Why Neighbor_Search?
Neighbor_Search is a program for calculating the "neighborhood" of a given compound. The neighborhood refers to those compounds which are structurally similar or "close."
To be precise, Neighbor_Search calculates the "distance" between compounds in the Daylight fingerprint space, using a measure of similarity known as the "Tanimoto coefficient." By now you must be saying "Hot-diggity!" Once these distances are known, then the compounds are sorted and the nearest neighbors, or neighborhood, may be reported.
While the neighborhood sometimes resembles the results of a conventional substructure search, the neighbor concept is actually a more "fuzzy" notion than a substructure, so that Neighbor_Search may be used as a complimentary tool to conventional substructure searching.
Neighbor_Search was developed by:
TJ O'Donnell
Tom Doman
Michael Cibulskis
of the Structural and Computational Chemistry Group