PREDICTION OF PROTEIN STRUCTURAL CLASSES BY MODIFIED MAHALANOBIS DISCRIMINANT ALGORITHM

We first discuss quantitative rules for determining the protein struct ural classes based on their secondary structures. Then we propose a mo dification of the least Mahalanobis distance method for prediction of protein classes. It is a generalization of a quadratic discriminant fu nction to the case of degenerate covariance matrices. The resubstituti on tests and leave-one-out tests are carried out to compare several me thods. When the class sample sizes or the covariance matrices of diffe rent classes are significantly different, the modified method should b e used to replace the least Mahalanobis distance method. Two lemmas fo r the derivation of our new algorithm are proved in an appendix.