it follows that functional changes in developmental gene regulatory networks have to be considered as a primary mechanism for evolutionary process. We here discuss some of the recent advances in gene regulatory network biology and how they have affected our current understanding of development, evolution, and regulatory genomics.

Ancestral state reconstruction by comparative analysis of a GRN kernel operating in echinoderms

Diverse sampling of organisms across the five major classes in the phylum Echinodermata is beginning to reveal much about the structure and function of gene regulatory networks (GRNs) in development and evolution. Sea urchins are the most studied clade within this phylum, and recent work suggests there has been dramatic rewiring at the top of the skeletogenic GRN along the lineage leading to extant members of the euechinoid sea urchins. Such rewiring likely accounts for some of the observed developmental differences between the two major subclasses of sea urchins-cidaroids and euechinoids. To address effects of topmost rewiring on downstream GRN events, we cloned four downstream regulatory genes within the skeletogenic GRN and surveyed their spatiotemporal expression patterns in the cidaroid Eucidaris tribuloides. We performed phylogenetic analyses with homologs from other non-vertebrate deuterostomes and characterized their spatiotemporal expression by quantitative polymerase chain reaction (qPCR) and whole-mount in situ hybridization (WMISH). Our data suggest the erg-hex-tgif subcircuit, a putative GRN kernel, exhibits a mesoderm-specific expression pattern early in Eucidaris development that is directly downstream of the initial mesodermal GRN circuitry. Comparative analysis of the expression of this subcircuit in four echinoderm taxa allowed robust ancestral state reconstruction, supporting hypotheses that its ancestral function was to stabilize the mesodermal regulatory state and that it has been co-opted and deployed as a unit in mesodermal subdomains in distantly diverged echinoderms. Importantly, our study supports the notion that GRN kernels exhibit structural and functional modularity, locking down and stabilizing clade-specific, embryonic regulatory states.

Sea Urchin Morphogenesis

In the sea urchin morphogenesis follows extensive molecular specification. The specification controls the many morphogenetic events and these, in turn, precede patterning steps that establish the larval body plan. To understand how the embryo is built it was necessary to understand those series of molecular steps. Here an example of the historical sequence of those discoveries is presented as it unfolded over the last 50 years, the years during which major progress in understanding development of many animals and plants was documented by CTDB. In sea urchin development a rich series of experimental studies first established many of the phenomenological components of skeletal morphogenesis and patterning without knowledge of the molecular components. The many discoveries of transcription factors, signals, and structural proteins that contribute to the shape of the endoskeleton of the sea urchin larva then followed as molecular tools became available. A number of transcription factors and signals were discovered that were necessary for specification, morphogenesis, and patterning. Perturbation of the transcription factors and signals provided the means for assembling models of the gene regulatory networks used for specification and controlled the subsequent morphogenetic events. The earlier experimental information informed perturbation experiments that asked how patterning worked. As a consequence it was learned that ectoderm provides a series of patterning signals to the skeletogenic cells and as a consequence the skeletogenic cells secrete a highly patterned skeleton based on their ability to genotypically decode the localized reception of several signals. We still do not understand the complexity of the signals received by the skeletogenic cells, nor do we understand in detail how the genotypic information shapes the secreted skeletal biomineral, but the current knowledge at least outlines the sequence of events and provides a useful template for future discoveries.

Transcriptional networks associated with 5-alpha-dihydrotestosterone in the fathead minnow (Pimephales promelas) ovary

Androgens play a significant role in regulating oogenesis in teleost fishes. The androgen dihydrotestosterone (DHT) is a potent non-aromatizable androgen involved in sexual differentiation in mammals; however, its actions are not well understood in teleost fish. To better characterize the physiological role of DHT in the fathead minnow (FHM) ovary on a temporal scale, in vitro assays for 17β-estradiol (E2) production were conducted in parallel with microarray analysis. Ovarian explants were incubated at different concentrations of DHT (10(-6), 10(-7), and 10(-8)M DHT) in three separate experiments conducted at 6, 9, and 12h. DHT treatment resulted in a rapid and consistent increase in E2 production from the ovary at all three time points. Therefore, DHT may act to shift the balance of metabolites in the steroidogenic pathway within the ovary. Major biological themes affected by DHT in the ovary in one or more of the time points included those related to blood (e.g. vasodilation, blood vessel contraction, clotting), lipids (e.g. lipid storage, cholesterol metabolism, lipid degradation) and reproduction (e.g. hormone and steroid metabolism). Gene networks related to immune responses and calcium signaling were also affected by DHT, suggesting that this androgen may play a role in regulating these processes in the ovary. This study detected no change in mRNA levels of steroidogenic enzymes (cyp19a1, star, 11βhsd, 17βhsd, srd5a isoforms), suggesting that the observed increase in E2 production is likely more dependent on the pre-existing gene or protein complement in the ovary rather than the de novo expression of transcripts. This study increases knowledge regarding the roles of DHT and androgens in general in the teleost ovary and identifies molecular signaling pathways that may be associated with increased E2 production.

Identifying Gene Regulatory Networks in Arabidopsis by In Silico Prediction, Yeast-1-Hybrid, and Inducible Gene Profiling Assays

A system-wide understanding of gene regulation will provide deep insights into plant development and physiology. In this chapter we describe a threefold approach to identify the gene regulatory networks in Arabidopsis thaliana that function in a specific tissue or biological process. Since no single method is sufficient to establish comprehensive and high-confidence gene regulatory networks, we focus on the integration of three approaches. First, we describe an in silico prediction method of transcription factor-DNA binding, then an in vivo assay of transcription factor-DNA binding by yeast-1-hybrid and lastly the identification of co-expression clusters by transcription factor induction in planta. Each of these methods provides a unique tool to advance our understanding of gene regulation, and together provide a robust model for the generation of gene regulatory networks.

Information theory in systems biology. Part I: Gene regulatory and metabolic networks

"A Mathematical Theory of Communication", was published in 1948 by Claude Shannon to establish a framework that is now known as information theory. In recent decades, information theory has gained much attention in the area of systems biology. The aim of this paper is to provide a systematic review of those contributions that have applied information theory in inferring or understanding of biological systems. Based on the type of system components and the interactions between them, we classify the biological systems into 4 main classes: gene regulatory, metabolic, protein-protein interaction and signaling networks. In the first part of this review, we attempt to introduce most of the existing studies on two types of biological networks, including gene regulatory and metabolic networks, which are founded on the concepts of information theory.

Transcriptional master regulator analysis in breast cancer genetic networks

Gene regulatory networks account for the delicate mechanisms that control gene expression. Under certain circumstances, gene regulatory programs may give rise to amplification cascades. Such transcriptional cascades are events in which activation of key-responsive transcription factors called master regulators trigger a series of gene expression events. The action of transcriptional master regulators is then important for the establishment of certain programs like cell development and differentiation. However, such cascades have also been related with the onset and maintenance of cancer phenotypes. Here we present a systematic implementation of a series of algorithms aimed at the inference of a gene regulatory network and analysis of transcriptional master regulators in the context of primary breast cancer cells. Such studies were performed in a highly curated database of 880 microarray gene expression experiments on biopsy-captured tissue corresponding to primary breast cancer and healthy controls. Biological function and biochemical pathway enrichment analyses were also performed to study the role that the processes controlled - at the transcriptional level - by such master regulators may have in relation to primary breast cancer. We found that transcription factors such as AGTR2, ZNF132, TFDP3 and others are master regulators in this gene regulatory network. Sets of genes controlled by these regulators are involved in processes that are well-known hallmarks of cancer. This kind of analyses may help to understand the most upstream events in the development of phenotypes, in particular, those regarding cancer biology.

Brachyury and SMAD signalling collaboratively orchestrate distinct mesoderm and endoderm gene regulatory networks in differentiating human embryonic stem cells

The transcription factor brachyury (T, BRA) is one of the first markers of gastrulation and lineage specification in vertebrates. Despite its wide use and importance in stem cell and developmental biology, its functional genomic targets in human cells are largely unknown. Here, we use differentiating human embryonic stem cells to study the role of BRA in activin A-induced endoderm and BMP4-induced mesoderm progenitors. We show that BRA has distinct genome-wide binding landscapes in these two cell populations, and that BRA interacts and collaborates with SMAD1 or SMAD2/3 signalling to regulate the expression of its target genes in a cell-specific manner. Importantly, by manipulating the levels of BRA in cells exposed to different signalling environments, we demonstrate that BRA is essential for mesoderm but not for endoderm formation. Together, our data illuminate the function of BRA in the context of human embryonic development and show that the regulatory role of BRA is context dependent. Our study reinforces the importance of analysing the functions of a transcription factor in different cellular and signalling environments.

Are ta-siRNAs only originated from the cleavage site of miRNA on its target RNAs and phased in 21-nt increments?

Trans-acting siRNAs (ta-siRNAs) are a class of small RNAs playing crucial roles in the regulation of plant gene expression. According to the canonical model, specific miRNA-guided cleavage of a TAS transcript triggers and sets the registry for the subsequent production of ta-siRNAs at 21-nt increments from the cleavage site. However, a previously validated 22-nt ta-siR2140 indicated that ta-siRNAs might be initiated from other phase increments and registers, which resulted in massive ta-siRNAs missing in the canonical model. To test this hypothesis, we employed high-throughput sequencing data to thoroughly identify the miR173-triggered ta-siRNAs from TAS1/TAS2 transcripts. As a result, thousands of phased siRNAs not generated through the canonical pathway were identified and 110 novel siRNA-target interactions were further validated based on degradome sequencing data. Based on these results, we propose that the canonical biogenesis model of ta-siRNAs should be modified in order to recruit the previously unidentified ta-siRNA candidates.