My research focuses on the design of statistical and computational algorithms to discover and explore the genomic sequence elements related to gene regulations. We are especially interested in Chromatin ImmunoPrecipitation coupled with DNA microarray analysis (ChIP-chip) to study the effect of transcription factor binding on transcription regulation.

ChIP-chip studies the in vivo transcription factor binding locations in the genome. Recent commercial high density oligonucleotide arrays that tile all the non-repetitive human genomic sequences allow biologists to conduct unbiased genome-wide ChIP-chip experiments. However, they also generate massive amounts of data. We design algorithms to identify genomic regions bound by the transcription factor in ChIP-chip on Affymetrix tiled arrays. Our method relies on estimating the array probe behavior by considering probe sequence and copy number in the genome. This method can detect transcription factor binding from a single ChIP-chip experiment without using mismatch probes, control experiments, or replicated samples. This is a huge time and money saver for labs testing their ChIP-chip antibodies and protocols, and finding failed replicates contaminating their data.

Once the transcription factor binding locations are determined, we develop a publicly available web server that allows biologists to download genomic sequences, mask sequence repeats, generate sequence conservation plots, map nearby genes, and look for enriched sequence motifs that are bound by the transcription factor and its cooperative binding partners. In the past few years, we have excelled in computational sequence motif finding and has developed a number of widely-used motif-finding algorithms for different biological applications. For the current study, we integrate ChIP-chip, gene expression microarray, and sequence motif finding to predict the genes regulated by the transcription factor. Our final and most challenging goal in this study is to generalize the global combinatorial regulatory mechanisms of the transcription factor. For this we plan to also incorporate genome-wide high-throughput epigenetic information, such as histone modification, DNA methylation, and nucleosome localization information. We are currently collaborating with many Harvard laboratories to use ChIP-chip to study the regulation of key cancer transcription factors.

In addition, we are interested in bacteria operon prediction, alternative splicing, non-coding RNA regulation.

Ph.D., 2002, Stanford University