Gene Regulatory Network Analysis Research Articles

Qualitative analysis of gene regulatory networks by temporal logic

In this article we propose a novel formalism to model and analyse gene regulatory networks using a well-established formal verification technique. We model the possible behaviours of networks by logical formulae in linear temporal logic (LTL). By checking the satisfiability of LTL, it is possible to check whether some or all behaviours satisfy a given biological property, which is difficult in quantitative analyses such as the ordinary differential equation approach. Owing to the complexity of LTL satisfiability checking, analysis of large networks is generally intractable in this method. To mitigate this computational difficulty, we developed two methods. One is a modular checking method where we divide a network into subnetworks, check them individually, and then integrate them. The other is an approximate analysis method in which we specify behaviours in simpler formulae which compress or expand the possible behaviours of networks. In the approximate method, we focused on network motifs and presented approximate specifications for them. We confirmed by experiments that both methods improved the analysis of large networks.

Combined analysis of gene regulatory network and SNV information enhances identification of potential gene markers in mouse knockout studies with small number of samples.

RNA-sequencing is widely used to measure gene expression level at the whole genome level. Comparing expression data from control and case studies provides good insight on potential gene markers for phenotypes. However, discovering gene markers that represent phenotypic differences in a small number of samples remains a challenging task, since finding gene markers using standard differential expressed gene methods produces too many candidate genes and the number of candidates varies at different threshold values. In addition, in a small number of samples, the statistical power is too low to discriminate whether gene expressions were altered by genetic differences or not. In this study, to address this challenge, we purpose a four-step filtering method that predicts gene markers from RNA-sequencing data of mouse knockout studies by utilizing a gene regulatory network constructed from omics data in the public domain, biological knowledge from curated pathways, and information of single-nucleotide variants. Our prediction method was not only able to reduce the number of candidate genes than the differentialy expressed gene-only filtered method, but also successfully predicted significant genes that were reported in research findings of the data contributors.

Gene networks and transcription factor motifs defining the differentiation of stem cells into hepatocyte-like cells

The differentiation of stem cells to hepatocyte-like cells (HLC) offers the perspective of unlimited supply of human hepatocytes. However, the degree of differentiation of HLC remains controversial. To obtain an unbiased characterization, we performed a transcriptomic study with HLC derived from human embryonic and induced stem cells (ESC, hiPSC) from three different laboratories. Genome-wide gene expression profiles of ESC and HLC were compared to freshly isolated and up to 14days cultivated primary human hepatocytes. Gene networks representing successful and failed hepatocyte differentiation, and the transcription factors involved in their regulation were identified. Gene regulatory network analysis demonstrated that HLC represent a mixed cell type with features of liver, intestine, fibroblast and stem cells. The "unwanted" intestinal features were associated with KLF5 and CDX2 transcriptional networks. Cluster analysis identified highly correlated groups of genes associated with mature liver functions (n=1057) and downregulated proliferation associated genes (n=1562) that approach levels of primary hepatocytes. However, three further clusters containing 447, 101, and 505 genes failed to reach levels of hepatocytes. Key TF of two of these clusters include SOX11, FOXQ1, and YBX3. The third unsuccessful cluster, controlled by HNF1, CAR, FXR, and PXR, strongly overlaps with genes repressed in cultivated hepatocytes compared to freshly isolated hepatocytes, suggesting that current in vitro conditions lack stimuli required to maintain gene expression in hepatocytes, which consequently also explains a corresponding deficiency of HLC. The present gene regulatory network approach identifies key transcription factors which require modulation to improve HLC differentiation.

Systems genetics identifies Sestrin 3 as a regulator of a proconvulsant gene network in human epileptic hippocampus

Gene-regulatory network analysis is a powerful approach to elucidate the molecular processes and pathways underlying complex disease. Here we employ systems genetics approaches to characterize the genetic regulation of pathophysiological pathways in human temporal lobe epilepsy (TLE). Using surgically acquired hippocampi from 129 TLE patients, we identify a gene-regulatory network genetically associated with epilepsy that contains a specialized, highly expressed transcriptional module encoding proconvulsive cytokines and Toll-like receptor signalling genes. RNA sequencing analysis in a mouse model of TLE using 100 epileptic and 100 control hippocampi shows the proconvulsive module is preserved across-species, specific to the epileptic hippocampus and upregulated in chronic epilepsy. In the TLE patients, we map the trans-acting genetic control of this proconvulsive module to Sestrin 3 (SESN3), and demonstrate that SESN3 positively regulates the module in macrophages, microglia and neurons. Morpholino-mediated Sesn3 knockdown in zebrafish confirms the regulation of the transcriptional module, and attenuates chemically induced behavioural seizures in vivo.

Gene regulatory network analysis reveals differences in site-specific cell fate determination in mammalian brain.

Neurogenesis—the generation of new neurons—is an ongoing process that persists in the adult mammalian brain of several species, including humans. In this work we analyze two discrete brain regions: the subventricular zone (SVZ) lining the walls of the lateral ventricles; and the subgranular zone (SGZ) of the dentate gyrus (DG) of the hippocampus in mice and shed light on the SVZ and SGZ specific neurogenesis. We propose a computational model that relies on the construction and analysis of region specific gene regulatory networks (GRNs) from the publicly available data on these two regions. Using this model a number of putative factors involved in neuronal stem cell (NSC) identity and maintenance were identified. We also demonstrate potential gender and niche-derived differences based on cell surface and nuclear receptors via Ar, Hif1a, and Nr3c1. We have also conducted cell fate determinant analysis for SVZ NSC populations to Olfactory Bulb interneurons and SGZ NSC populations to the granule cells of the Granular Cell Layer. We report 31 candidate cell fate determinant gene pairs, ready to be validated. We focus on Ar—Pax6 in SVZ and Sox2—Ncor1 in SGZ. Both pairs are expressed and localized in the suggested anatomical structures as shown by in situ hybridization and found to physically interact. Finally, we conclude that there are fundamental differences between SGZ and SVZ neurogenesis. We argue that these regulatory mechanisms are linked to the observed differential neurogenic potential of these regions. The presence of nuclear and cell surface receptors in the region specific regulatory circuits indicate the significance of niche derived extracellular factors, hormones and region specific factors such as the oxygen sensitivity, dictating SGZ and SVZ specific neurogenesis.

Abstract IA5: Using cross-species analysis of genome-wide regulatory networks to identify drivers of cancer malignancy

Abstract We have been interested in using systems biology approaches for cross-species interrogations of cancer phenotypes. To identify regulatory drivers of prostate cancer malignancy, we have performed cross-species analysis of complementary human and mouse gene regulatory networks (interactomes) assembled from molecular profiles of human tumors and genetically engineered prostate cancer mouse models, respectively. Genome-wide analysis of these interactomes reveals extensive conservation of their transcriptional programs, particularly those known to be essential for prostate tumorigenesis. Cross-species interactome interrogation and subsequent experimental validation have elucidated a synergistic interaction of FOXM1 and CENPF that drives prostate cancer malignancy. Silencing of both FOXM1 and CENPF, but not either individually, abrogates prostate tumor growth in vivo via cooperative regulation of key signaling pathways in prostate cancer progression. Furthermore, co-expression of FOXM1 and CENPF is a highly robust prognostic indicator of disease outcome. Taken together, these analyses suggest that genome-wide cross-species interrogation of regulatory networks represents a valuable new strategy to identify causal mechanisms of human cancer. Citation Format: Cory Abate-Shen. Using cross-species analysis of genome-wide regulatory networks to identify drivers of cancer malignancy. [abstract]. In: Proceedings of the AACR Special Conference: The Translational Impact of Model Organisms in Cancer; Nov 5-8, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2014;12(11 Suppl):Abstract nr IA5.

Systematic identification of transcriptional and post-transcriptional regulations in human respiratory epithelial cells during influenza A virus infection.

BackgroundRespiratory epithelial cells are the primary target of influenza virus infection in human. However, the molecular mechanisms of airway epithelial cell responses to viral infection are not fully understood. Revealing genome-wide transcriptional and post-transcriptional regulatory relationships can further advance our understanding of this problem, which motivates the development of novel and more efficient computational methods to simultaneously infer the transcriptional and post-transcriptional regulatory networks.ResultsHere we propose a novel framework named SITPR to investigate the interactions among transcription factors (TFs), microRNAs (miRNAs) and target genes. Briefly, a background regulatory network on a genome-wide scale (~23,000 nodes and ~370,000 potential interactions) is constructed from curated knowledge and algorithm predictions, to which the identification of transcriptional and post-transcriptional regulatory relationships is anchored. To reduce the dimension of the associated computing problem down to an affordable size, several topological and data-based approaches are used. Furthermore, we propose the constrained LASSO formulation and combine it with the dynamic Bayesian network (DBN) model to identify the activated regulatory relationships from time-course expression data. Our simulation studies on networks of different sizes suggest that the proposed framework can effectively determine the genuine regulations among TFs, miRNAs and target genes; also, we compare SITPR with several selected state-of-the-art algorithms to further evaluate its performance. By applying the SITPR framework to mRNA and miRNA expression data generated from human lung epithelial A549 cells in response to A/Mexico/InDRE4487/2009 (H1N1) virus infection, we are able to detect the activated transcriptional and post-transcriptional regulatory relationships as well as the significant regulatory motifs.ConclusionCompared with other representative state-of-the-art algorithms, the proposed SITPR framework can more effectively identify the activated transcriptional and post-transcriptional regulations simultaneously from a given background network. The idea of SITPR is generally applicable to the analysis of gene regulatory networks in human cells. The results obtained for human respiratory epithelial cells suggest the importance of the transcriptional, post-transcriptional regulations as well as their synergies in the innate immune responses against IAV infection.Electronic supplementary materialThe online version of this article (doi:10.1186/1471-2105-15-336) contains supplementary material, which is available to authorized users.

Finite-state discrete-time Markov chain models of gene regulatory networks

In this study, Markov chain models of gene regulatory networks (GRN) are developed. These models make it possible to apply the well-known theory and tools of Markov chains to GRN analysis. A new kind of finite interaction graph called a combinatorial net is introduced to represent formally a GRN and its transition graphs constructed from interaction graphs. The system dynamics are defined as a random walk on the transition graph, which is a Markov chain. A novel concurrent updating scheme (evolution rule) is developed to determine transitions in a transition graph. The proposed scheme is based on the firing of a random set of non-steady-state vertices in a combinatorial net. It is demonstrated that this novel scheme represents an advance in asynchronicity modeling. The theorem that combinatorial nets with this updating scheme can asynchronously compute a maximal independent set of graphs is also proved. As proof of concept, a number of simple combinatorial models are presented here: a discrete auto-regression model, a bistableswitch, an Elowitz repressilator, and a self-activation model, and it is shown that these models exhibit well-known properties.

Inference of gene regulatory networks incorporating multi-source biological knowledge via a state space model with L1 regularization.

Comprehensive understanding of gene regulatory networks (GRNs) is a major challenge in the field of systems biology. Currently, there are two main approaches in GRN analysis using time-course observation data, namely an ordinary differential equation (ODE)-based approach and a statistical model-based approach. The ODE-based approach can generate complex dynamics of GRNs according to biologically validated nonlinear models. However, it cannot be applied to ten or more genes to simultaneously estimate system dynamics and regulatory relationships due to the computational difficulties. The statistical model-based approach uses highly abstract models to simply describe biological systems and to infer relationships among several hundreds of genes from the data. However, the high abstraction generates false regulations that are not permitted biologically. Thus, when dealing with several tens of genes of which the relationships are partially known, a method that can infer regulatory relationships based on a model with low abstraction and that can emulate the dynamics of ODE-based models while incorporating prior knowledge is urgently required. To accomplish this, we propose a method for inference of GRNs using a state space representation of a vector auto-regressive (VAR) model with L1 regularization. This method can estimate the dynamic behavior of genes based on linear time-series modeling constructed from an ODE-based model and can infer the regulatory structure among several tens of genes maximizing prediction ability for the observational data. Furthermore, the method is capable of incorporating various types of existing biological knowledge, e.g., drug kinetics and literature-recorded pathways. The effectiveness of the proposed method is shown through a comparison of simulation studies with several previous methods. For an application example, we evaluated mRNA expression profiles over time upon corticosteroid stimulation in rats, thus incorporating corticosteroid kinetics/dynamics, literature-recorded pathways and transcription factor (TF) information.

Genome-wide analysis of the skeletogenic gene regulatory network of sea urchins

Genome-wide analysis of the skeletogenic gene regulatory network of sea urchins

The tissue-specific transcriptomic landscape of the mid-gestational mouse embryo.

Differential gene expression is a prerequisite for the formation of multiple cell types from the fertilized egg during embryogenesis. Understanding the gene regulatory networks controlling cellular differentiation requires the identification of crucial differentially expressed control genes and, ideally, the determination of the complete transcriptomes of each individual cell type. Here, we have analyzed the transcriptomes of six major tissues dissected from mid-gestational (TS12) mouse embryos. Approximately one billion reads derived by RNA-seq analysis provided extended transcript lengths, novel first exons and alternative transcripts of known genes. We have identified 1375 genes showing tissue-specific expression, providing gene signatures for each of the six tissues. In addition, we have identified 1403 novel putative long noncoding RNA gene loci, 439 of which show differential expression. Our analysis provides the first complete transcriptome data for the mouse embryo. It offers a rich data source for the analysis of individual genes and gene regulatory networks controlling mid-gestational development.

Genetic variation in aldo‐keto reductase 1D1 (AKR1D1) affects expression and activity of multiple cytochrome P450s (292.1)

Human liver gene regulatory network analysis identified a cytochrome P450 subnetwork with AKR1D1 as its key regulatory driver. This study assessed the biologic importance of AKR1D1 [key enzyme in synthesis of bile acids, ligand activators of FXR, PXR and CAR, known P450 regulators] to hepatic P450 expression. Overexpression of AKR1D1 in primary human hepatocytes led to increased expression of CYP3A4, 2C8, 2C9, 2C19, and 2B6. Conversely, AKR1D1 knockdown decreased expression of these P450s. Resequencing AKR1D1 from donor livers identified a 3'‐UTR (rs1872930) SNP that was significantly associated with higher AKR1D1 mRNA. AKR1D1 3'‐UTR‐luciferase reporter studies showed that variant allele resulted in higher luciferase activity, suggesting that the SNP increases AKR1D1 mRNA stability and/or translation efficiency. AKR1D1 3'‐UTR SNP was significantly associated with increased hepatic mRNA of multiple P450s (CYP3A4, 2C8, 2C9, 2C19, and 2B6) and CYP3A4, 2C8, 2C19, and 2B6 activities. After adjusting for multiple testing, association remained significant for AKR1D1, CYP2C9, and CYP2C8 mRNA and CYP2C8 activity. These results provide new insights into the variation in expression and activity of P450s that can account for interindividual differences in drug metabolism/efficacy. In conclusion, we provide the first experimental evidence supporting a role for AKR1D1 as a key genetic regulator of P450 network.

Genetic variation in aldo‐keto reductase 1D1 (AKR1D1) affects expression and activity of multiple cytochrome P450s (LB595)

Human liver gene regulatory network analysis identified a cytochrome P450 subnetwork with AKR1D1 as its key regulatory driver. This study assessed the biologic importance of AKR1D1 [key enzyme in synthesis of bile acids, ligand activators of FXR, PXR and CAR, known P450 regulators] to hepatic P450 expression. Overexpression of AKR1D1 in primary human hepatocytes led to increased expression of CYP3A4, 2C8, 2C9, 2C19, and 2B6. Conversely, AKR1D1 knockdown decreased expression of these P450s. Resequencing AKR1D1 from donor livers identified a 3'‐UTR (rs1872930) SNP that was significantly associated with higher AKR1D1 mRNA. AKR1D1 3'‐UTR‐luciferase reporter studies showed that variant allele resulted in higher luciferase activity, suggesting that the SNP increases AKR1D1 mRNA stability and/or translation efficiency. AKR1D1 3'‐UTR SNP was significantly associated with increased hepatic mRNA of multiple P450s (CYP3A4, 2C8, 2C9, 2C19, and 2B6) and CYP3A4, 2C8, 2C19, and 2B6 activities. After adjusting for multiple testing, association remained significant for AKR1D1, CYP2C9, and CYP2C8 mRNA and CYP2C8 activity. These results provide new insights into the variation in expression and activity of P450s that can account for interindividual differences in drug metabolism/efficacy. In conclusion, we provide the first experimental evidence supporting a role for AKR1D1 as a key genetic regulator of P450 network.

Comparative metabolism of cellulose, sophorose and glucose in Trichoderma reeseiusing high-throughput genomic and proteomic analyses

BackgroundThe filamentous fungus Trichoderma reesei is a major producer of lignocellulolytic enzymes utilized by bioethanol industries. However, to achieve low cost second generation bioethanol production on an industrial scale an efficient mix of hydrolytic enzymes is required for the deconstruction of plant biomass. In this study, we investigated the molecular basis for lignocellulose-degrading enzyme production T. reesei during growth in cellulose, sophorose, and glucose.ResultsWe examined and compared the transcriptome and differential secretome (2D-DIGE) of T. reesei grown in cellulose, sophorose, or glucose as the sole carbon sources. By applying a stringent cut-off threshold 2,060 genes were identified as being differentially expressed in at least one of the respective carbon source comparisons. Hierarchical clustering of the differentially expressed genes identified three possible regulons, representing 123 genes controlled by cellulose, 154 genes controlled by sophorose and 402 genes controlled by glucose. Gene regulatory network analyses of the 692 genes differentially expressed between cellulose and sophorose, identified only 75 and 107 genes as being specific to growth in sophorose and cellulose, respectively. 2D-DIGE analyses identified 30 proteins exclusive to sophorose and 37 exclusive to cellulose. A correlation of 70.17% was obtained between transcription and secreted protein profiles.ConclusionsOur data revealed new players in cellulose degradation such as accessory proteins with non-catalytic functions secreted in different carbon sources, transporters, transcription factors, and CAZymes, that specifically respond in response to either cellulose or sophorose.

Unraveling Regulation of the Small Heat Shock Proteins by the Heat Shock Factor HvHsfB2c in Barley: Its Implications in Drought Stress Response and Seed Development

The rapid increase in heat shock proteins upon exposure to damaging stresses and during plant development related to desiccation events reveal their dual importance in plant development and stress tolerance. Genome-wide sequence survey identified 20 non-redundant small heat shock proteins (sHsp) and 22 heat shock factor (Hsf) genes in barley. While all three major classes (A, B, C) of Hsfs are localized in nucleus, the 20 sHsp gene family members are localized in different cell organelles like cytoplasm, mitochondria, plastid and peroxisomes. Hsf and sHsp members are differentially regulated during drought and at different seed developmental stages suggesting the importance of chaperone role under drought as well as seed development. In silico cis-regulatory motif analysis of Hsf promoters showed an enrichment with abscisic acid responsive cis-elements (ABRE), implying regulatory role of ABA in mediating transcriptional response of HvsHsf genes. Gene regulatory network analysis identified HvHsfB2c as potential central regulator of the seed-specific expression of several HvsHsps including 17.5CI sHsp. These results indicate that HvHsfB2c is co-expressed in the central hub of small Hsps and therefore it may be regulating the expression of several HvsHsp subclasses HvHsp16.88-CI, HvHsp17.5-CI and HvHsp17.7-CI. The in vivo relevance of binding specificity of HvHsfB2C transcription factor to HSE-element present in the promoter of HvSHP17.5-CI under heat stress exposure is confirmed by gel shift and LUC-reporter assays. Further, we isolated 477 bp cDNA from barley encoding a 17.5 sHsp polypeptide, which was predominantly upregulated under drought stress treatments and also preferentially expressed in developing seeds. Recombinant HvsHsp17.5-CI protein was expressed in E. coli and purified to homogeneity, which displayed in vitro chaperone activity. The predicted structural model of HvsHsp-17.5-CI protein suggests that the α-crystallin domain is evolutionarily highly conserved.

JAK2-Centered Interactome Hotspot Identified by an Integrative Network Algorithm in Acute Stanford Type A Aortic Dissection

The precise mechanisms underlying dissections, especially those without connective tissue diseases or congenital vascular diseases, are incompletely understood. This study attempted to identify both the expression profile of the dissected ascending aorta and the interactome hotspots associated with the disease, using microarray technology and gene regulatory network analysis. There were 2,737 genes differentially expressed between patients with acute Stanford type A aortic dissection and controls. Eight interactome hotspots significantly associated with aortic dissection were identified by an integrative network algorithm. In particular, we identified a JAK2-centered expression module, which was validated in an independent gene expression microarray data set, and which was characterized by over-expressed cytokines and receptors in acute aortic dissection cases, indicating that JAK2 may play a key role in the inflammatory process, which potentially contributes to the occurrence of acute aortic dissection. Overall, the analytical strategy used in this study offered the possibility to identify functional relevant network modules and subsequently facilitated the biological interpretation in the complicated disease.

Genome-wide analysis of the skeletogenic gene regulatory network of sea urchins

A central challenge of developmental and evolutionary biology is to understand the transformation of genetic information into morphology. Elucidating the connections between genes and anatomy will require model morphogenetic processes that are amenable to detailed analysis of cell/tissue behaviors and to systems-level approaches to gene regulation. The formation of the calcified endoskeleton of the sea urchin embryo is a valuable experimental system for developing such an integrated view of the genomic regulatory control of morphogenesis. A transcriptional gene regulatory network (GRN) that underlies the specification of skeletogenic cells (primary mesenchyme cells, or PMCs) has recently been elucidated. In this study, we carried out a genome-wide analysis of mRNAs encoded by effector genes in the network and uncovered transcriptional inputs into many of these genes. We used RNA-seq to identify >400 transcripts differentially expressed by PMCs during gastrulation, when these cells undergo a striking sequence of behaviors that drives skeletal morphogenesis. Our analysis expanded by almost an order of magnitude the number of known (and candidate) downstream effectors that directly mediate skeletal morphogenesis. We carried out genome-wide analysis of (1) functional targets of Ets1 and Alx1, two pivotal, early transcription factors in the PMC GRN, and (2) functional targets of MAPK signaling, a pathway that plays an essential role in PMC specification. These studies identified transcriptional inputs into >200 PMC effector genes. Our work establishes a framework for understanding the genomic regulatory control of a major morphogenetic process and has important implications for reconstructing the evolution of biomineralization in metazoans.

Modelling and analysis of gene regulatory networks based on the G-network

G-networks are a class of stochastic models that have had a broad range of applications ranging from the performance analysis of computer systems and networks to the modelling of gene regulatory networks. Gene regulatory networks consist of thousands of genes and proteins which are dynamically interacting with each other. Once these regulatory structures are revealed, it is necessary to understand their dynamical behaviours since pathway activities could be changed by their given conditions. This review mainly focuses on a stochastic GRN modelling techniques based on G-networks which provide the analytical steady-state solution of a system for efficient GRN dynamics modelling. Three applications of the G-network model to GRNs show that this novel approach can serve to detect abnormalities from protein expression data, and that they can help to explicit the behaviour of complicated GRN models by dividing the gene regulatory processes into DNA and protein layers.

Nonlinear quantitative radiation sensitivity prediction model based on NCI-60 cancer cell lines.

We proposed a nonlinear model to perform a novel quantitative radiation sensitivity prediction. We used the NCI-60 panel, which consists of nine different cancer types, as the platform to train our model. Important radiation therapy (RT) related genes were selected by significance analysis of microarrays (SAM). Orthogonal latent variables (LVs) were then extracted by the partial least squares (PLS) method as the new compressive input variables. Finally, support vector machine (SVM) regression model was trained with these LVs to predict the SF2 (the surviving fraction of cells after a radiation dose of 2 Gy γ-ray) values of the cell lines. Comparison with the published results showed significant improvement of the new method in various ways: (a) reducing the root mean square error (RMSE) of the radiation sensitivity prediction model from 0.20 to 0.011; and (b) improving prediction accuracy from 62% to 91%. To test the predictive performance of the gene signature, three different types of cancer patient datasets were used. Survival analysis across these different types of cancer patients strongly confirmed the clinical potential utility of the signature genes as a general prognosis platform. The gene regulatory network analysis identified six hub genes that are involved in canonical cancer pathways.

Detecting cellular reprogramming determinants by differential stability analysis of gene regulatory networks

BackgroundCellular differentiation and reprogramming are processes that are carefully orchestrated by the activation and repression of specific sets of genes. An increasing amount of experimental results show that despite the large number of genes participating in transcriptional programs of cellular phenotypes, only few key genes, which are coined here as reprogramming determinants, are required to be directly perturbed in order to induce cellular reprogramming. However, identification of reprogramming determinants still remains a combinatorial problem, and the state-of-art methods addressing this issue rests on exhaustive experimentation or prior knowledge to narrow down the list of candidates.ResultsHere we present a computational method, without any preliminary selection of candidate genes, to identify reduced subsets of genes, which when perturbed can induce transitions between cellular phenotypes. The method relies on the expression profiles of two stable cellular phenotypes along with a topological analysis stability elements in the gene regulatory network that are necessary to cause this multi-stability. Since stable cellular phenotypes can be considered as attractors of gene regulatory networks, cell fate and cellular reprogramming involves transition between these attractors, and therefore current method searches for combinations of genes that are able to destabilize a specific initial attractor and stabilize the final one in response to the appropriate perturbations.ConclusionsThe method presented here represents a useful framework to assist researchers in the field of cellular reprogramming to design experimental strategies with potential applications in the regenerative medicine and disease modelling.