From owner-chemistry#* at *#ccl.net Mon Jul 11 12:49:28 2005 From: "CCL" To: CCL Subject: CCL: CCL Macromodel conformation search Message-Id: <-28876-050711121545-11837-gJKokDZHWSap5rm98+rrIw(_)server.ccl.net> X-Original-From: Eric Bennett Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed MIME-Version: 1.0 Sent to CCL by: Eric Bennett khunter(_)mta.ca wrote: >Was wondering if it was possible to do a conformational search with Macromodel >where one is changing the puckering in a five membered ring. > > You can search ring conformations using either (a) the LMCS [for a few hundres atoms] or LMC2 [for very large systems] directives or (b) RCA4 and TORS directives in the command file. LMCS is easier and tends to be more efficient for large rings, although it may not matter for a five membered ring. If you are setting up the job in Maestro, then under the "CSearch" tab of the Conformational Search window you need to pick a 'Method' that includes "Low Mode". If you are searching a single molecule then you probably either want "Mixed MCMM/Low Mode" (if you are defining other torsions to be searched) or "Pure Low Mode" (if you are not defining torsions). LowMode distorts the molecule along some cominbation of its calculated vibrational modes and will perturb any atoms that are not "frozen" (much to my consternation it *will* try to move atoms which are "fixed"). As compared to simple torsion searching, LowMode is also more likely to flip chiral centers so be sure to define chiral centers if your molecule has any. Some LMCS references: J. Comput. Chem. 20(15), 1671-1684 (1999) JACS 118(21), 5011-5019 (1996) -- Eric Bennett, Center for Drug Design, U of Minnesota I am not a vegetarian because I love animals. I am a vegetarian because I hate plants. - A. Whitney Brown