From romano@bellatrix.pcl.ox.ac.uk Thu Jul 9 07:26:49 1998 Received: from oxmail4.ox.ac.uk (oxmail4.ox.ac.uk [163.1.2.33]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id HAA14720 Thu, 9 Jul 1998 07:26:46 -0400 (EDT) Received: from bellatrix.pcl.ox.ac.uk by oxmail4 with SMTP (PP); Thu, 9 Jul 1998 12:26:35 +0100 Received: by bellatrix.pcl.ox.ac.uk (940816.SGI.8.6.9/940406.SGI.AUTO) for chemistry@www.ccl.net id LAA23721; Thu, 9 Jul 1998 11:19:53 GMT From: romano@bellatrix.pcl.ox.ac.uk (Romano Kroemer) Message-Id: <199807091119.LAA23721@bellatrix.pcl.ox.ac.uk> Subject: partial charges for large molecules To: chemistry@www.ccl.net Date: Thu, 9 Jul 1998 12:19:52 +0100 (BST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Dear all, could somebody point me to references for the following problem/method: I have a large molecule and want to calculate partial charges with a fairly sophisticated method. As the molecule is too large, I divide it into moderately sized _overlapping_ pieces, calculate the partial charges for each of these chunks, and derive then the charges for the whole molecule. The "overlapping pieces" procedure is necessary in order to avoid "boundary problems". Comments on the advantages and drawbacks of such an approach (such as the perturbations of the charges on one end of the molecule by the charges on the other end when the molecule is not linear) are of course welcome. Thank you very much in advance for your input. I will post a summary to the list. Yours sincerely, Romano Kroemer -- =============================================================================== Dr. Romano T. Kroemer Phys. & Theoret. Chem. Lab. University of Oxford South Parks Rd. | Tel: ++44-1865-275475 Oxford OX1 3QZ | Fax: ++44-1865-275410 England, U.K. | Email: romano@bellatrix.pcl.ox.ac.uk =============================================================================== From bernhold@npac.syr.edu Thu Jul 9 12:10:50 1998 Received: from snake.npac.syr.edu (snake.npac.syr.edu [128.230.162.60]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id MAA20455 Thu, 9 Jul 1998 12:10:48 -0400 (EDT) Received: from localhost (bernhold@localhost) by snake.npac.syr.edu (950413.SGI.8.6.12/8.6.6) with SMTP id MAA10010; Thu, 9 Jul 1998 12:10:31 -0400 Message-Id: <199807091610.MAA10010@snake.npac.syr.edu> X-Authentication-Warning: snake.npac.syr.edu: Host localhost didn't use HELO protocol Reply-to: bernhold@npac.syr.edu To: KATAGIRI Masahiko cc: chemistry@www.ccl.net, T.Somasundaram@qub.ac.uk cc: bernhold@npac.syr.edu Subject: Re: CCL:LaPack on SX4 In-reply-to: Message from KATAGIRI Masahiko of "Thu, 09 Jul 1998 14:46:44 +0900." <199807090546.OAA02882@momokusa.nrim.go.jp> Content-Type: text/plain; charset="us-ascii" Date: Thu, 09 Jul 1998 12:10:31 -0400 From: "David E. Bernholdt" Optimized libraries, such as LAPACK, BLAS, etc. are often (but not exclusively) provided by the hardware vendor. Therefore you might check with the people running your SX4 to see if they're already on the system, or if there's any possibility of purchasing them. http://www.netlib.org is a useful place to look for numerical libraries, and this happens to be the official home of the BLAS, LAPACK and some other libraries. This is the place where you can find SOURCE CODE for many freely available libraries, but of course that may not provide the best performance for your platform. In the LAPACK area on Netlib, they even provide precompiled libraries for a number of platforms, though not SX4. But again, these are simply compiled versions of the source code, probably with reasonable compiler switches w.r.t. optimization -- they are not the highly optimized libraries that other vendors often provide. Good luck -- David E. Bernholdt | Email: bernhold@npac.syr.edu Northeast Parallel Architectures Center | Phone: +1 315 443 3857 111 College Place, Syracuse University | Fax: +1 315 443 1973 Syracuse, NY 13244-4100 | URL: http://www.npac.syr.edu From hinsen@dirac.cnrs-orleans.fr Thu Jul 9 12:33:16 1998 Received: from dirac.cnrs-orleans.fr (dirac.cnrs-orleans.fr [163.9.6.67]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id MAA20967 Thu, 9 Jul 1998 12:33:14 -0400 (EDT) Received: (from hinsen@localhost) by dirac.cnrs-orleans.fr (AIX4.3/UCB 8.7/8.7) id SAA13326; Thu, 9 Jul 1998 18:33:00 +0200 (DFT) Date: Thu, 9 Jul 1998 18:33:00 +0200 (DFT) Message-Id: <199807091633.SAA13326@dirac.cnrs-orleans.fr> From: Konrad Hinsen To: laaksone@csc.fi CC: CHEMISTRY@www.ccl.net In-reply-to: (message from Leif Laaksonen on Wed, 8 Jul 1998 15:35:35 +0300 (GFT Daylight Time)) Subject: Re: CCL:Do the Windows NT have any impact on the comp chem field? References: > My question is now whether Windows NT and Linux already play a significant > role in the computation chemistry field? Are the Windows NT and Linux As for Linux, I'd say definitely yes. From a technical point of view, a PC running Linux is just another member of the big Unix family, with most of the common Unix software available, including more and more commercial programs. Due to their excellent price-performance ratio, more and more research groups are using Linux machines, at least for use personal workstations, and sometimes even in clusters for heavy number crunching. As for Windows NT, the future is more difficult to predict. Most of the computational science software comes from the Unix world, and porting it to NT is not trivial. Moreover, most people will have to work with some Unix variety anyway for heavy-duty computing, and this won't change in the foreseeable future (has Cray already announced an NT version of the T3E?). So using NT means learning two systems, with no immediate advantage. I expect computational scientists to stick to Unix for the same reason that office users stick to Windows: familiarity and availability of application programs. An exception might be dedicated graphics machines, for which the operating system doesn't matter much. The main market I see for NT is people whose job description is science management rather than active science. They might prefer NT for its better provision with office software, and want a certain range of scientific applications (mostly visualization) as well. This could be the group that SGI is aiming at with its announced NT machines. -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsen@cnrs-orleans.fr Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69 Rue Charles Sadron | Fax: +33-2.38.63.15.17 45071 Orleans Cedex 2 | Deutsch/Esperanto/English/ France | Nederlands/Francais ------------------------------------------------------------------------------- From bmurray@jasper.uor.edu Thu Jul 9 19:03:01 1998 Received: from jasper.uor.edu (JASPER.UOR.EDU [192.251.139.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id TAA28388 Thu, 9 Jul 1998 19:03:00 -0400 (EDT) Received: by jasper.uor.edu; id AA09293; Thu, 9 Jul 1998 16:04:17 -0700 Message-Id: <35A54C0C.6C1D@uor.edu> Date: Thu, 09 Jul 1998 16:02:36 -0700 From: Barbara Murray Reply-To: bmurray@jasper.uor.edu Organization: University of Redlands X-Mailer: Mozilla 3.0Gold (Win16; U) Mime-Version: 1.0 To: chemistry@www.ccl.net Subject: CCL: looking for molecular modeling program on web or free References: <35A2DC15.EC2498DD@info.combichem.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I am part of a national committee running an undergraduate pharmaceutical course on-line in the fall. We will be taking the students through all the steps that are involved in getting a drug to market. One of the things we want the students to do is to be able to draw and look at their chosen molecule in some sort of modeling program. Now, they don't really need to do calculations on this molecule; just look at it and move it around. Calculations would only be icing on the cake. Students from many universities all over the country (maybe the world) will be taking this course since it is a web based course, and these students will have a wide variety of computer capabilities. That is why I would like to find a simple program that can be downloaded from the web or run from the course's web board. I thought this would be easy since I knew about Chime and Rasmol, but I soon discovered that they are only visualization programs to look at someone else's molecules (like PDB files). Please correct me if I'm wrong. Does anyone out there have a suggestion? Ideas? -- *********************************************************************** Barbara Murray bmurray@uor.edu Chemistry Department 909-793-2121 ext 2374 University of Redlands FAX 909-793-2029 1200 E Colton, PO Box 3080 Redlands, CA 92373-0999