A new approach to the automatic identification of candidates for ligand receptor sites in proteins: (I). Search for pocket regions.

The work presented here is aimed at the topographical analysis of localized regions of receptor proteins leading to the identification of pocket areas (superficial depressions or internal cavities), which may play the role of receptor sites. An algorithm is described that yields complete information about the position of each cavity or superficial depression relative to any point of the protein molecules, as well as detailed information on the atoms constituting it. The applicability of this algorithm to the automatic identification of candidate receptor sites in a receptor protein is also discussed using the typical receptor structure dihydrofolate reductase-methotrexate complex.