Design of New Chemoinformatic Tools for the Analysis ofVirtual Screening Studies: Application to Tubulin Inhibitors

摘要

Virtual screening (VS) experiments yield huge numbers of configurations (conformations plus rotations plus translations). In order to extract important structural information from such a complex database, new chemoinformatic tools are urgently needed. We have clustered and classified by means of "ad hoc" semiautomatic chemoinformatic tools the poses arising from docking experiments conducted on more than 700,000 compounds on tubulin. The results obtained in this way have been compared with those achieved by visual inspection protocols in an attempt to develop new useful tools.