Computational Regulatory Genomics

Development of complex organisms critically depends of the regulation of gene activity across dozens, hundreds or millions of cells. This regulation ensures the fidelity of the development of complex structures and overall function of the organism. Its disruption often leads to disease. We investigate how regulatory information is encoded in the DNA, and how it is communicated across cells in multicellular organisms. The main objects of our research are:
• The structure and function of gene promoters - parts of DNA at the beginning ofeach gene that bind molecular machinery that transcribes DNA into RNA;
• The function and genomic distribution of gene regulatory elements - parts of DNA at varying distances from promoters that affect the rate/level of transcription;
• The function of transcription factors - proteins that bind to specific motifs in the regulatory elements and this way affect the activity of genes;
• Transcription factor proteins are themselves products of gene transcription and translation, and the transcription of genes that encode transcription factors is regulated by other transcription factors and often by that factor itself. The complex regulatory interrelationships of this kind are known as transcriptional regulatory networks (TRNs);
• The association of different modes of regulation with epigenetic marks - chemicalmodifications of DNA or protein complexes it is packaged with in chromosomes - and their inheritance.

The central interests of the Computational Regulatory Genomics group are genomics and epigenomics of gene regulation in higher eukaryotes. We discovered several fundamental features of genes under developmental regulation and established the link between the type of core promoter and its responsiveness to long-range regulation, and developed a number of widely used computational methods and data resources for studying the regulatory content of genomes. Our current aim is to understand the molecular and mechanistic basis for different modes of gene regulation associated with the control of cell cycle, multicellular processes and regulation in terminally differentiated cells. In addition to our models of long-range regulation, the newly available epigenomic data contains crucial information for distinguishing these modes in development, differentiation and disease.
Our core research program is to identify the features of genes under different modesand regimes of gene regulation. Different genes have regulatory territories that vastly differ in size, are often nested, and exhibit distinct patterns of epigenetic modifications, or succession of such patterns. By investigating genomic regulatory territories, we want to establish sequence features and epigenomic patterns and sequence features associated with different modes of gene regulation. We want to find out which properties of genes, are responsible for the correct sorting and assignment of inputs in complex, nonlinear gene loci. Among the most exciting hypotheses we are currently investigating are the fundamentally different modes of regulation of genes in dividing vs non-dividing, terminally differentiated cells, and time-sharing of core promoters. We plan to investigate how complex input from the cell environment is interpreted by megabase-sized cis-regulatory arrays of these genes to result in tightly coordinated development of complex multicellular structures.
Our primary focus is biological insight. Nonetheless, we develop reusable methods for our analysis, to be shared by the community. This has proven to be a highly productive approach for subsequent analyses. In the forthcoming period our method development will focus on i) Transcription factor binding and transcription factor complexes: We are developing a computational toolbox for the analysis and visualisation of large sets of transcription factor binding sites and other regulatory elements and their integration with other types of genome-wide data; ii) Computational methodology for comparative epigenomics – The next several years will witness the rise of the need to analyse hundreds or thousands of epigenomes simultaneously - across tissues, time courses, epidemiological cohorts or different species. Computational capacity expertise and tools for the analysis of this scale are currently largely unavailable. Building on and inspired by methods for the segmentation of epigenomic regions, we plan to build methods for the characterisation of individual regulatory elements and their ensembles around genes under complex regulation.