From chemistry-request@server.ccl.net Thu May 27 06:54:13 1999 Received: from probity.mcc.ac.uk (probity.mcc.ac.uk [130.88.200.94]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA12393 for ; Thu, 27 May 1999 06:54:03 -0400 Received: from nessie.mcc.ac.uk ([130.88.200.20] ident=root) by probity.mcc.ac.uk with esmtp (Exim 1.92 #3) id 10mxjx-0009G0-00; Thu, 27 May 1999 11:50:21 +0100 Received: from serenity.mcc.ac.uk (serenity.mcc.ac.uk [130.88.200.93]) by nessie.mcc.ac.uk (8.8.8/8.8.8) with ESMTP id LAA17564; Thu, 27 May 1999 11:50:00 +0100 (BST) (envelope-from mbdtsnm@mchhpf.ch.man.ac.uk) Received: from [130.88.12.28] (helo=hpf.ch.man.ac.uk) by serenity.mcc.ac.uk with smtp (Exim 1.92 #3) id 10mxjR-000CcK-00; Thu, 27 May 1999 11:49:49 +0100 Date: Thu, 27 May 1999 11:46:33 +0100 Message-Id: <16730.9905271046@hpf.ch.man.ac.uk> From: "Nathaniel (noj) Malcolm" To: wtian@uoguelph.ca Cc: chemistry@ccl.net, jgoddard@uoguelph.ca, huwata@chem.ucla.edu In-Reply-To: (message from WeiQuan Tian on Tue, 25 May 1999 10:30:14 -0400 (EDT)) Subject: Re: CCL:CASMP2(10,10) Problems (fwd) Reply-To: Noj.Malcolm@man.ac.uk X-Beard: Beard v5.3 X-Child: Bethany WeiQuan, just a note of caution, the G98 CASSCF +MP2 is NOT the CASPT2 method of Roos et al., its the OVB-MP2 method ......... noj -- Dr. Noj Malcolm Theory Group, (but only until may 31st) Dept. of Chemistry, University of Manchester, Oxford Road, Manchester. M13 9PL e-mail noj.malcolm@man.ac.uk From chemistry-request@server.ccl.net Thu May 27 09:57:24 1999 Received: from mallard.duc.auburn.edu (mallard2.duc.auburn.edu [131.204.2.23]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id JAA14049 for ; Thu, 27 May 1999 09:57:24 -0400 Received: from localhost (youngd2@localhost) by mallard.duc.auburn.edu (8.8.8+Sun/8.8.8) with SMTP id IAA00255; Thu, 27 May 1999 08:53:49 -0500 (CDT) X-Authentication-Warning: mallard.duc.auburn.edu: youngd2 owned process doing -bs Date: Thu, 27 May 1999 08:53:48 -0500 (CDT) From: David Young X-Sender: youngd2@mallard2 To: Peter Shenkin cc: computational chemistry list Subject: Re: CCL:Average conformation; was, A question In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, If there is a best way to find some sort of molecular average, I would think it would be a weighted average, where the weights are those predicted by a Boltzmann distribution. If the MD simulation was done well, this should be what you get from equally weighted MD trajectories. Dave Young youngd2@mail.auburn.edu ------------------------------------------------------------------------ Live long and maintain marketable job skills. ------------------------------------------------------------------------ On Wed, 26 May 1999, Peter Shenkin wrote: > Hi, > > You could get the "average" conformation by doing an rms superposition > of all conformations on the first and averaging coordinates, or by > doing a more sophisticated multiconformer superposition and then > averaging coordinates. Simon Kearsley has published a multiple > conformation superposition method. So have we, in our XCluster > paper (J. Comp. Chem., v15, pp899-916, 1994). Ours involves looking > at all pairwise superpositions. > > However you do the superposition, the "average" conformation, defined > as above, might not be terribly useful. For instance, if half the > conformations have a +60-degree torsion and the other half have a > -60-degree torsion, the average might have a 0-degree torsion, and > thus would not resemble any conformation in the ensemble. OTOH, > this might be what you want for some applications. > > It is sometimes more useful to use the "most representative" structure; > this is the actual structure from the data set that has the smallest > rms interatomic displacement from the average structure. It's like > using the number from a list of numbers which is closest to the > average, rather than using the average itself. > > In XCluster, we first try to find meaningful clusters (maybe there > would be two, one with the +60 and one with the -60 torsion). Then > we write out the the most representative structure from each > cluster. Or we can write out the average structure, if that's > what you really want. > > Hope this helps, > -P. > > On Wed, 26 May 1999, Richard Wood wrote: > > > Hi all on the CCL > > I have a question for you. I have run 16 molecular dynamics > > simulations of the same molecule, each one using a different set of > > initial velocities. From these MD simulations, I have calculated various > > properties and obtained their averages. I have found the lowest PE for > > each MD sim and the corresponding conformation. My question is this: is > > there any way that I can generate an "average" conformation that would > > give me these "average" properties? > > > > I will post all replies. > > > > Richard > > > > > > > > > > > > > > -- > ********* Peter S. Shenkin; Schrodinger, Inc.; (201)433-2014 x111 ********* > *********** shenkin@schrodinger.com; http://www.schrodinger.com *********** > > -= This is automatically added to each message by mailing script =- > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > > From chemistry-request@server.ccl.net Thu May 27 10:07:37 1999 Received: from schrodinger.com (root@thermidore.schrodinger.com [192.156.98.99]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id KAA14204 for ; Thu, 27 May 1999 10:07:37 -0400 Received: from sandra.schrodinger.com (sandra.schrodinger.com [206.231.140.250]) by schrodinger.com (8.8.5/8.8.5) with ESMTP id HAA09079; Thu, 27 May 1999 07:09:12 -0700 Received: from localhost (shenkin@localhost) by sandra.schrodinger.com (8.8.5/8.8.5) with ESMTP id KAA12193; Thu, 27 May 1999 10:08:40 -0400 X-Authentication-Warning: sandra.schrodinger.com: shenkin owned process doing -bs Date: Thu, 27 May 1999 10:08:40 -0400 (EDT) From: Peter Shenkin To: David Young cc: computational chemistry list Subject: Re: CCL:Average conformation; was, A question In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, Yes, for a conf search, what you say makes some sense. It really does depend on just what you want to do with the result. -P. On Thu, 27 May 1999, David Young wrote: > > Hi, > > If there is a best way to find some sort of molecular average, I > would think it would be a weighted average, where the weights are those > predicted by a Boltzmann distribution. If the MD simulation was done > well, this should be what you get from equally weighted MD trajectories. > > Dave Young > youngd2@mail.auburn.edu > > ------------------------------------------------------------------------ > > Live long and maintain marketable job skills. > > ------------------------------------------------------------------------ > > > On Wed, 26 May 1999, Peter Shenkin wrote: > > > Hi, > > > > You could get the "average" conformation by doing an rms superposition > > of all conformations on the first and averaging coordinates, or by > > doing a more sophisticated multiconformer superposition and then > > averaging coordinates. Simon Kearsley has published a multiple > > conformation superposition method. So have we, in our XCluster > > paper (J. Comp. Chem., v15, pp899-916, 1994). Ours involves looking > > at all pairwise superpositions. > > > > However you do the superposition, the "average" conformation, defined > > as above, might not be terribly useful. For instance, if half the > > conformations have a +60-degree torsion and the other half have a > > -60-degree torsion, the average might have a 0-degree torsion, and > > thus would not resemble any conformation in the ensemble. OTOH, > > this might be what you want for some applications. > > > > It is sometimes more useful to use the "most representative" structure; > > this is the actual structure from the data set that has the smallest > > rms interatomic displacement from the average structure. It's like > > using the number from a list of numbers which is closest to the > > average, rather than using the average itself. > > > > In XCluster, we first try to find meaningful clusters (maybe there > > would be two, one with the +60 and one with the -60 torsion). Then > > we write out the the most representative structure from each > > cluster. Or we can write out the average structure, if that's > > what you really want. > > > > Hope this helps, > > -P. > > > > On Wed, 26 May 1999, Richard Wood wrote: > > > > > Hi all on the CCL > > > I have a question for you. I have run 16 molecular dynamics > > > simulations of the same molecule, each one using a different set of > > > initial velocities. From these MD simulations, I have calculated various > > > properties and obtained their averages. I have found the lowest PE for > > > each MD sim and the corresponding conformation. My question is this: is > > > there any way that I can generate an "average" conformation that would > > > give me these "average" properties? > > > > > > I will post all replies. > > > > > > Richard > > > > > > > > > > > > > > > > > > > > > > -- > > ********* Peter S. Shenkin; Schrodinger, Inc.; (201)433-2014 x111 ********* > > *********** shenkin@schrodinger.com; http://www.schrodinger.com *********** > > > > -= This is automatically added to each message by mailing script =- > > CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To Admins > > MAILSERV@ccl.net -- HELP CHEMISTRY or HELP SEARCH > > CHEMISTRY-SEARCH@ccl.net -- archive search | Gopher: gopher.ccl.net 70 > > Ftp: ftp.ccl.net | WWW: http://www.ccl.net/chemistry/ | Jan: jkl@ccl.net > > > > > > > > > > > > > -- ********* Peter S. Shenkin; Schrodinger, Inc.; (201)433-2014 x111 ********* *********** shenkin@schrodinger.com; http://www.schrodinger.com *********** From chemistry-request@server.ccl.net Thu May 27 12:18:11 1999 Received: from alpha.luc.ac.be (alpha.luc.ac.be [193.190.2.30]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id MAA15830 for ; Thu, 27 May 1999 12:18:11 -0400 Received: from sbg-ts (sbg-ts.luc.ac.be [193.190.1.15]) by alpha.luc.ac.be (8.9.3/8.9.3) with SMTP id SAA04446 for ; Thu, 27 May 1999 18:16:24 +0200 (MET DST) Date: Thu, 27 May 1999 18:16:24 +0200 (MET DST) Message-Id: <199905271616.SAA04446@alpha.luc.ac.be> X-Sender: skwasnie@mail.luc.ac.be X-Mailer: Windows Eudora Pro Version 2.1.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: chemistry@ccl.net From: sergiusz kwasniewski Subject: summary: excitation energies Hi all, a couple of weeks ago I asked the following question: _____________________________________________________ ---0--- CCL:excitation energies > Does anyone know something about the difference between the excitation > energies calculated by RHF CIS and semi-empirical techniques like ZINDO/S > CIS on organic structures? I've done some calculations on aromatic > structures, but the spectra are not comparable at all! Is there a reason for > that behaviour? The following answers where provided to me: ___________________________________________ ---1--- From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: sergiusz kwasniewski Date: Tue, 27 Apr 1999 10:20:45 +0100 Subject: Re: CCL:excitation energies CC: chemistry@ccl.net Dear Sergiusz, HF CIS calculations are generally in error because of neglect of electronic correlation effects. These effects tend to stabilize excited states more than the ground state, and the result is that HF CIS frequently predicts excitation energies that are much too large. Semiempirical procedures like ZINDO/S are especially parameterized to account for the effects of electron correlation (in particular, by empirical adjustment of the parameters used to estimate electron repulsion terms), and these procedures thus predict transition energies of the right order of magnitude, at least for the class of compunds and types of transition for which they were parameterized. By the way, have you tried to apply Time Dependent Density Functional Theory (TD DFT) as implemented in Gaussian 98? For example, TD B3LYP does a great job for electronic transitions. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Phone: +45 4674 2000 (RUC) Department of Chemistry +45 4674 2710 (direct) Roskilde University (RUC) Fax: +45 4674 3011 P.O.Box 260 E-Mail: JSL@virgil.ruc.dk DK-4000 Roskilde, Denmark http://www.rub.ruc.dk/dis/chem/psos/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ---2--- Hi, for the semi-empirical techniques, I've been using the whole space (Energy criterium is 1000 eV), while for the ab initio methods the basis set is 6-311G(d,p) within the FC approximation... But I'm concerned about the low energy part of the spectrum, and then the criterium inside the semi-empirical techniques doesn't change much the appearence of the spectrum. Did anyone encounter these inconsistencies? (More information about this, was inside my last mail) Thanks in advance Sergiusz At 11:57 AM 4/26/99 -0300, Sergio Emanuel Galembeck wrote: >Dear Sergiusz, > What basis are you using? I don't have experience >on RHF/CIS, but the results depends on basis sets >and on the choice of active space. > Best regards, > Sergio >PS: Please, sumarize the answers. >============================================== >Sergio Emanuel Galembeck >e-mail: segalemb@usp.br >=============================================== _________________________________________________ Sergiusz Kwasniewski LUC SBG/TS Universitaire Campus Gebouw D 3590 Diepenbeek BELGIUM tel(direct): 011/268315 fax : 011/268301 email : sergiusz.kwasniewski@luc.ac.be _________________________________________________ From chemistry-request@server.ccl.net Thu May 27 12:50:07 1999 Received: from theory.tc.cornell.edu (THEORY.TC.CORNELL.EDU [128.84.30.174]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id MAA16372 for ; Thu, 27 May 1999 12:50:07 -0400 Received: from tc.cornell.edu (REN.TC.CORNELL.EDU [128.84.244.22]) by theory.tc.cornell.edu (8.8.4/8.8.3/CTC-1.0) with ESMTP id MAA15807 for ; Thu, 27 May 1999 12:46:44 -0400 Sender: richard@tc.cornell.edu Message-ID: <374D76F3.AE9E2952@tc.cornell.edu> Date: Thu, 27 May 1999 12:46:43 -0400 From: Richard Gillilan X-Mailer: Mozilla 4.5 [en] (X11; I; IRIX 6.2 IP22) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@server.ccl.net Subject: DX goes open source Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sorry if this has already been posted. I'm, temporarily off the list but thought this would be of real interest. The IBM Data Explorer visualization software that we have used for a number of years has officially gone open source and can be obtained free from the web site (OpenDX.org). Richard Gillilan Cornell Theory Center ------------------------------------------------- The long awaited announcement is now public. Source code will officially be released in two days. http://www.ibm.com/dx As a founding member of the OpenDX.org, I'm happy to answer any questions that aren't answered on the website(s). Cheers, Chris Pelkie Scientific Visualization Producer Cornell Theory Center 618 Rhodes Hall Cornell University Ithaca, NY 14853-3801 chrisp@tc.cornell.edu From chemistry-request@server.ccl.net Thu May 27 14:24:46 1999 Received: from cornelius.INS.CWRU.Edu (root@cornelius.INS.CWRU.Edu [129.22.8.216]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA17180 for ; Thu, 27 May 1999 14:24:46 -0400 Received: from po.cwru.edu (hxt10@chem51374.CHEM.CWRU.Edu [129.22.200.174]) by cornelius.INS.CWRU.Edu with ESMTP (8.8.8+cwru/CWRU-3.6) id OAA07602; Thu, 27 May 1999 14:21:22 -0400 (EDT) (from hxt10@po.cwru.edu for ) Sender: hxt10@cornelius.INS.CWRU.Edu Message-ID: <374D8D53.E3899DE1@po.cwru.edu> Date: Thu, 27 May 1999 18:22:12 +0000 From: Gavin Tsai Organization: CWRU X-Mailer: Mozilla 4.6 [en] (X11; I; Linux 2.0.36 i686) X-Accept-Language: en MIME-Version: 1.0 To: CHEMISTRY@ccl.net Subject: Ahlrich's VTZ basis set Content-Type: text/plain; charset=big5 Content-Transfer-Encoding: 7bit Hi, CCLers: I tried to download Ahlrich's VTZ basis set from tchibm3.chemie.uni-karlsruhe.de which they allowed anonymous logging. Unfortunately, the anonymous is rejected. The VTZ basis set also does not list on his 1992 JCP paper. Any help would be appreciated!! Gavin Tsai From chemistry-request@server.ccl.net Thu May 27 14:39:10 1999 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA17369 for ; Thu, 27 May 1999 14:39:10 -0400 Received: from krakow.ccl.net (krakow.ccl.net [192.148.249.195]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id OAA12601; Thu, 27 May 1999 14:35:42 -0400 (EDT) Date: Thu, 27 May 1999 14:35:43 -0400 (EDT) From: Jan Labanowski To: Gavin Tsai cc: CHEMISTRY@ccl.net Subject: Re: Ahlrich's VTZ basis set In-Reply-To: <374D8D53.E3899DE1@po.cwru.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 27 May 1999, Gavin Tsai wrote: > Hi, CCLers: > > I tried to download Ahlrich's VTZ basis set from > tchibm3.chemie.uni-karlsruhe.de which they allowed anonymous logging. > Unfortunately, the anonymous is rejected. The VTZ basis set also does > not list on his 1992 JCP paper. Any help would be appreciated!! > > Gavin Tsai > Check the venerable {:-)} CCL archives: http://www.ccl.net/cca/data/basis-sets/ Jan jkl@ccl.net Jan K. Labanowski | phone: 614-292-9279, FAX: 614-292-7168 Ohio Supercomputer Center | Internet: jkl@ccl.net 1224 Kinnear Rd, | http://www.ccl.net/chemistry.html Columbus, OH 43212-1163 | http://www.ccl.net/ From chemistry-request@server.ccl.net Thu May 27 15:37:48 1999 Received: from gateway.bioreason.com (bioreason.com [207.108.245.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id PAA18020 for ; Thu, 27 May 1999 15:37:48 -0400 Received: from bioreason.com (val [192.168.1.51]) by gateway.bioreason.com (8.9.3/8.9.3) with ESMTP id NAA00448; Thu, 27 May 1999 13:26:29 -0600 Sender: dalke@bioreason.com Message-ID: <374D9DFB.636E7699@bioreason.com> Date: Thu, 27 May 1999 13:33:15 -0600 From: Andrew Dalke Organization: Bioreason, Inc. X-Mailer: Mozilla 4.05C-SGI [en] (X11; U; IRIX64 6.5 IP30) MIME-Version: 1.0 To: Richard Gillilan CC: chemistry@server.ccl.net Subject: Re: CCL:DX goes open source References: <374D76F3.AE9E2952@tc.cornell.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Richard Gillilan said: > The IBM Data Explorer visualization software that we have used > for a number of years has officially gone open source and can > be obtained free from the web site (OpenDX.org). And if anyone is interested, here are instructions for getting the source to compile on IRIX 6.5 using the 7.2 C/C++ compiler. 1) download the source (and examples) from http://www.research.ibm.com/dx/download/ 2) The configuration code uses the value of SHELL rather than /bin/sh, so if you have a non-sh compatible shell (like tcsh) you are best off starting 'sh' then doing SHELL=/bin/sh export SHELL 3) If you want to use SGI's C++ compiler over GNU, you also need to set the CXX environment variable to "CC", as in CXX=CC export CXX 4) Configure by running "./configure" 5) The default 'make' on SGIs isn't very good. Use "smake" or GNU make ("gmake" or "gnumake") instead. This can best be done by adding #!/usr/sbin/smake to the top of the newly created Makefile. The normal make will restart with the #! indicated make if given. This is needed for compiling dx because otherwise every time you run "make" it will try to recompile some of the files. This is annoying because it will try to recompile even when using "make install" after the system has been built. 6) The file "src/exec/libdx/memory.c" didn't compile because of the code: === #ifdef SYSTYPE_SVR4 if (inv->inv_class==INV_MEMORY && inv->inv_type==INV_MAIN) { physmem = (uint)((double)inv->inv_state / 1024. / 1024.); break; } #else if (inv->class==INV_MEMORY && inv->type==INV_MAIN) { physmem = (uint)((double)inv->inv_state / 1024. / 1024.); break; } #endif /* SVR4 */ === It is supposed to use the first block but instead uses the second. Easiest fix is to replace #ifdef with #ifndef . 7) Once the "#!" was added, "make" it. 8) Become root and "make install". There is a problem with the code -- some of it assumes data files are in "/usr/lpp/dx" instead of "/usr/local/dx" and it will complain about missing files during startup. The only fix I know so far is to make a symbolic link, as in cd /usr ln -s local lpp 9) Some of the help files are sym linked to themselves. val> pwd /usr/local/dx/help val> ls -l GarHelpDir BMHelpDir lrwxr-xr-x 1 root sys 28 May 27 10:48 BMHelpDir -> /usr/local/dx/help/MBHelpDir lrwxr-xr-x 1 root sys 29 May 27 10:48 GarHelpDir -> /usr/local/dx/help/GarHelpDir There is no fix for this, but it hasn't been a problem (yet?). 10) To use the sample files, set the environment variable DXDATA to point to your dxsamples-4.0/data directory. 11) Start the program by running "/usr/local/dx/bin_sgi/startupui". There may be other ways to start it, but I'm not sure what they are. Also, looking at the memory usage with "gmemusage", at least some of the time DX wants up to 400 MB of memory. If you run "dxui" directly, it seems to keep that memory need, but starting >from the startupui makes that need be only transient. I don't understand the behaviour, but it may be because dxui isn't meant to be started directly. Then take a look at some of the Chemistry extensions developed by Richard Gililan and others at the Theory Center, under http://www.tc.cornell.edu/DX/ Andrew Dalke dalke@bioreason.com