Prediction of Interface Residues in Protein?Protein Hetero Complexes?

Many data mining technique have been proposed for fulfilling various knowledge discover task in order to achieve the goal of retrieving useful information for user. Sequence based protein is understanding and identification of protein binding interface is a challenging task, in this protein system imbalanced distribution between positive sample (interface) and negative sample (non-interface).This paper proposed method that can be predict protein interaction sites in hereto-complexes. Mahalanobis distance measure (namely, a pseudo distance) termed as locally centered Mahalanobis distance, derived by centering the covariance matrix at each data sample rather than at the data centroid as in the classical covariance matrix. The probability of the class label of the residue instance (PCL), and the importance of within-class and between-class (IWB) residue instances. That is, an SVM classifier trained on an imbalanced dataset the data sets without outliers are taken as input for a support vector machine (SVM) ensemble. The proposed SVM ensemble trained on input data without outliers performs better than that with outliers.