Poster Sessions

Bioinfo-8

Sarosh Fatakia

S. N. Fatakia, S. Costanzi, C. C. Chow

Computing highly correlated positions in mutual information space for G protein coupled-receptors: further evidence of a ligand binding cavity

G protein-coupled receptors (GPCRs) are a superfamily of seven transmembrane-spanning proteins involved in a wide array of physiological functions and are the most common targets of pharmaceuticals. Using a multiple sequence alignment of the transmembrane (TM) domains, we calculated the mutual information between all inter-TM pairs of aligned positions and ranked the pairs by mutual information. Invoking the statistically significant top ranked pairs of positions, a mutual information graph was constructed whose vertices corresponded to TM positions and pairs with statistically significant mutual information represented its edges. Positions with statistically significant high degree (i.e. had significant mutual information with a large number of other positions) were found to line a well defined inter-helical binding cavity for class A as well as class C GPCRs, for which the natural ligands bind to their extracellular N-terminal domains. Such positions were not found for class B GPCRs confirming the observation that ligands that bind to an inter-helical binding cavity are not known to exist for class B. Our algorithm may be helpful for localizing topologically conserved binding cavities in other protein families.