Conotoxin Superfamily Prediction Based on Diffusion Maps and dHKNN

Conotoxins show prospects for being potent pharmaceuticals in the treatment of some serious disease. Accurate prediction of conotoxin superfamily would have many important applications in biological research and clinical medicine. In this study, we propose a novel dHKNN method to predict conotoxin superfamily. Firstly, we extract the protein's sequential features composed of physicochemical properties, evolutionary information, predicted secondary structures and amino acid composition. Then we use the diffusion maps for dimensionality reduction.At last, with considering the local density information in the diffusion space, the dHKNN is proposed based on the K-local hyperplane distance nearest neighbor subspace classifier method for predicting conotoxin superfamilies. An overall accuracy of 91.90% is obtained through the jackknife cross-validation test which is higher than present methods.