Project: Systems approach to gene regulation biology through nuclear receptors

Nuclear receptors (NRs) are key factors regulating fundamental cell fate decisions during organogenesis, growth, homeostatic tissue maintenance and renewal. Through influencing the expression of genes within complex regulatory networks, NRs affect a diverse spectrum of physiological and pathological processes, including differentiation, cellular homeostasis, cancer and metabolic diseases. Prime examples are estrogen-dependent breast cancer and androgen-dependent prostate cancer. Transcription of NR-regulated genes is a complex, tightly regulated process where distinct NRs, in conjunction with other transcription factors (TFs), the basal transcription machinery and covalent modifications to chromatin, collectively act to regulate gene expression. The major objectives of SYNERGY (Systems approach to gene regulation biology through nuclear receptors) are to characterize the roles of four nuclear receptors (NRs), RNA polymerase II and four histone marks in tumor cells and in normal breast and prostate cells. We will determine NR binding through ChIP-seq; gene expression with RNA-seq and place these datasets in context with DNA methylation and histone marks at multiple time points. These measurements will provide unique temporal datasets that will be used to design and implement computational methods to (i) identify genes regulated by NRs, (ii) infer the mechanisms of NR-triggered gene regulation, and (iii) identify pathways, biological processes and gene regulatory networks that the NR-responsive genes are involved in. SYNERGY is built upon interactive cycles between experimental (Henk Stunnenberg, Olli A. Jänne, George Reid) and modeling oriented (Sampsa Hautaniemi, Magnus Rattray, Antti Honkela, Genomatix Ltd.) groups. The models will be extensively validated during the project, and the predictions emerging from the models will be used to direct experiments that lead to more comprehensive understanding of gene regulation.

Acronym Synergy
Project Results
(after finalisation)
Systems approach to gene regulation biology through nuclear receptors (SYNERGY) consortium focused on comprehensive, quantitative and predictive understanding of nuclear receptor (NR) regulated gene regulatory networks in breast and prostate cancers. The overall objective was to derive experimentally validated, dynamic computational models that (i) describe the synergy of NRs and other TFs with modifications to chromatin, (ii) predict gene regulation and (iii) characterize downstream effects of the NR-regulated genes. SYNERGY has generated a massive amount of deep sequencing data, including the first time-course (10 time points) for a human cell line. All data generate in SYNERGY are freely available at A particularly important feature of SYNERGY was to find ways to connect the basic research findings to clinical data. The key findings of SYNERGY in this respect were 1) identification of an estrogen receptor responsive gene (ATAD3B) whose high activity is associated with decreased breast cancer patient survival (objective iii), 2) finding a new paradigm for a protein (FoxA1) related to the androgen receptor mediated gene regulation (objective i). Both findings were verified in large patient cohorts consisting of several hundreds of cancer patients. As estrogen receptor and androgen receptor are NRs that are the most important variables in clinical decision making in breast cancer and prostate cancer, respectively, SYNERGY is a prime example of the use of systems biology with biomedical data to achieve medically important results. The SYNERGY project led to the development of several new computational tools for the analysis and modelling of deep sequencing data. The key tools developed were (1) SPINLONG for finding ER responsive genes; (2) GROK for fast analysis and processing of deep sequence data; (3) The BitSeq tool for transcript-level deconvolution of RNA-Seq data and differential expression analysis across conditions; (4) Gaussian process methods for testing for temporal changes in high-throughput sequencing data and especially splicing using RNA-Seq data; (5) Gaussian process methods for modelling the dynamics of RNA-Pol II and for linking this dynamics to mRNA production (objective i) and (6) A novel Bayesian method for inferring the targets of multiple co-regulating transcription factor proteins from time course data (objective ii). This last method is being applied to the analysis of a new time course dataset in which certain protein-protein interactions have been perturbed in order to identify their downstream regulatory targets. There is an ongoing effort in follow-up projects to model the full dataset with highly coupled probabilistic latent variable models (objectives i and ii) to better characterise non-local enhancer-promoter interactions.
Network ERASysBio+
Call ERASysBio+-2008-01

Project partner

Number Name Role Country
1 University of Helsinki Coordinator Finland
2 University of Helsinki Partner Finland
3 Radboud University Nijmegen Partner Netherlands
4 Institute of Molecular Biology Partner Germany
5 Genomatix Software GmbH Partner Germany
6 University of Manchester Partner United Kingdom
7 Helsinki University of Technology Partner Finland