A SURFACE OF MINIMUM AREA METRIC FOR THE STRUCTURAL COMPARISON OF PROTEINS

A new method for comparing protein structures, based on a minimal surf ace metric, is developed. A virtual polypeptide backbone is created by joining consecutive C-alpha atoms in a protein structure. The minimal surface between the virtual backbones of two proteins (the Area Funct ional) is determined numerically using an iterative triangulation stra tegy The first protein is then rotated and translated in space until t he smallest minimal surface is obtained. Such a technique yields the o ptimal structural superposition between two protein segments. It requi res no initial sequence alignment, is relatively insensitive to insert ions and deletions, and obviates the need to select a gap penalty. The optimal minimal area can then be converted to the Area-C-alpha distan ce,measured in angstroms, to determine the structural similarity This technique has been applied to a large class of proteins and is able to detect not only small-scale differences between closely related prote ins but also large-scale topological similarities between evolutionary unrelated proteins that lack any obvious sequence homology: To measur e the similarity between structurally dissimilar proteins, an addition al measure (the Fit Comparison) is developed. This is a scale-invarian t measure of a structural similarity that is useful for determining to pological similarities between dissimilar proteins with unrelated sequ ences. (C) 1996 Academic Press Limited