A new multi-level thresholding algorithm for finding peaks in ChIP-Seq data

Combination of Chromatin immunoprecipitation (ChIP) and high-throughput sequencing (Seq) is one of the most effective ways for studying the genome-wide binding sites, which helps to understand protein-DNA interactions on a genomic-wide scale. Several algorithms have been proposed to take advantage of ChIP-Seq abilities. However, despite many elaborations, it is still difficult to accurately determine individual peaks corresponding to each binding event. We developed a constrained multi-level thresholding algorithm which has the ability of targeting regions within a user-defined range. We show that out proposed method has higher accuracy in detecting enriched regions by assessing its performance relative to other previously proposed peak finders.