Analyzing ChIP-seq Data based on Multiple Knowledge Sources for Histone Modification

摘要

ChIP-seq is able to capture the genomic profilesfor histone modification by combining chromatinimmunoprecipitation (ChIP) with next generationsequencing. However, enriched regions generated from peakfinding algorithms are evaluated only based on the limitedknowledge acquired from manually examining the relevantbiological literature. This paper proposes a novelframework of incorporating multiple knowledge sources,consisting of information extracted from biologicalliterature, Gene Ontology, and microarray data, in order toprecisely analyze ChIP-seq data for histone modification.The information is combined in a unified probabilisticmodel to rerank the enriched regions generated from peakfinding algorithms. Through filtering the reranked enrichedregions using some predefined threshold, more reliable andprecise results could be generated. The combination of themultiple knowledge sources with the peaking findingalgorithm produces a new paradigm for ChIP-seq dataanalysis.