Prediction of cystine connectivity using SVM

Abstract

Abstract:One of the major contributors to protein structures is the formation of disulphide bonds between selected pairs of cysteines at oxidized state. Prediction of such disulphide bridges from sequence is challenging given that the possible combination of cysteine pairs as the number of cysteines increases in a protein.Here, we describe a SVM (support vector machine) model for the prediction of cystine connectivity in a protein sequence with and without a priori knowledge on their bonding state. We make use of a new encoding scheme based on physico-chemical properties and statistical features (probability of occurrence of each amino acid residue in different secondary structure states along with PSI-blast profiles). We evaluate our method in SPX (an extended dataset of SP39 (swiss-prot 39) and SP41 (swiss-prot 41) with known disulphide information from PDB) dataset and compare our results with the recursive neural network model described for the same dataset.Keywords: disulphide bridges; prediction; protein fold; SVM model; SPX dataset Background:The completion of the human genome project shows a significant gap between the protein sequence and known structure space. Determination of protein structures using conventional X-ray crystallography and NMR (nuclear magnetic resonance) techniques is not adequate to cover the sequence space in the context of drug discovery. Hence, protein structure prediction using computational methods is becoming critical. However, prediction of protein tertiary structure from sequence is non-trivial and is generally achieved by dividing the problem into finite levels of secondary structures and super secondary structures.

Soft Computing Methods for Disulfide Connectivity Prediction

“…This vector includes the Euclidean distance between all the oxidized cysteines of the training proteins. Finally, Shilton et al 8 elaborate an encoding scheme based on physicochemical properties and statistical features. These properties are hydrophobicity and polarity according to the scales described by Kyte and Doolitle 21 and Grantham, 22 respectively.…”

“…Properties used in the literature are hydrophobicity, polarity, volume of residues, graph shape index, and isoelectric point, among others. Shilton et al 8 and Song et al 9 include amino acid properties as input data.…”

Prediction of disulfide connectivity in proteins with support vector machine

“…The results showed that the distribution of cysteine residues in a protein sequence provides valuable information for predicting the disulfide connectivity pattern more precisely. Previously, another machine learning technique, support vector machine, was employed to predict the disulfide connectivity pattern of proteins (Rama et al, 2005;Tsai et al, 2005). In this paper, we also adopted SVM to predict the disulfide connectivity pattern of proteins.…”

“…Different machine learning techniques such as NN Vullo and Frasconi, 2004) and SVM (Rama et al, 2005;Tsai et al, 2005) have been used to solve this problem. Previously, Fariselli and Casadio (2001) have transformed this problem into an undirected graph, where vertices represent the bonded cysteine residues and weights of edges are the bonding probabilities for the corresponding cysteine pairs.…”

The Disulfide Connectivity Prediction with Support Vector Machine and Behavior Knowledge Space

“…Ferre and P. Clote, 2005;Pier Luigi Martelli et al, 2002;Alessandro Vullo and Paolo Frasconi, 2004;Castrense Savojardo et al, 2013) and support vector machine (SVM) (Yu-Ching Chen et al, 2004;Yu-Ching Chen and Jenn-Kang Hwang, 2005;P. Frasconi et al, 2002;Jayavardhana Rama G. L. et al, 2005;Hsuan-Liang Liu and ShihChieh Chen, 2007;Chih-Hao Lu et al, 2007;ChiHung Tsai et al, 2005;Marc Vincent et al, 2008).…”