Study Of The Functional Consequences Of Single Amino-acid Substitution In Human Cytochrome P450

The cytochrome P450 superfamily (CYP) is a large and diverse group of powerful enzymes containing the heme group. In the human body, they perform many key functions including the synthesis of endogenous substances and the metabolism of exogenous chemicals. As CYP enzymes can metabolize approximately 90% of clinically used drugs, the polymorphisms of CYP genes have been considered as a dominant causative factor for the inter-individual differences in drug metabolism. Therefore, the prediction of functional consequences of nonsynonymous SNP which causes single amino-acid substitution in human cytochrome P450 is of great importance to the understanding of what causes different drug metabolism capabilities among individuals and may be helpful in developing personalized drugs for different individuals.In this study, we use support vector machine (SVM) combined with genetic algorithm to construct the predictive model whereby we can predict the functional consequences of nsSNP of nine major human drug-metabolizing CYPs with PDB structure including CYP1A2, CYP2A6, CYP2A13, CYP2B6, CYP2C8, CYP2C9, CYP2D6, CYP2E1 and CYP3A4 based on both the structural and physicochemical properties of the enzyme and those of the substrate. In addition, we developed a strategy to calculate two kinds of protein descriptors using the amino-acid indices. Compared with the amino-acid indices, the protein descriptors can fully consider the change caused by the substitution of single amino acid. After multi-round feature selection, five protein descriptors and eight substrate descriptors were selected to construct the final model. The validation results exhibited that our model achieved the predictive accuracy of 88.1%, sensitivity of 93.1% and specificity of 72.7%. The main improvement of our model over existing ones is the relatively high accuracy and the consideration of the structural information and substrate's properties into the prediction. As a result, our model can facilitate the functional analysis of nsSNPs. Before conducting experiment, we can get a preliminary idea whether an unknown single amino-acid substitution causes functional changes in human CYPs.