Title: Statistical model for identification of ChIP-enriched regions in epigenomic data

Abstract:

Eukaryotic genomes are organized into chromatin. Chromatin structure and function is regulated by a myriad of post-translational modifications on histone tails of the nucleosomes, the fundamental unit of chromatin. These modifications play critical role in transcriptional regulation, cell identity determination and maintenance, The genome-scale mapping of modified histones or histone variants has recently been fueled by the development of Chromatin immunoprecipitation followed by ultra-high throughput sequencing technology (ChIP-Seq). Many chromatin modification profiles are diffuse, calling for sensitive and efficient method to identify ChIP-enriched regions in the genome. I will present a method that aims to achieve this goal by detecting spatial clustering of signals. Using a genomic background model of random reads, this method provides precise formulas for accurate and efficient assessment of the statistical significance. I will then discuss applications of this method in statistical analysis of ChIP-Seq data and in extracting biological information.