论文部分内容阅读
The Encyclopedia of DNA Elements (ENCODE) Consortium has generated hundreds of high throughput genomic datasets with the goal of cataloging functional elements in the human genome.Our goal was to integrate these complex data types to annotate regulatory elements and their target genes.We then used these annotations to functionally characterize non-coding variants associated with human disease.We began by developing an unsupervised method for predicting enhancer-like regions,which we evaluated using experimentally validated data from transgenic mouse assays.Using only H3K27ac ChIP-seq and DNase-seq data,our method shows high concordance with more complex genome segmentation methods with the advantage that it can be applied across many more cell and tissue types.