Promoters located at the 5'-ends of genes play a critical role in regulating transcriptional initiation. Emerging evidence suggests that a significant fraction of approximately 30,000 human genes likely contain alternative promoters, which produce more elaborate regulation of gene expression in different tissues, cell-types and/or developmental stages. Despite vast information available for the human genome sequences, a comprehensive approach for identifying and characterizing alternative promoters of gene loci is still lacking. One effective approach is to analyze orthologous sequences in both mouse and human genomes. It has not been thoroughly explored as to what extent the 5'-end regulatory regions of a locus show sequence similarity between human and rodents. We hypothesize that the basal (core) promoter regions necessary for gene transcription are conserved between human and mouse. Computational approaches will be used to mine both human and mouse genomes to identify functional orthologous sequences. The derived information will be added into a prototype database for mammalian promoters, called MPromDb, developed by us. Next, these computationally derived sequences will be experimentally verified. We further hypothesize that a functional genomic sequence is in a euchromatic environment that presents an open chromatin configuration, allowing for the access of transcription factors for driving gene expression. Instead of taking the usual route to analyze gene expression, we will determine the chromatin status of a test genomic sequence as a measure for a functional promoter. The chromatin immunoprecipitation (ChIP) microarray, or the so-called ChlP-on-chip assay, previously developed in our laboratory will be extended to simultaneously assess the euchromatin status of approximately 1,500 putative promoters of 250 pairs of human and mouse orthologous genes. The combination of computational, statistical and highthroughput experimental approaches proposed in this grant application will help better characterize the gene regulatory regions in the human genome, one of the grand challenges of future genome research. Specifically we will: (1) Develop computational tools for annotating experimentally known alternative promoters and first exons. (2) Conduct ChlP-on-chip and luciferase assays to verify computationally annotated alternative promoter sequences of human and mouse orthologous genes, and (3) Develop computational methods to detect alternative promoters and first exons in the human and mouse genomes.
