THE STRUCTURAL ALIGNMENT BETWEEN 2 PROTEINS - IS THERE A UNIQUE ANSWER

Structurally similar bur sequentially unrelated proteins have been dis covered and rediscovered by many re searchers, using a variety of stru cture comparison tools. For several pairs of such proteins, existing s tructural alignments obtained from the literature, as well as alignmen ts prepared using several different similarity criteria, are compared with each other. It is shown that, in general, they differ from each o ther, with differences increasing with diminishing sequence similarity . Differences are particularly strong between alignments optimizing gl obal similarity measures, such as RMS deviation between C alpha atoms, and alignments focusing on more local features, such as packing or in teraction pattern similarity. Simply speaking, by putting emphasis on different aspects of structure, different structural alignments show t he unquestionable similarity in a different way. With differences betw een various alignments extending to a point where they can differ at a ll positions, analysis of structural similarities leads to contradicto ry results reported by groups using different alignment techniques. Th e problem of uniqueness and stability of structural alignments is furt her studied with the help of visualization of the suboptimal alignment s. It is shown that alignments are often degenerate and whole families of align ments can be generated with almost the same score as the ''o ptimal alignment.'' However, for some similarity criteria, specially t hose based on side-chain positions, rather than C alpha positions, ali gnments in some areas of the protein are unique. This opens the questi on of how and if the structural alignments can be used as ''standards of truth'' for protein comparison.