Identification of transcriptional modules in budding yeast

One focus of postgenomic research concerns the regulation of gene expression through signal transduction and combinatorial control of gene transcription. Protein-protein interactions and protein-DNA interactions provide the dynamic backbones of regulatory networks which define the complex response of a cell to different environmental conditions. More recently, large scale experiments have resulted in a plentitude of interaction data for model organisms. Genome-wide location analysis experiments [6, 4] have generated a comprehensive map of transcription factor binding sites and in high-throughput assays researchers have also identified binary and complex interactions in budding yeast [7, 3, 1, 2]. While such genome-wide data suffers from both false-positive and false-negative error rates, it provides valuable information to guide and direct future research that will have to validate these putative findings. In this work we investigate the correlation of protein-DNA binding data with experimental findings for protein-protein interactions. We identify modules of co-regulating transcription factors and find that synergistic transcription factor pairs interact preferentially. Furthermore, we use sequence data to give predictions for binding sites which were not yet identified in experiments.