Statistical analysis of residue variability in cytochrome c

Abstract

This paper presents a quantitative measure of residue variability patterns of the cytochrome c ensemble using statistical analysis. From variations of residues (localized or correlated) among sites in the available cytochrome c sequences, residue variability patterns are derived. Then with the chi-square contingency tables, quantitative and qualitative comparisons are drawn between (1) the residue variability patterns and (2) the molecular and taxonomical features of the molecules. The results, besides providing statistical evidence in support of some observations made by previous researchers, reveal new findings, including some significant conformational, chemical and taxonomical characteristics of several seemingly hypervariate sites. The correlation of the “internal” locations with invariant sites, and that of the “exterior” locations with the highly correlated sites, suggest that the selection pressure in general favors the conservation of localized chemical functions on the internal portion of the sequence, and point to the conservation of the correlated patterns in external locations for various taxonomical groups. The inclusion of the correlated variability patterns, the overall objective, comprehensive and systematic framework, together with the significant analytical results, yields a better understanding of the structure of the molecule and its conservatism.