TOP: a new method for protein structure comparisons and similarity searches

In order to facilitate the three-dimensional structure comparison of protei ns, software for making comparisons and searching for similarities to prote in structures in databases has been developed. The program identifies the r esidues that share similar positions of both main-chain and side-chain atom s between two proteins. The unique functions of the software also include d atabase processing via Internet- and Web-based servers for different types of users. The developed method and ifs friendly user interface copes with m any of the problems that frequently occur in protein structure comparisons, such as detecting structurally equivalent residues, misalignment caused by coincident match of C-alpha atoms, circular sequence permutations, tedious repetition of access, maintenance of the most recent database, and inconve nience of user interface. The program is also designed to cooperate with ot her tools in structural bioinformatics, such as the 3DB Browser software [P rilusky (1998), Protein Data Bank Q. Newslett. 54, 3-4] and the SCOP databa se [Murzin, Brenner, Hubbard & Chothia (1995). J. Mel. Biol. 247, 536-540], for convenient molecular modelling and protein structure analysis. A simil arity ranking score of 'structure diversity' is proposed in order to estima te the evolutionary distance between proteins based on the comparisons of t heir three-dimensional structures. The function of the program has been uti lized as a part of an automated program for multiple protein structure alig nment. In this paper, the algorithm of the program and results of systemati c tests are presented and discussed.