From chemistry-request@server.ccl.net Wed Mar 29 01:49:44 2000 Received: from mail.virco.be ([195.207.78.10]) by server.ccl.net (8.8.7/8.8.7) with SMTP id BAA07799 for ; Wed, 29 Mar 2000 01:49:17 -0500 Received: from [192.168.2.168] by mail.virco.be (CommuniGate SMTP 2.7.2) with ESMTP id S.001841210212 for ; Wed, 29 Mar 2000 08:39:04 +0100 Sender: jan@ccl.net Message-ID: <38E1A80A.BF7B0CD8@tibotec.be> Date: Wed, 29 Mar 2000 08:51:54 +0200 From: jan de kerpel Organization: Tibotec X-Mailer: Mozilla 4.07C-SGI [en] (X11; I; IRIX64 6.5 IP30) MIME-Version: 1.0 To: chemistry@ccl.net Subject: CCL:Clustering compounds from large databases Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear CCL members, I would like to know the availability (both freeware as commercial) of software able to cluster chemical compounds based on physical (e.g. logP), structural (smiles string/2D/SDF) and biological screening data. The number of compounds in the database is typically 100.000 to 150.000 or larger. The questions are : 1) available software based on 2D clustering and, is there already software capable to work in 3D on this large amount of compounds (or smaller databases)? 2) where can I find information about the algorithms, and are there other than SMILES strings used to generate structure keys? 3) has anybody an idea of the ratio of cpu usage when doing 2D versus 3D clustering on this amount of compounds, and what are the expected bottlenecks? Thanks! Jan -- ************************************************************* TIBOTEC Dep. Medicinal Chemistry Gen. De Wittelaan 11B 3 B-2800 Mechelen (BELGIUM) tel : (32) 15 286.300 fax : (32) 15 286.349 email : jan.dekerpel@tibotec.be *************************************************************