Grant: Operating Grant

There are hundreds of different types of cells in our body but almost all of them have an identical set of genes. The reason that heart cells are different from liver cells is differences in gene activity. A better understanding of the regulation of gene activity would not only improve our ability to predict why certain mutations cause cancer but also improve our ability to fight disease-causing human parasites. The general principle of gene regulation is that regulatory proteins recognize codes encoded in the gene, or nearby the gene, in the genome. By knowing what these regulatory codes are, and which proteins recognize these codes, we can more easily study gene regulation. Recently it has become clear that most gene regulation activity occurs post-transcriptionally, i.e., after the gene has been copied from DNA (where it is stored in our genome) to messenger RNA. This means that many important regulatory codes are encoded in RNA sequence and are being recognized by RNA-binding proteins. Although it is often possible recognize whether or not a protein will bind an RNA code based on its amino acid sequence, it has proven difficult to tell exactly what that RNA code is. So, although sequencing an organism's genome will reveal what RNA-binding proteins it has, it won't tell us what codes these proteins recognize. Our specific aim is solve this problem. We propose to use high volume biochemical methods to determine the RNA binding preferences of hundreds of carefully selected RNA-binding proteins from a diverse set of organisms. Based on these measured binding preferences, we then propose to use advanced computational methods to design an algorithm that will predict the binding preferences for arbitrary protein sequences. We will validate this algorithm by testing some of its predictions.