Application of high quality amino acid indices to AMS 3.0: A update note

Abstract

In this article, we are showing the application of high quality indices of amino acids to improve the performance of AutoMotif Server (AMS) for prediction of phosphorylation sites in proteins. The latest version of AMS 3.0 is developed using artificial neural network (ANN) method. The query protein sequence is dissected into overlapping short sequence segments and then represented it by ten different amino acid indices, which are various physicochemical and biochemical properties of amino acids. However, the selection of amino acid indices has done based on literature survey. Hence, this fact motivated us to use the recently proposed high quality amino acid indices for AMS 3.0. High quality amino acid indices have been developed after analyzing the AAindex database using fuzzy clustering methods. The significant differences in the performance are observed by boosting the precision and recall values of four major protein kinase families like CDK, CK2, PKA and PKC in comparison with the currently available state-of-the-art methods. © 2013 Springer.