Postodctoral Research Fellowship in Biological Informatics FY2006

Siggers Trevor W, Medford MA

Investigators

Abstract

Project Title: "Using Protein-Binding Microarrays and Computational Methods to Investigate Affinity-dependent Mechanisms of Cis-regulatory Elements" This project is awarded under the Postdoctoral Research Fellowships in Biological Informatics Program for 2006. Transcription factors are proteins with affinities for short DNA sequences and by binding to specific occurrences of these sequences in the genome can turn genes on and off. The Fellow will conduct his research in the lab of Martha Bulyk at the Brigham and Woman's Hospital, Harvard University to develop a practical technology to measure the affinity of a protein for all possible DNA binding sequences in an unbiased manner. Protein-binding microarrays (PBM) are small glass slides to which short, double-stranded DNA molecules (~50 base pairs) are affixed in an array of tens of thousands of spots, each spot containing many copies of a single DNA sequence. Proteins are applied to the microarray and allowed to bind to the DNA in each spot. The amount of protein bound to each spot (i.e. sequence) is subsequently quantified using a fluorescently-labeled antibody applied to the microarray. Using a newly-developed, universal PBM developed in the lab, a technology will be developed to measure the absolute binding affinities of a protein to each possible 9 base pair DNA sequence in an unbiased fashion. The complete binding-affinity profiles of 40 yeast transcription factors to all possible 9 base pair sequences will be determined using the PBM technology. This large dataset will be used to investigate the accuracy of a commonly used matrix representation for describing transcription factor binding affinities, and to examine how the binding affinity of transcription factors to individual genomic sites affects the regulation of associated genes. Matrix scores will be compared with PBM-measured affinities to assess the accuracy of the scores, to examine assumptions underlying the matrix representation, and to examine possible accuracy biases for transcription factors from different protein structural classes. For the 40 yeast transcription factors, PBM-measured affinities will be analyzed for previously-identified DNA sequences that are bound upstream of ~800 genes in vivo to understand the role of binding affinities in gene regulation. The proposed research plan will provide the Fellow with valuable experience and training in studies (both computational and experimental) on control and specificity in gene regulation, which will form a foundation for his future independent career development.