A Support Vector Machine Based Model for Predicting Heparin-Binding Proteins Using XB Patterns as Features

摘要

Heparin is a highly sulphated and negatively charged polysaccharides belonging to the glycosamino-glycans (GAGs) family, used in medical treatments as an anticoagulant. Although many heparin-binding proteins have been identified, there are still many proteins needing to be classified as heparin-binding or not. Many studies have been aimed at prediction of heparin binding patterns in the primary structure of proteins, however, still no model has emerged which reasonably predicts proteins as heparin-binding or not. The main objective of this study is to predict heparin-binding proteins from their amino acid sequence information. A supervised learning algorithm based on support vector machine (SVM) is applied to two data sets; one training set used to create the model and one testing set used to validate and test accuracy of the model. For the testing set, the model achieves 75.36% accuracy in predicting heparin-binding proteins. The current model uses 66 XB patterns as features.