W0103

A New Algorithm for Protein Structure Comparison which Probes Functional Similarity. Philip E. Bourne and Ilya N. Shindyalov, San Diego Supercomputer Center, University of California, San Diego, 9500 Gilman Drive, La Jolla CA, 92093

The 3-D comparison of protein structures is a non-trivial problem (Godzik 1996), yet the result is important to crystallographers trying to better understand the function of a protein being studied, or indeed to determine whether their protein exhibits a unique fold. While many methods have been proposed to address this problem (see (Holm and Sander 1994) for a review) they are computationally expensive, not robust enough, or not intended for wide use. Exceptions are the Dali (Holm and Sander 1997) and VAST (Hogue, Ohkawa et al. 1996) methods used to calculate lookup tables describing structure similarity, FSSP and structure neighbors in Entrez, respectively. These tables are available via the Web. A new algorithm, combinatorial extension (CE) of the optimal path will be described (Shindyalov and Bourne 1998). CE is available at http://cl.sdsc.edu/ce.html. Unlike Dali and VAST which are based on geometry and secondary structure, CE can also include information about the conserved positions and physicochemical properties. The application to the alignment of all protein kinase structures will be described.

Godzik, A. (1996). "The structural alignment between two proteins: is there a unique answer?" Protein Sci 5(7): 1325-38.

Hogue, C. W., H. Ohkawa, et al. (1996). "A dynamic look at structures: WWW-Entrez and the Molecular Modeling Database." Trends Biochem Sci 21(6): 226-9.

Holm, L. and C. Sander (1994). "Searching protein structure databases has come of age." Proteins 19(3): 165-73.

Holm, L. and C. Sander (1997). "Dali/FSSP classification of three-dimensional protein folds." Nucleic Acids Res 25(1): 231-4.

Shindyalov, I. N. and P. E. Bourne (1998). "Protein Structure Alignment by Incremental Combinatorial Extension of the Optimum Path." Protein Engineering Accepted.