From Eugene.Leitl@lrz.uni-muenchen.de Sat May 3 11:40:31 1997 Received: from sunsrv5.lrz-muenchen.de for Eugene.Leitl@lrz.uni-muenchen.de by www.ccl.net (8.8.3/950822.1) id LAA01070; Sat, 3 May 1997 11:17:38 -0400 (EDT) Received: from sun6.lrz-muenchen.de by sunsrv5.lrz-muenchen.de; Sat, 3 May 97 17:17:37 +0200 Received: by sun6.lrz-muenchen.de (5.x/SMI-SVR4) id AA02456; Sat, 3 May 1997 17:17:36 +0200 Date: Sat, 3 May 1997 17:17:36 +0200 (MET DST) From: Eugene Leitl X-Sender: ui22204@sun6 To: chemistry Subject: ???: molecular dynamics, where to turn for information Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, yesterday I've picked up "The Art of Molecular Dynamics" by D.C. Rappaport. As a newbie, I found it so far to be very good. Question is: are the techniques described therein truly state of the art? If one is interested in MD merely, which other must reads are out there? Which journals should one read. I noticed this forum focuses on quantum/ab initio stuff mostly. If one is interested in long-term, large-system dynamics, is there a different forum for it? Thanks! -- Eugene Leitl From rvenable@deimos.cber.nih.gov Sat May 3 15:40:40 1997 Received: from deimos.cber.nih.gov for rvenable@deimos.cber.nih.gov by www.ccl.net (8.8.3/950822.1) id PAA06000; Sat, 3 May 1997 15:28:29 -0400 (EDT) Received: from localhost by deimos.cber.nih.gov with SMTP (1.37.109.14/16.2) id AA033037722; Sat, 3 May 1997 15:28:42 -0400 Date: Sat, 3 May 1997 15:28:42 -0400 (EDT) From: Rick Venable To: Eugene Leitl Cc: chemistry Subject: Re: CCL:???: molecular dynamics, where to turn for information In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sat, 3 May 1997, Eugene Leitl wrote: > yesterday I've picked up "The Art of Molecular Dynamics" by D.C. > Rappaport. As a newbie, I found it so far to be very good. Question is: > are the techniques described therein truly state of the art? If one is > interested in MD merely, which other must reads are out there? Which > journals should one read. A "must read" book from the technical theory point of view is "Computer Simulation of Liquids", by M.P. Allen and D.J. Tildesley Oxford University Press (Clarendon Press) ISBN 0-19-855375-7 ISBN 0-19-855645-4 (pbk) While there are several more general books which discuss MD of proteins and other biological systems, if you're looking for the real details, it's hard to find a better single reference to the ideas and math underlying MD and MC simulations of condensed phases. > I noticed this forum focuses on quantum/ab initio stuff mostly. If one is > interested in long-term, large-system dynamics, is there a different forum > for it? There are some application specific mailing lists and/or newsgroups, for programs such as X-Plor, CHARMM, AMBER, etc.; unless your actually using one of the software package in question, they aren't that interesting. While at times the CCL list seems to be a surrogate for the Gaussian helpdesk, there are sometimes items of more general interest, or pertaining to other areas of computational chemistry, such as MD and MC simulations. If you find a better list or newsgroup for MD, I'd be interested in hearing about it. -- Rick Venable =====\ |=| "Eschew Obfuscation" FDA/CBER Biophysics Lab |____/ |=| Bethesda, MD U.S.A. | \ / |=| ( Not an official statement or rvenable@deimos.cber.nih.gov | \ / |=| position of the FDA; for that, http://nmr1.cber.nih.gov/ \/ |=| see http://www.fda.gov ) From nash@chem.wisc.edu Thu May 1 13:40:02 1997 Received: from mail5.doit.wisc.edu for nash@chem.wisc.edu by www.ccl.net (8.8.3/950822.1) id MAA18780; Thu, 1 May 1997 12:44:05 -0400 (EDT) Received: from audumla.students.wisc.edu by mail5.doit.wisc.edu; id LAA30758; 8.7.5/50; Thu, 1 May 1997 11:43:58 -0500 Received: from [144.92.180.52] by audumla.students.wisc.edu; id LAA118940; 8.6.9W/42; Thu, 1 May 1997 11:42:24 -0500 X-Sender: jrnash@144.92.104.66 Message-Id: In-Reply-To: <33662A8E.6D54@worldnet.att.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Thu, 1 May 1997 11:40:44 -0500 To: "Jeffrey J. Ayres" From: "John R. Nash" Subject: Re: Babel Cc: CHEMISTRY@www.ccl.net Jeff (and the rest of CCL) - >The Babel program seems very easy to use. For instance >converting >molgen files to pdb files. Babel is indeed very easy to use, especially for freeware. Since it is freeware, however, the quality of the individual modules is to some extent based on how much testing was done by the authors and third parties. For example, the converters to and from Gaussian are very good, since lots of people used them and gave feedback to the authors (who nicely went back to the code and fixed things). The CACAO converter is, unfortunately, much less used, and it still has some problems. I'm not sure what the current status of Babel is; the authors have received their PhD's and are in the industrial world now, and I haven't seen an update in some time. You might want to email the "babel" address at Arizona that's mentioned in the README to see if anyone's answering. I'll do some checking myself. Since the source code has always been available, it seems to me that development could resume if someone else picked it up; maybe I'll have a look at the code after I graduate. >However it stumbles converting molgen or >mopac output files into C.A.C.A.O. files. The commands >babel -imolgen butane.mol -ocaccrt butane.in >babel -imolgen butane.mol -ocaccrt butane.cor >babel -imolgen butane.mol -ocaccrt butane.bin >where each tried separately. The last two files were inputted directly >into the CACAO program without doing an extended Huckel(eh). >The program issued the error message binary file corrupted, also the >error message does not recognize atom type 6. The extension you want for either internal or cartesian input to CACAO is .in. The .cor and .bin are generated by the EH program when the EHMO calculation is run. Reading a file.in into CACAO without its .cor and .bin will cause it to complain that it can't find the supporting files, but you can still use it to view the structure to make sure that the geometry transferred correctly. If you want to look at orbitals, you have to run the EHMO program (EH . FILENAME); babel is only a geometry converter, and it can't do anything with checkpoint files from qm programs. I would suggest using the CACAO cartesian converter of Babel (as you are), since the internal coordinates converter seems more broken. What you will get out is close to the input file you want, but there are still things needed to be tweaked. The input for CACAO 4.0 is Fortran format, so every space counts (for example, carbon is "C" not "C" since the element field is two characters long). Supposedly, a free-format CACAO is in the works... any one seen an update? My advice is to do what I do: use babel to convert to CACAO cartesian, then go through the input file carefully, using one of the CACAO example files that works as a guide. The manual tells explicitly what the program expects, so have that open when you do it, too. This is hardly ideal (I hate Fortran format!), but I have gotten it to work for me. One more note: if you want to look at Walsh diagrams, you have to use internal coordinates (z-matrix). As I said, the Babel CACAO internal converter isn't quite working, so it won't be much help. My suggestion is to convert to Gaussian z-matrix and exit the file so it looks like a CACAO one, keeping in mind that the way CACAO and Gaussian start and label their zmat's is slightly different. Good luck, and I hope this helps! -john nash almost a PhD, uw-madison chemistry -==-John R. Nash-==-nash@chem.wisc.edu-==-UW-Madison Chem. Dept-==- --- "When in doubt, roll!" --- From ccl@www.ccl.net Fri May 2 10:47:18 1997 Received: from bedrock.ccl.net for ccl@www.ccl.net by www.ccl.net (8.8.3/950822.1) id KAA25421; Fri, 2 May 1997 10:09:13 -0400 (EDT) Received: from dns1.uga.edu for buyong@ibmnla.chem.uga.edu by bedrock.ccl.net (8.8.3/950822.1) id KAA09343; Fri, 2 May 1997 10:08:28 -0400 (EDT) From: Received: from ibmnla.chem.uga.edu (ibmnla.chem.uga.edu [128.192.5.8]) by dns1.uga.edu (8.8.5/8.8.3) with SMTP id KAA13594; Fri, 2 May 1997 10:08:27 -0400 Received: from localhost by ibmnla.chem.uga.edu (AIX 3.2/UCB 5.64/4.03) id AA16470; Fri, 2 May 1997 10:08:20 -0400 Date: Fri, 2 May 1997 10:08:17 -0400 (EDT) To: "Joseph W. Bausch" Cc: chemistry@ccl.net Subject: Re: CCL:1-3 kcal/mol In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII My two points regarding quantum calculation convergency: 1. The standard way to test reliability of calculations: i) do calculations with different basis sets(with increasing quality ii) do calculations starting from HF(SCF) then MP2,MP3,MP4....CCSD(T)... iii) then you will know the trend of convergency. For large system, single point calculations at the correlation level are recommended. You must do this way if you are uneasy about your calculations reliability. 2. convergency varies with chemical systems. It is rather easy for well behavoired molecules (closed shell main elements molecules + nonconjugated). Care should be allway excised for open shell molecules or molecules with transition metals. I agree with the point that the relative energies are more reliable than absolute energy. To make it more clear, I would say that it is pretty reliable to obtain relative energies between isomers with similar bonds(especially for rotamers:)). ================================================================= Dr. Buyong Ma buyong@ibmnla.chem.uga.edu Computational Center for Molecular Structure and Design Department of Chemistry University of Georgia Athens, Georgia 30602 USA Voice (706) 542-2044 ================================================================= On Thu, 24 Apr 1997, Joseph W. Bausch wrote: > > > My group has been using Gaussian 94 to optimize the geometries of a variety > of polyboranes and heteroatom substituted polyboranes. We usually go up to > the MP2(FULL)/6-31G* level and calculate the relative energies of isomeric > systems at this level with a zero point energy correction from a HF/6-31G* > level calculation. We have found a handful of systems where two tautomers > exist on the potential energy surface differing only by ~ 1-3 kcal/mol. > Most of these tautomeric pairs are relatively similar structurally except > for the placement of a single hydrogen. > > Question 1: is the level of theory we are operating at accurate enough to > characterize these energy differences as being significant? > > Question 2: if these energy differences are meaningful (or if we can do > them at a level that is more reliable), is there a way to extract out from > the calculations why one tautomer is more stable than the other? > > Thanks in advance, > Joseph W. Bausch > bausch@rs6chem.chem.vill.edu > > > > > --- > Administrivia: This message is automatically appended by the mail exploder: > CHEMISTRY@www.ccl.net: Everybody | CHEMISTRY-REQUEST@www.ccl.net: Coordinator > MAILSERV@www.ccl.net: HELP CHEMISTRY or HELP SEARCH | Gopher: www.ccl.net 73 > Anon. ftp: www.ccl.net | CHEMISTRY-SEARCH@www.ccl.net -- archive search > Web: http://www.ccl.net/chemistry.html > --- > > From robert@pauli.utmb.edu Fri May 2 16:40:18 1997 Received: from nmr.utmb.edu for robert@pauli.utmb.edu by www.ccl.net (8.8.3/950822.1) id QAA28973; Fri, 2 May 1997 16:20:40 -0400 (EDT) Received: from pauli.utmb.edu by nmr.utmb.edu via SMTP (931110.SGI/930416.SGI.AUTO) for chemistry@www.ccl.net id AA21598; Fri, 2 May 97 15:15:06 -0600 Received: (from robert@localhost) by pauli.utmb.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) id PAA22396; Fri, 2 May 1997 15:13:31 -0500 From: "Robert Fraczkiewicz" Message-Id: <9705021513.ZM22394@pauli.utmb.edu> Date: Fri, 2 May 1997 15:13:31 -0500 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: chemistry@www.ccl.net Subject: Release of RAMVIB: software for modeling molecular vibrations Cc: jkl@ccl.net Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Dear Netters, I am pleased to announce the availability of the RAMVIB software package, free for academic users. RAMVIB is a set of FORTRAN programs that I wrote some time ago to ease pains of interpreting resonance Raman spectra of metalloproteins and metalloporphyrins. It uses normal mode analysis approach to model vibrations of molecules. RAMVIB can be downloaded from CCL archives: ftp://www.ccl.net/pub/chemistry/software/SGI/ramvib/ What makes RAMVIB different from existing NMA software? The partial answer to this question is listed below: 1) Automatic removal of coordinate redundancies by a new algorithm. 2) A novel method of solving the inverse spectral problem (a.k.a. force field refinement) that is convergent even in the case of strong correlation among force constants. It delivers high quality fits due to the regularization error reduction. 3) It contains an unique set of diagnostic programs for the purpose of testing numerical condition of input matrices to eliminate mathematical singularities and perform selection of force constants. 4) An input can be constructed for any combination of isotopomers and structurally similar molecules as long as they share common subset of force constants. 5) It calculates significance matrices of frequencies, isotope shifts and potential energy distributions with respect to force constants (as a part of the initial process of force constants selection). 6) An automatic force constant selection is performed using a new concept of "fitting potential". 7) Automatic sorting and matching of normal modes for different isotopomers. 8) Output of atomic displacement vectors and motion animation files in XYZ format readable by XMol to illustrate calculated normal modes. RAMVIB works under SGI IRIX 5.2 and 6.2 and uses three functions specific for SGI FORTRAN: iargc() - number of command line arguments getarg() - returns command line argument in a string variable etime() - returns elapsed CPU time It can be ported to other platforms after these functions are replaced by appropriate local equivalents. The paper describing all details of the package has been accepted for publication in Journal of Molecular Structure and is currently in press. I will post the exact reference. Robert Fraczkiewicz robert@nmr.utmb.edu ----------------------------------- University of Texas Medical Branch Galveston, TX 77555