A Streamlined and Generalized Analysis of Chromatin ImmunoPrecipitation Paired-End diTag Data

Comprehensive, accurate and detailed maps of transcription factor binding sites (TFBS) help to unravel the transcriptional regulatory relationship between genes and transcription factors. The recently developed sequencing-based genome-wide approach ChIP-PET (Chromatin ImmunoPrecipitation coupled with Paired-End diTag analysis) permits accurate and unbiased mapping of TF-DNA interactions. In this paper we outline a methodical framework to analyze ChIP-PET sequence data to identify most likely binding regions. Mathematical formulations were derived to streamline and strengthen the analysis. We established a more faithful noise distribution estimation that leads to the adaptive threshold scheme. The algorithms were evaluated using three real-world datasets. Using motif enrichment as indirect evidence and additional ChIP-qPCR validations, the overall performance was consistently satisfactory.