Research Areas - Computational Epigenetics
We perform research at the interface of bioinformatics and molecular genetics, developing and appling tailored computational methods in order to address questions in the context of mammalian epigenetics and epigenomes.
Epigenetics is an umbrella term for heritable gene regulation that is not directly encoded in the DNA sequence. Because of its clinical relevance in cancer, inflammation and mental disorders, epigenetic mechanisms have recently received a boost of attention, and several large human epigenome projects are now underway. Starting to work on computational epigenetics in 2004, we were among the first bioinformaticians to address epigenetic questions by computational methods. By means of advanced bioinformatic methods such as statistical learning and prediction, we could show that epigenetic regulation is deeply rooted in specific properties of the genomic DNA sequence (Bock et al. 2006; Bock et al. 2007; Bock et al. 2008), a result that was a surprise for many molecular biologists. We also developed several widely used software tools that faciliate epigenetic and epigenomic research, such as BiQ Analyzer, MethMarker and EpiGRAPH. These and other topics are also summarized in a recent review paper on computational epigenetics that we wrote for the Bioinformatics journal (Bock and Lengauer 2008).
Our current work focuses on three areas. First, we developed and continue to extend an epigenome prediction software (http://epigraph.mpi-inf.mpg.de), which brings the use of advanced statistical and machine learning methods within reach of epigeneticists with little bioinformatic experience. Second, we contribute bioinformatic analysis and prediction know-how to an EU project on cancer epigenetics (http://www.cancerdip.eu/), and we develop a comprehensive bioinformatic pipeline for discovery and optimization of epigenetic cancer biomarkers (Mikeska et al. 2007, Viré et al. submitted, Brenner et al. submitted, Schüffler et al. in preparation). Third, we develop methods and perform analyses that compare epigenetic modifications in different species and tissues, in order to understand the origins and molecular function of epigenetic regulation in mammals.
Our long-term goals are to develop advanced bioinformatic methods that help bench researchers make sense of (epi-) genome datasets, and to use these methods in collaboration with biomedical and clinical researchers to bring epigenetic diagnostics and treatment for diseases such as cancer, chronic inflammation or mental disorders into the clinic.