Epigenomic research has become one the fastest evolving fields in molecular biology. Epigenetic effects control the packaging of DNA in the nucleus thereby deeply influencing gene expression. They also play crucial roles in cell differentiation and aberrant patterns are associated with cancer, mental disorders and autoimmune diseases. However, our understanding of the epigenomic code is still limited. A main obstacle to its decoding is the requirement for immensely data-intense experiments, as epigenomic configurations embrace multiple different marks which form an intricate interplay that varies between cell types. Advances in high-throughput sequencing resulted in a plethora of complex data sets and computational methods are called upon to solve pressing questions for their analysis and modeling. In this project, we will develop machine learning tools to combine epigenomic measurements with computational sequence analysis. This will provide us with a better understanding of the extent to which DNA sequence controls the establishment of epigenomic marks. It will also serve as a credible basis for data integration. We will next combine data from different cell lines and analyze them simultaneously. In particular, we will examine the effect of a certain transfactor, by comparing histone marks in wild type ES cells with mutants that lack the DNA-binding protein Cfp-1, which is known to play a role in the formation of epigenomic marks at active promoters. To shed light on the epigenomic impact on the transcriptome the framework will eventually be complemented by adding expression data. The proposed research will doubly benefit the applicant by introducing her to a new field of application, as well as a wider class of computational techniques. We believe this work to be of scientific importance, as the employed machine learning approaches are likely to lead to new insights in epigenome research with immense potential consequences in addressing key biomedical issues.