Mahalanobis distance with radial basis function network on protein secondary structures

In this paper, the radial basis function (RBF) network method with the Mahalanobis distance was applied to predict the content of protein secondary structure elements. A study of the Mahalanobis-RBF with different window sizes on the dataset developed by Qian-Sejnowski is given. The RBF network predicts each position in turn based on a local window of residues, by sliding this window along the length of the sequence. Comparison of Gaussian-RBF and Mahalanobis-RBF on the Qian dataset shows that the Mahalanobis distance in using RBF gives better results in the prediction of secondary structure for local sequence structural state.