|Quanwei Zhang|| at 11:00
Affiliation: Northwestern University
Host: Ivan Ovcharenko
Transcription factor binding sites identification and histone methylation analyses with ChIP-seq data
Genome-wide mapping of transcription factors is essential for a full understanding of gene regulation. Besides, nucleosome positioning and its dynamic modification also play a key role in gene regulation. Owning to the rapid progress in next-generation sequencing technology, ChIP-seq (chromatin immunoprecipitation followed by sequencing) data offer higher resolution, less noise and greater coverage compared with its array-based predecessor ChIP-chip data. In this talk I will present the basic ChIP-seq data process steps and some methods used in our ongoing projects.
For the transcription factors, it is crucial to identify their candidate binding sites. But it is a great challenge to find the true sites from background noise, especially for the single-end reads. We proposed a peak detection algorithm, which effectively makes use of the tag distribution information on both strands. The model can not only keep the tags contribute to the candidate binding sites, but also decrease the chance to involve in background tags. By comparison with some recent algorithms, our methods showed higher detection sensitivity.
Unlike transcription factor binding sites, which are sparse and narrow, histone methylation regions are much wider and dense. So the analysis methods were somewhat different with those used for transcription factors. We have a project, in which we are trying to analyze how MMSET (Multiple Myeloma SET domain) regulate gene expression through changes in histone methylation. Here I will present the methods we used and some results from this project.