Flexible Similarity Searching

Most of the early systems for 3D substructure searching and those described earlier assumed that all of the molecules in the database that was to be searched were completely rigid or could be represented by a single (presumably low-energy) conformation, that is, the systems took little account of the conformational flexibility that characterized most 3D molecules. Such an approach was appropriate because the only databases that were generally available 

It is a common approach to investigate a special case of some scientific problem before considering how the initial solution can be generalized to encompass all of the occurrences of that problem it is hence hardly surprising that the 3D similarity measures that have been described thus far have involved only rigid structures. However, even with such a constraint, the results that have been obtained to date are sufficiently encouraging to suggest that the next major step is to consider how the measures can be extended to encompass conformational flexibility. 

We now describe an important application of 3D similarity searching, one that is rather different in concept from those discussed thus fer. Specifically, we discuss the use of similarity-based techniques to identify molecules in a database of 3D structures that are complementary to a biological receptor site 

The demonstration that DOCK scores are positively correlated with binding affinities has led to widespread use of the program to identify those structures in a database that are most likely to fit the receptor site on steric grounds.Usage is exemplified by a recent paper that discusses the design of potential HIV-1 protease inhibitors.The search used the CONCORD structures for 9561 molecules from the SmithKline Beecham database that contained two oxygen atoms within 5.87 1.0 A of each other, 

Smellie et al. replaced the sphere-matching algorithm in DOCK with a clique-detection procedure in which the hydrogen-bond donor and acceptor atoms of a ligand are docked onto corresponding atoms in the protein. The 

---

Raymond, J.W. and Willett, P. Similarity searching in databases of flexible 3D structures using smoothed bounded distance matrices. /. Chem. Inf. Comput. Sci. 2003, 43, 908-916. 

Abrahamian, E., Fox, P.C., Naerum, L., Christensen, I.T., Thogersen, H., and Clark, R.D. Efficient generation, storage, and manipulation of fully flexible pharmacophore multiplets and their use in 3D similarity searching./. Chem. 

Pearson, W. R., Flexible sequence similarity searching with the FASTA3 program package. Methods Mol Biol, 2000. 132 p. 185-219. 

The retrieval of all 3D structures from a database considered to be similar to a given target structure is comparable with 2D similarity searching. 3D similarity searching raises the problem of conformational flexibility. Schuffenhauer et analysed the BIOSTER database by similarity search using 2D fingerprints and molecular field descriptors. A comprehensive overview on pharmacophore perception and 3D database searches is given in references 101 and 102. 

Renner, S., Schwab, C.H., Gasteiger, J., and Schneider, G. (2006) Impact of conformational flexibility on three-dimensional similarity searching using... 

---