A discriminative framework for detecting remote protein homologies

A new method for detecting remote protein homologies is introduced and show n to perform well in classifying protein domains by SCOP superfamily. The m ethod is a variant of support vector machines using a new kernel function. The kernel function is derived from a generative statistical model for a pr otein family, in this case a hidden Markov model. This general approach of combining generative models like HMMs with discriminative methods such as s upport vector machines may have applications in other areas of biosequence analysis as well.