Prediction of functional nonsynonymous single nucleotide polymorphisms in human G-protein-coupled receptors

G-protein-coupled receptors (GPCRs) are found in a wide range of organisms and are central to a cellular signaling network that regulates many basic physiological processes. GPCRs are the focus of a significant amount of current pharmaceutical research because they play a key role in many diseases. In this paper, we predict the functional nonsynonymous single nucleotide polymorphisms (nsSNPs) in human GPCRs by defining optimal attributes and using a decision tree method. The predictive power of each attribute was evaluated. A subset of sequences with optimal attributes was obtained using the decision tree method combined with a genetic search algorithm. The subset contains both sequence-based and structure-based information, and the information for each subset consists of a conservation score, the location of the mutation, the BLOSUM62 substitution matrix score, as well as the hydrophobicity change, the solvent accessibility, and the buried charge. Seven important rules were derived from the decision tree. A total of 166 functional nsSNPs in human GPCRs from the dbSNP have been predicted using the optimal attributes subset.