Neural networks with resilient propagation for protein secondary structure prediction

Abstract

In this paper a neural network secondary structure prediction algorithm is presented. A novel input encoding, based on multidimensional scaling applied to a modified Percent Accepted Mutation amino acid substitution matrix, is developed and shown to be superior to an arbitrary input encoding. Both decimal valued and binary input encodings are compared. Two neural network learning algorithms, Resilient Propagation and Learning Vector Quantization, which have not previously been applied to the problem of protein secondary structure prediction, are compared.