From owner-chemistry@ccl.net Fri Feb 13 09:21:01 2009 From: "Vincent Leroux vincent.leroux_._loria.fr" To: CCL Subject: CCL: Chemical library analysis tool Message-Id: <-38608-090213082314-26455-d3jRwebDKMPwr3DYJMgDWQ() server.ccl.net> X-Original-From: Vincent Leroux Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Fri, 13 Feb 2009 14:22:54 +0100 MIME-Version: 1.0 Sent to CCL by: Vincent Leroux [vincent.leroux|a|loria.fr] Hi Sangeetha, These two papers describe interesting programs being developed by academics : http://dx.doi.org/10.1021/ci7001507 (ParaSurf/ParaFit) http://dx.doi.org/10.1007/s11030-006-9033-5 (ScreeningAssistant) The representations of molecules being used are completely different (property-mapped spherical harmonics surfaces of conformers and short descriptors fingerprints respectively), so results of the two programs might be complementary. For clustering I'd rather use the ParaFit, and for quick diversity assessment ScreeningAssistant. If you want to build subsets from large databases for virtual screening, then ParaSurf/ParaFit is fine if you want to select candidates for optimization by selecting the database molecules most 3D-structurally similar to known hits, or at least if you have prior SAR knowledge that allows you to select the most interesting molecular clusters for examination first. If this is not the case, then you might just want to build the most diverse database drug-like subset of a given (much lower) size. ScreeningAssistant is perfectly good for this, see http://dx.doi.org/10.1007/s00894-007-0257-9 Do not hesitate to contact the corresponding authors for more information. VL Sangeetha Vimal srdshigella-x-gmail.com a écrit : > Sent to CCL by: "Sangeetha Vimal" [srdshigella,gmail.com] > Hi all, > > Can someone suggest me a chemical library analysis tool, which can help in clustering/diversity analysis of chemical compounds in the database. > > Thanks > Sangeetha > >