NSiteMatch: Prediction of Binding Sites of Nucleotides by Identifying the Structure Similarity of Local Surface Patches.

In this study, we propose a nucleotide-binding site predictor, namely, NSiteMatch. The NSiteMatch algorithm integrates three different strategies: geometrical analysis, energy calculation, and template comparison. Unlike a traditional template-based predictor, which identifies global similarity between target structure and template, NSiteMatch concerns the local similarity between a surface patch of the target protein and the binding sites of template. To this end, NSiteMatch identifies more templates than traditional template-based predictors. The NSiteMatch predictor is compared with three representative methods, Findsite, Q-SiteFinder, and MetaPocket. An extensive evaluation demonstrates that NSiteMatch achieves higher success rates than Findsite, Q-SiteFinder, and MetaPocket, in prediction of binding sites of ATP, ADP, and AMP.
PMID: 28811833 [PubMed - in process]

Source: Computational and Mathematical Methods in Medicine - August 18, 2017 Category: Statistics Tags: Comput Math Methods Med Source Type: research