DF/HCC Program Affiliation

Biostatistics

Research Abstract

My lab studies transcriptional regulatory networks, both in model organisms and in the human genome. We are developing and applying computational methods to identify likely DNA regulatory elements in mammalian genomes. We are also developing & applying new microarray technologies for identification of TF binding sites. The study of regulatory regions in higher eukaryotes is difficult in metazoan organisms, in part because roughly 99% of the human genome is noncoding. In addition, even general themes regarding the genomic locations of DNA regulatory elements are still unknown. Adding to the difficulty, regulatory elements have been found far upstream of coding regions, within introns, and even downstream of the genes they regulate.

To permit a genome-wide scan of all possible TF binding sites in a given genome, my group has developed a new in vitro DNA microarray-based technology (“protein binding microarrays (PBMs)”) that allows high-throughput characterization of the DNA binding site sequence specificities of proteins in a single day. We are currently applying this technology to study human tissue-specific TFs, with a focus on TFs important in skeletal muscle gene expression.

One of our computational searches for regulatory elements is focused on conserved noncoding sequences surrounding mouse and human orthologs with similar tissue-specific mRNA expression patterns or related functional annotation, as these genes are most likely to be regulated by orthologous TFs. We are currently applying this technology to identify candidate human tissue-specific motifs.

A second computational approach for identifying potential DNA regulatory elements is to focus on identifying over-represented clusterings of TF binding sites, as regulatory complexity in higher eukaryotes is achieved in part through combinatorial interactions between TFs. We have developed software (“ModuleFinder”) that identifies statistically significant clusterings of TF binding sites that are conserved across related genomes. Using ModuleFinder, we have predicted novel tissue-specific transcriptional enhancers in fly and also in human. Using another new algorithm we have developed ("CodeFinder"), we have systematically examined a large set of Boolean combinations of TF binding sites to ascertain each TF’s relative contribution to and specificity for a given transcriptional regulatory model.

The results of our experiments and analyses will provide valuable data on cellular regulatory networks, including the construction of a more detailed connectivity map of the human transcriptional regulatory network. Novel cis regulatory elements and corresponding TFs will be identified. Our studies will permit a better understanding of the locations and organization of regulatory elements in higher eukaryotic genomes, and will aid in understanding the regulatory complexity resulting from combinatorial interactions of TFs.

Publications

• Philippakis AA, He FS, Bulyk ML.Modulefinder: a tool for computational discovery of cis regulatory modules.Nat Biotechnol 2005:519-30.
  15759656
• Mukherjee S, Berger MF, Jona G, Wang XS, Muzzey D, Snyder M, Young RA, Bulyk ML.Rapid analysis of the DNA-binding specificities of transcription factors with DNA microarrays.Nat Genet 2004 Dec;36(12):1331-9.
  15543148
• Bulyk ML, McGuire AM, Masuda N, Church GM.A motif co-occurrence approach for genome-wide prediction of transcription-factor-binding sites in Escherichia coli.Genome Res 2004 Feb;14(2):201-8.
  14762058
• Bulyk ML, Huang X, Choo Y, Church GM.Exploring the DNA-binding specificities of zinc fingers with DNA microarrays.Proc Natl Acad Sci U S A 2001 Jun 19;98(13):7158-63.
  11404456