Project Summary: The large majority of genetic variations (GVs) occur in non-coding regions particularly in intergenic regions. Understanding the functions of the GVs and revealing their regulatory impact on gene expression remain a great challenge because it is not trivial to link GVs to their target genes and consider collaborative effect of individual GVs. The availability of large amount of the ENCODE and Roadmap Epigenomics Project data provides an unprecedented opportunity to tackle these challenges. We will develop new computational methods to predict long-range promoter-enhancer interactions from epigenomic data. These predicted promoter-enhancer interactions will be integrated with the other ENCODE and Roadmap Epigenomics Project data including histone modification, ChIP-seq of DNA binding proteins, RNA-seq and open chromatin data to construct genetic networks that represent cell-type specific regulatory interactions. These networks will be used to annotate the GVs identified in patient samples to reveal disease-related GVs. Once completed, the proposed study will provide a suite of new computational methods for integrative analysis of the ENCODE/Epigenome Roadmap data and establish a resource of the digested ENCODE/Roadmap Epigenomics Project data. The proposed computational framework is general and can be easily applied to other public data.