Predicting Drug-Target Interaction Networks Based on Functional Groups and Biological Features

Abstract

BackgroundStudy of drug-target interaction networks is an important topic for drug development. It is both time-consuming and costly to determine compound-protein interactions or potential drug-target interactions by experiments alone. As a complement, the in silico prediction methods can provide us with very useful information in a timely manner.Methods/Principal FindingsTo realize this, drug compounds are encoded with functional groups and proteins encoded by biological features including biochemical and physicochemical properties. The optimal feature selection procedures are adopted by means of the mRMR (Maximum Relevance Minimum Redundancy) method. Instead of classifying the proteins as a whole family, target proteins are divided into four groups: enzymes, ion channels, G-protein- coupled receptors and nuclear receptors. Thus, four independent predictors are established using the Nearest Neighbor algorithm as their operation engine, with each to predict the interactions between drugs and one of the four protein groups. As a result, the overall success rates by the jackknife cross-validation tests achieved with the four predictors are 85.48%, 80.78%, 78.49%, and 85.66%, respectively.Conclusion/SignificanceOur results indicate that the network prediction system thus established is quite promising and encouraging.

Proteome-wide Prediction of Self-interacting Proteins Based on Multiple Properties

“…Here, mRMR method is used for feature selection (26,27 where p(x,y) is the joint probabilistic distribution function, and p(x) and p(y) are the respective marginal probability distribution functions (26). MI quantifies the mutual dependence between two random variables, i.e.…”

High performance set of PseAAC and sequence based descriptors for protein classification

“…Therefore, the jackknife test has been increasingly and widely adopted by investigators to test the power of various prediction methods (see, e.g. (Chen et al, 2009;Chou and Shen, 2007b;Chou and Shen, 2008b;Ding and Zhang, 2008;Esmaeili et al, 2010;He et al, 2010;Jiang et al, 2008;Lin, 2008;Lin et al, 2008;Qiu et al, 2009;Zeng et al, 2009;Zhou, 1998;Zhou et al, 2007)). However, to reduce the computational time, we adopted the 2-fold cross-validation in this study as done by many investigators with SVM as the prediction engine.…”

Molecular biocoding of insulin – amino acid Ser

“…They reveal to us that the process of sequencing in bio-macromolecules is conditioned and determined not only through biochemical, but also through cybernetic and information principles. Many studies have indicated that analysis of protein sequence codes and various sequencebased prediction approaches, such as predicting drug-target interaction networks [14], predicting functions of proteins [15,18], analysis and prediction of the metabolic stability of proteins [16], predicting the network of substrate-enzymeproduct triads [7], membrane protein type prediction [1,2,5]. protein structural class prediction [4,12], protein secondary structure prediction [6,11], enzyme family class prediction [3,11], identifying cyclin proteins [20], protein subcellular location prediction [9,10,17,19] , among many others as summarized in a recent review [15] , can timely provide very useful information and insights for both basic research and drug design and hence are widely welcome by science community.…”