Biological meaning of nonlinear mapping of substitution matrices in metric spaces

Abstract

Based on the Information Theory, a distance function between amino acids is induced to model the affinity relationship between them. Tables with mapping coordinates have been obtained by using nonlinear multidimensional scaling for different dimensions. These mapping coordinates are meaningless virtual data, but a high relationship with physical and chemical properties is found. The main conclusion is that the number of effective characteristics involved in substitution matrices is low. The hypothesis that hydrophobicity and secondary structure propensities are very important characteristics involved in substitution matrices is reinforced by the analysis of the results.