Accurate sequence-based prediction of catalytic residues

Abstract

Our method (CRpred) uses sequence-based features and the sequence-derived PSI-BLAST profile. We used feature selection to reduce the dimensionality of the input (and explain the input) to support vector machine (SVM) classifier that provides predictions. Tests on eight datasets and side-by-side comparison with six modern structure- and sequence-based predictors show that CRpred provides predictions with quality comparable to current structure-based methods and better than sequence-based methods. The proposed method obtains 15-19% precision and 48-58% TP (true positive) rate, depending on the dataset used. CRpred also provides confidence values that allow selecting a subset of predictions with higher precision. The improved quality is due to newly designed features and careful parameterization of the SVM. The features incorporate amino acids characterized by the highest and the lowest propensities to constitute catalytic residues, Gly that provides flexibility for catalytic sites and sequence motifs characteristic to certain catalytic reactions. Our features indicate that catalytic residues are on average more conserved when compared with the general population of residues and that highly conserved amino acids characterized by high catalytic propensity are likely to form catalytic sites. We also show that local (with respect to the sequence) hydrophobicity contributes towards the prediction.

iCataly-PseAAC: Identification of Enzymes Catalytic Sites Using Sequence Evolution Information with Grey Model GM (2,1)

“…Meanwhile, we can see that the To further approve its power, let's compare iCatalyPseAAC with the existing predictor in this area. Predictor CRpred (Zhang et al 2008) is based on sequence predictor, method EXIA (Chien and Huang 2012) is based on the residue side chain structure and sequence conservation method. Because our predictor is only one with a publicly accessible web server, the best way to compare them is though practical application.…”

“…To promote the biological community, a web server of iCataly-PseAAC was constructed, which can be freely accessible at http://www.jcibioinfo.cn/iCataly-PseAAC. (Zhang et al 2008) b From (Chien and Huang 2012) c From (Petrova and Wu 2006;Youn et al 2007) …”

“…Protein Data Bank (PDB) also provides the catalytic sites annotation for enzymes (Berman et al 2002;Berman et al 2000). The sequence information (Dou et al 2011;Ota et al 2003;Zhang et al 2008;Dou et al 2010;Chea and Livesay 2007;Fischer et al 2008)and the structure information (Bartlett et al 2002;Torrance et al 2005;Zvelebil and Sternberg 1988) have been used in predicting catalytic sites, respectively, or together. Chien et al developed a structure-based method based on residue side chain orientations and backbone flexibility of enzyme structure (2012).…”

Analysis and Prediction of RNA-Binding Residues Using Sequence, Evolutionary Conservation, and Predicted Secondary Structure and Solvent Accessibility

“…Similarly as in [63], PSSM and EntWOP features were obtained from the PSSM and WOP vectors and were computed using logistic function f ij (a ij )=1/(1+exp(-a ij )) and entropy estimate EntWOP i = There are 15 20 = 300 PSSM and 15 EntWOP features for each predicted residue. In Kim's work [21], structurederived information concerning adjacent residues was used to predict RBRs, which resulted in an improved prediction quality.…”

“…The motivation behind the choice of the SVM comes from wide-spread applications of SVM in various bioinformatics problems, such as prediction of secondary structure [93,94], catalytic residues [63], subcellular localization [95,96], protein-protein interaction site [97], and the successful application in the existing method for RBR predictions [7,11,[22][23][24]. We note that four most recent sequence-based predictors of the RNA-binding residues are based on the SVM classifier, see Table 1.…”