Bioinformatics Seminar: Deciphering the contribution of cis-regulatory variants to gene expression patterns

Speaker

Dr. Joshua Rest
University of Chicago

Abstract

Gene expression is controlled combinatorially by transcription factors that bind to variable motifs in a gene's promoter. The role of motif variability in the cis-regulatory code and the resulting gene expression network has not heretofore been systematically studied. To explore this, I developed a computational framework integrating gene expression, binding site motif predictions, sequence and expression evolution, and functional gene annotations from yeast. A modular pipeline where intermediate results are stored in a relational database was used to explore different statistics and filters and to facilitate rapid reanalysis. Data was prepared using MySQL and Perl, analyzed using R and C, and visualized using R. Calculations were distributed over a cluster of computers in order to accommodate the large number of permutations required for determining significance. I show that functional variants of binding site motifs are associated with condition-specific differential expression, and that they play a combinatorial role in determining gene expression across different environmental and cellular conditions. The function of these variants is often conserved, and their evolution associated with change in gene expression. The power of binding site variants to predict gene expression is tested by experimentally switching between binding site variants through site-directed mutagenesis in yeast.