From mn1@helix.nih.gov Thu Jul 23 11:08:57 1998 Received: from helix.nih.gov (helix.nih.gov [128.231.2.3]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id LAA17098 Thu, 23 Jul 1998 11:08:57 -0400 (EDT) Received: (from mn1@localhost) by helix.nih.gov (8.8.5/8.8.5) id LAA05372; Thu, 23 Jul 1998 11:08:56 -0400 (EDT) Date: Thu, 23 Jul 1998 11:08:56 -0400 (EDT) From: "M. Nicklaus" Message-Id: <199807231508.LAA05372@helix.nih.gov> To: chemistry@www.ccl.net Subject: Enhanced NCI Database Browser Cc: mn1@helix.nih.gov Enhanced NCI Database Browser ============================= This is to announce the public availability of the Enhanced NCI Database Browser. After the beta test has been concluded, 246182 structures of the open part of the NCI structural database are available for searching through a Web Browser interface. Two servers that mirror each other have been set up, one in the U.S., and one in Europe. Their URLs are: http://www2.ccc.uni-erlangen.de/ncidb/ for Europe, http://cactvs.cit.nih.gov/ncidb/ for the U.S. A total of 246,182 structures (about twice the size of earlier catalogs) are searchable by formula, name (where present), CAS RN (for about 50% of the compounds), AIDS and cancer screening data (where available), substructures and numerous other criteria. SMILES notation is used both for input of (sub-) structures and in the output of the hit lists. A Java structure editor, kindly provided by Peter Ertl of Novartis Crop Protection, allows to draw structures and structure fragments used in the searches. Structure export to plug-ins and external viewers is supported in a variety of 2D and 3D formats. If needed, 3D coordinates are computed by a CORINA structure generator module. The database engine behind the Enhanced NCI Database Browser is the toolkit CACTVS (http://schiele.organik.uni-erlangen.de/cactvs/index.html). The set of structures is essentially the same as available as a SMILES file that can be downloaded from the Web site of the Developmental Therapeutics Program, DCTDC, National Cancer Institute (http://epnws1.ncifcrf.gov:2345/dis3d/3ddatabase/nci_smil.html). Certain features of the NCI Database Browser are known to cause problems with Internet Explorer (IE), particularly when using the structure editor. These difficulties are not a bug of the system but are related to the way IE handles JavaScript, and are therefore not under our control. This is a collaborative effort between Wolf-Dietrich Ihlenfeldt of Computer Chemistry Center, University of Erlangen-Nuernberg, Erlangen, Germany, and the Laboratory of Medicinal Chemistry, National Cancer Institute, National Institutes of Health, Bethesda, USA. Please send bug reports, and any other feedback, to Wolf-Dietrich Ihlenfeldt (Wolf-Dietrich.Ihlenfeldt@CCC.Chemie.Uni-Erlangen.DE), and a CC to Marc C. Nicklaus (mn1@helix.nih.gov). Wolf-Dietrich Ihlenfeldt Marc C. Nicklaus Computer Chemistry Center, Lab. of Medicinal Chemistry, University of Erlangen-Nuernberg National Cancer Institute, NIH D-91052 Erlangen (Germany) Bethesda, MD 20892 (USA) ------------------------------------------------------------------------ Marc C. Nicklaus National Institutes of Health E-mail: mn1@helix.nih.gov Bldg 37, Rm 5B29 Phone: (301) 402-3111 37 Convent Dr, MSC 4255 Fax: (301) 496-5839 BETHESDA, MD 20892-4255 USA Laboratory of Medicinal Chemistry, National Cancer Institute, & Center for Molecular Modeling, Ctr. for Information Technology, NIH ------------------------------------------------------------------------ From tsonchev@skin6.chem.wisc.edu Thu Jul 23 12:32:29 1998 Received: from skin6.chem.wisc.edu (skin6.chem.wisc.edu [128.104.70.254]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id MAA18429 Thu, 23 Jul 1998 12:32:29 -0400 (EDT) Received: by skin6.chem.wisc.edu; id AA19644; Thu, 23 Jul 1998 11:32:43 -0500 Date: Thu, 23 Jul 1998 11:32:43 -0500 Message-Id: <9807231632.AA19644@skin6.chem.wisc.edu> From: Evelina Tsoncheva To: chemistry@www.ccl.net Subject: Help on using CHARMM needed I have to use the CHARMM softwear for my current project. I am not familiar with the program, and since the .doc files submitted with the code provide very few, simple examples, I believe that it would be very helpful for my work if somebody could send me some input files for CHARMM runs (dynamics, collecting trajectories, calculating correlation fxns and average bond lengths/angles. If this is not possible, I would appreciate any other help. Please, contact me at tsonchev@chem.wisc.edu Thanks. Evelina Tsoncheva From fparnold@balihai.uchicago.edu Thu Jul 23 12:46:56 1998 Received: from balihai.uchicago.edu (balihai.uchicago.edu [128.135.136.49]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id MAA18703 Thu, 23 Jul 1998 12:46:55 -0400 (EDT) Received: from localhost by balihai.uchicago.edu via SMTP (950413.SGI.8.6.12/931108.SGI.ANONFTP) for id LAA03590; Thu, 23 Jul 1998 11:46:56 -0500 Date: Thu, 23 Jul 1998 11:46:56 -0500 (CDT) From: "Fred P. Arnold" To: Computational Chemistry List Subject: Segmented Contractions: How To Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, This is probably a trivial question, but given an uncontracted gaussian basis set, what is the best way to go about constructing a segmented contraction, suitable for use with standard programs such as GAMESS or G94? I have a medium(?) sized basis set for 4th row main-group elements (Stromberg, A., et al., J. Comp. Chem., 1983, 4, 181), and I would like to arrive at a suitable contraction so that it could be used with the DZ set already present in GAMESS. I realize that I can get a feel for which exponents should be in which contractions by looking at the coefficients of the functions from an uncontracted calculation, but I'm a bit stuck where to go next. Thank you for your time. -Fred Frederick P. Arnold, Jr. "...But if you give theorists enough Advanced Research Systems time, they can predict anything." 5640 S. Ellis Ave Dr. M. S. Turner, U. Chicago. Chicago, IL 60637 From elewars@alchemy.chem.utoronto.ca Thu Jul 23 12:49:39 1998 Received: from alchemy.chem.utoronto.ca (alchemy.chem.utoronto.ca [142.150.224.224]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id MAA18759 Thu, 23 Jul 1998 12:49:39 -0400 (EDT) Received: (from elewars@localhost) by alchemy.chem.utoronto.ca (8.7.4/8.7.3) id MAA12609 for chemistry@www.ccl.net; Thu, 23 Jul 1998 12:49:37 -0400 (EDT) Date: Thu, 23 Jul 1998 12:49:37 -0400 (EDT) From: "E. Lewars" Message-Id: <199807231649.MAA12609@alchemy.chem.utoronto.ca> To: chemistry@www.ccl.net Subject: TWO-ELECT INTEGRALS, SUGGESTIONS Thurs 1998 July 23 Here are some suggestions I received about the two-electron integrals. I thank *everyone* who responded. E. Lewars ========================== The question was: Monday July 1998 Hello, Comparing some two-electron integrals in the book "Modern Quantum Chemistry", by Szabo and Ostlund (1989) with Gaussian 92 output I find what seem to be discrepancies: ------------ H2 molecule STO-3G bond length = 1.4 a.u. (0.741 A) I give the integrals in the order and the approximate format in which they appear in the book and in the G92 output: Book, page 162 G92 I J K L (11|11)=(22|22) = 0.7746 2 2 2 2 0.7746 (11|22) = 0.5697 2 2 2 1 0.4441 (21|11)=(22|21) = 0.4441 2 1 2 1 0.1485 (21|21) = 0.2970 2 2 1 1 0.2848 What is supposed to correspond to the Book's 0.5697 and 0.2970?? ----------------- HeH+ molecule STO-3G bond length = 1.4632 a.u. (0.7743 A) Book, page 172 G92 (11|11) = 1.3072 (22|11) = 0.6057 2 2 2 2 1.0557 2 2 1 1 0.5908 (21|11) = 0.4373 (22|21) = 0.3118 2 2 2 1 0.4440 2 1 1 1 0.3674 (21|21) = 0.1773 (22|22) = 0.7746 2 1 2 1 0.2243 1 1 1 1 0.7746 What is supposed to match 1.3072? and 0.6057? ....etc.? -------------------- I'm sure I may just be overlooking something _very simple_. But on the surface the numbers do not agree. Is one set of values wrong? Thanks E. Lewars ============================ SUGGESTIONS: Jul 14 Darko Babic (45) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 2/5 from Darko Babic Jul 14 '98 at 9:45 am X-Sender: dbabic@rudjer.irb.hr Date: Tue, 14 Jul 1998 09:45:24 +0200 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? Dear Prof. Lewars, At 17:56 1998.07.13 -0400, you wrote: > Book, page 162 G92 > I J K L >(11|11)=(22|22) = 0.7746 2 2 2 2 0.7746 >(11|22) = 0.5697 2 2 2 1 0.4441 >(21|11)=(22|21) = 0.4441 2 1 2 1 0.1485 >(21|21) = 0.2970 2 2 1 1 0.2848 > >What is supposed to correspond to the Book's 0.5697 and 0.2970?? Note that the Book's 0.5697 is (practically) exactly 2 x 0.2848 (G92) and that the Book's 0.2970 is exactly 2 x 0.1485 (G92), and that it holds just for the 'nondegenerate' integrals. So there must be some multiplication factor involved. However, this does not explain the other case (HeH+). Since the dif- ferences are present wherever He-orbital is involved, I would check if it is really equally represented in both cases (the Book and G92). Sincerely, Darko Babic Institute "Rudjer Boskovic" HR-10001 Zagreb, P.O.B.1016 Croatia ================ Jul 14 Javier Modrego (50) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 3/5 from Javier Modrego Jul 14 '98 at 11:31 am Date: Tue, 14 Jul 1998 11:31:14 +0200 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? At 17:56 -0400 13/7/1998, E. Lewars wrote: >Monday July 1998 > >Hello, > >Comparing some two-electron integrals in the book "Modern Quantum Chemistry", >by Szabo and Ostlund (1989) with Gaussian 92 output I find what seem to be >discrepancies: > >------------ > H2 molecule STO-3G bond length = 1.4 a.u. (0.741 A) > >I give the integrals in the order and the approximate format in which they >appear in the book and in the G92 output: > > Book, page 162 G92 > I J K L >(11|11)=(22|22) = 0.7746 2 2 2 2 0.7746 >(11|22) = 0.5697 2 2 2 1 0.4441 >(21|11)=(22|21) = 0.4441 2 1 2 1 0.1485 >(21|21) = 0.2970 2 2 1 1 0.2848 > >What is supposed to correspond to the Book's 0.5697 and 0.2970?? >----------------- Well, 0.5697 seems twice 0.2848 and 0.2970 two times 0.1485 but I have not a clue about the other set of numbers. Javier =============== Jul 14 Pedro Salvador Sed (91) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 4/5 from Pedro Salvador Sedano Jul 14 '98 at 11:59 am Message-ID: <35AB8E5C.81554C26@iqc.udg.es> Date: Tue, 14 Jul 1998 11:59:08 -0500 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? Hi, You must use the keyword NORAFF if you want Gaussian to print the 'pure' [I did use NORAFF--EL] TWO ELECTRON INTEGRALS, RESPONSES 1998 July Jul 14 Darko Babic (45) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 2/5 from Darko Babic Jul 14 '98 at 9:45 am X-Sender: dbabic@rudjer.irb.hr Date: Tue, 14 Jul 1998 09:45:24 +0200 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? Dear Prof. Lewars, At 17:56 1998.07.13 -0400, you wrote: > Book, page 162 G92 > I J K L >(11|11)=(22|22) = 0.7746 2 2 2 2 0.7746 >(11|22) = 0.5697 2 2 2 1 0.4441 >(21|11)=(22|21) = 0.4441 2 1 2 1 0.1485 >(21|21) = 0.2970 2 2 1 1 0.2848 > >What is supposed to correspond to the Book's 0.5697 and 0.2970?? Note that the Book's 0.5697 is (practically) exactly 2 x 0.2848 (G92) and that the Book's 0.2970 is exactly 2 x 0.1485 (G92), and that it holds just for the 'nondegenerate' integrals. So there must be some multiplication factor involved. However, this does not explain the other case (HeH+). Since the dif- ferences are present wherever He-orbital is involved, I would check if it is really equally represented in both cases (the Book and G92). Sincerely, Darko Babic Institute "Rudjer Boskovic" HR-10001 Zagreb, P.O.B.1016 Croatia ================ Jul 14 Javier Modrego (50) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 3/5 from Javier Modrego Jul 14 '98 at 11:31 am Date: Tue, 14 Jul 1998 11:31:14 +0200 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? At 17:56 -0400 13/7/1998, E. Lewars wrote: >Monday July 1998 > >Hello, > >Comparing some two-electron integrals in the book "Modern Quantum Chemistry", >by Szabo and Ostlund (1989) with Gaussian 92 output I find what seem to be >discrepancies: > >------------ > H2 molecule STO-3G bond length = 1.4 a.u. (0.741 A) > >I give the integrals in the order and the approximate format in which they >appear in the book and in the G92 output: > > Book, page 162 G92 > I J K L >(11|11)=(22|22) = 0.7746 2 2 2 2 0.7746 >(11|22) = 0.5697 2 2 2 1 0.4441 >(21|11)=(22|21) = 0.4441 2 1 2 1 0.1485 >(21|21) = 0.2970 2 2 1 1 0.2848 > >What is supposed to correspond to the Book's 0.5697 and 0.2970?? >----------------- Well, 0.5697 seems twice 0.2848 and 0.2970 two times 0.1485 but I have not a clue about the other set of numbers. Javier =============== Jul 14 Pedro Salvador Sed (91) Re: CCL:G:TWO ELECTRON INTEGRAL ERRCommand: Read MessageMessage 4/5 from Pedro Salvador Sedano Jul 14 '98 at 11:59 am Message-ID: <35AB8E5C.81554C26@iqc.udg.es> Date: Tue, 14 Jul 1998 11:59:08 -0500 To: "E. Lewars" Subject: Re: CCL:G:TWO ELECTRON INTEGRAL ERROR? Hi, You must use the keyword NORAFF if you want Gaussian to print the 'pure' [I did use NORAFF--EL] From smb@smb.chem.niu.edu Thu Jul 23 13:12:43 1998 Received: from cz2.chem.niu.edu (cz2.chem.niu.edu [131.156.9.10]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id NAA19070 Thu, 23 Jul 1998 13:12:42 -0400 (EDT) Received: from cz.chem.niu.edu by cz2.chem.niu.edu via SMTP (920330.SGI/920502.SGI.AUTO) for CHEMISTRY@www.ccl.net id AA18000; Thu, 23 Jul 98 11:22:57 -0500 Received: from smb.chem.niu.edu by cz.chem.niu.edu via SMTP (920330.SGI/890607.SGI) (for @cz2.chem.niu.edu:CHEMISTRY@www.ccl.net) id AA29610; Thu, 23 Jul 98 12:22:16 -0500 Received: from smb by smb.chem.niu.edu via SMTP (940816.SGI.8.6.9/940406.SGI) for id MAA26563; Thu, 23 Jul 1998 12:09:45 -0500 Sender: smb@smb.chem.niu.edu Message-Id: <35B76E58.794B@smb.chem.niu.edu> Date: Thu, 23 Jul 1998 12:09:44 -0500 From: Steven Bachrach X-Mailer: Mozilla 3.01 (X11; I; IRIX 5.3 IP12) Mime-Version: 1.0 To: computational chemistry list Subject: ChemInt 98 - Poster Session Added Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit A poster session has been added to the Chemistry and the Internet Conference (ChemInt98). Posters will be presented both Sunday and Monday after lunch. There will be just one poster session - all posters will be available at both times. Contributors who wish to show a poster at the meeting should email the title, the authors names, their affiliation and an abstract to Steven Bachrach, smb@smb.chem.niu.edu, prior to August 31, 1998. Authors should also indicate what medium they will be using for their poster, i.e. paper, computer, etc. The conference will be held from Sep. 12-15, 1998 in Irvine CA. For complete information, see http://www.ijc.com/ci1/ Steve Bachrach -- Steven Bachrach Department of Chemistry Northern Illinois University DeKalb, Il 60115 Phone: (815)753-6863 smb@smb.chem.niu.edu Fax: (815)753-4802 From Harald.Lanig@ccc.uni-erlangen.de Wed Jul 22 02:46:00 1998 Received: from faui45.informatik.uni-erlangen.de (root@faui45.informatik.uni-erlangen.de [131.188.2.45]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id CAA02415 Wed, 22 Jul 1998 02:45:58 -0400 (EDT) Received: from as1200.organik.uni-erlangen.de (as1200-1.organik.uni-erlangen.de [131.188.128.6]) by faui45.informatik.uni-erlangen.de (8.9.1/8.1.41-FAU) with ESMTP id IAA00056 for ; Wed, 22 Jul 1998 08:45:56 +0200 (MET DST) Received: from ccc.uni-erlangen.de (indigo11.organik.uni-erlangen.de [131.188.128.69]) by as1200.organik.uni-erlangen.de (8.8.7/8.1.11-FAU) with ESMTP id IAA20217; Wed, 22 Jul 1998 08:44:06 +0200 (MET DST) Sender: Harald.Lanig@ccc.uni-erlangen.de Message-ID: <35B58A9D.1EE8DE7@ccc.uni-erlangen.de> Date: Wed, 22 Jul 1998 08:45:49 +0200 From: Harald Lanig Organization: Computer-Chemie-Centrum Uni Erlangen/Nbg X-Mailer: Mozilla 4.04 [en] (X11; I; IRIX 5.3 IP22) MIME-Version: 1.0 To: CHEMISTRY@www.ccl.net, bear@ellington.Pharmacy.arizona.edu Subject: Summary: Zn within AMBER Content-Type: multipart/mixed; boundary="------------B381BA6FFFB59F6E893E0C01" This is a multi-part message in MIME format. --------------B381BA6FFFB59F6E893E0C01 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit -- --------------------------------------------------------------------- Dr. Harald Lanig Computer Chemie Centrum der Universitaet Erlangen/Nuernberg Institut fuer Organische Chemie I, Naegelsbachstr. 25 D-91052 Erlangen, Germany Phone +49(0)9131-85 6581 Fax -85 6565 mailto:lanig@ccc.uni-erlangen.de http://www.ccc.uni-erlangen.de/clark/lanig --------------------------------------------------------------------- --------------B381BA6FFFB59F6E893E0C01 Content-Type: text/plain; charset=us-ascii; name="znenz.txt" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="znenz.txt" Dear Netters, here are the (unchanged) answers to my question about Zn Parameters for AMBER. Thanks to everybody who replied to me. -Harry The original query: Dear netters, optimizing metalloenzymes (e.g. Zn) with AMBER needs parameters not included in the original 95 parameter set. Has anybody out there experiences about how to perform such calculations? Any help is highly appreciated! The answers: 111111111111111111111111111111111111111111111111111111111111111111111 Takanori Kanazawa takanori.kanazawa@pharma.Novartis.com The following literature may help you. (In the first two articles, modified version of AMBER force-field is used.) 1. W.C.Guida et al. J.Comp.Chem. 13(2),214-228(1992) 2. C.McMartin et al. J.Comp-Aided.Mol.Design 11,333-344(1997) 3. S.C.Hoops et al. JACS 113,8262-8270(1991) "http://www.amber.ucsf.edu/amber/0Net/ions" also might help you. 2222222222222222222222222222222222222222222222222222222222222222222222 Ulf Ryde Ulf.Ryde@teokem.lu.se Dear Dr. Lanig, I have performed quite a few simulations of metalloproteins (Cu, Zn) with Amber as well as with other programs, and I have also performed some studies of parameterisations and charge estimations nessesary for such studies. Some of my relevant references: U. Ryde (1995) Molecular dynamic simulations of alcohol dehydrogenase with a varying coordination number of the catalytic zinc ion. Proteins. Struct. Funct. Genet. 21, 40-56. U. Ryde (1995) On the role of Glu-68 in alcohol dehydrogenase. Protein Science 4, 1124-1132. U. Ryde (1996) The coordination of the catalytic zinc ion in alcohol dehydrogenase studied by combined quantum chemical and molecular mechanical calculations. J. Comput.-Aided Mol Design 10, 153-164. K. Pierloot, J. O. A. De Kerpel, U. Ryde & B. O. Roos (1997) Theoretical study of the electronic spectrum of plastocyanin. J. Am. Chem. Soc. 119, 218-226. E. Sigfridsson & U. Ryde (1998) A comparison of methods for deriving atomic charges from the electrostatic potential and moments. J. Comp. Chem. 19, 377-395. J. O. A. De Kerpel, K. Pierloot, U. Ryde & B. O. Roos (1998) Theoretical study of the structural and spectroscopic properties of stellacyanin. J. Phys. Chem. B, 102, 4638-4647. J. O. A. De Kerpel & U. Ryde (1998) Protein strain in blue copper proteins studied by free energy perturbations Proteins. Struct. Funct. Genet., submitted Some other references on Zn: E Clementi et al. J. Chem. Phys 72(80)260 Non-bonded potential for Zn with H2O and CO2 A Vedani et al. J. Comp. Chem. 7(86)701 A Vedani & DW Huhta, J. Am. Chem. Soc. 112(90)4759 Program Yeti; a special bonded potential for S, N and O ligands; Simulations of carbonic anhydrase R(Zn)=0.69 A; eps(Zn)=-0.014 kcal/mole KM Merz et al. J. Am. Chem. Soc. 113(91)406 SC Hoops et al. J. Am. Chem. Soc. 113(91)8262 KM Merz & L Banci, J. Phys. Chem. 100(96)17414 KM Merz & L Banci, J. Am. Chem. Soc. 119(97)863 Simplified bonded potential for Amber with His, H2O and OH-; Simulations of carbonic anhydrase R(Zn)=1.1 A; eps(Zn)=0.0125 kcal/mole PMT de Kok et al. Recl. Trav. Pays-Bas 107(88)355 NA Beijer et al. Ann. Ny Ac. Sci. 377(91)494 MM studies with Amber on alcohol dehydrogenase R(Zn)=1.4 A; eps(Zn)=0.01 kcal/mole R Reddy, et al, J. Am. Chem. Soc., in press. FEP calculations with Amber on Adenosine Deaminase R*=1.1 A, eps=0.1 kcal/mole, q(Zn)=+1.2 3333333333333333333333333333333333333333333333333333333333333333333 John W. Cox che5jwc@titan.vcu.edu Dr. Lanig, I've done 2 years work on platinum-DNA containing systems, using a derivative of the AMBER parameter set. Adding all the parameters for Pt (including 2 new atom types, PT and NM) had to be done manually. I started adding them into the parameter set used by HyperChem, and later into Insight/Discover. The text references for HyperChem were a good place to start, there are a few pages describing the addition of new atom types and their corresponding values (stretch, bend, etc.). As far as the values that will make up the additions for your metalloenzymes, they may be available in the literature. I'm not directly familiar with Zn values. If this helps, and you would like more details, etc., feel free to email me directly. There is also a Chem. Reviews article, Zimmer, M., Chem. Rev., 1995, 95, 2629 (I think this is the first page), that is a good reference. --------------B381BA6FFFB59F6E893E0C01-- From d.w.price@reading.ac.uk Wed Jul 22 08:36:57 1998 Received: from sumh1.rdg.ac.uk (pp@sumh1.rdg.ac.uk [134.225.16.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id IAA04254 Wed, 22 Jul 1998 08:36:56 -0400 (EDT) Received: from indyp (actually host scsgind2) by sumh1.rdg.ac.uk; Wed, 22 Jul 1998 13:36:47 +0100 Received: from reading.ac.uk by indyp via ESMTP (940816.SGI.8.6.9/930416.SGI) for id NAA13041; Wed, 22 Jul 1998 13:36:45 +0100 Sender: dave@reading.ac.uk Message-ID: <35B5DCDC.CD59B4C7@reading.ac.uk> Date: Wed, 22 Jul 1998 13:36:45 +0100 From: Dave Price Organization: none-what-so-ever X-Mailer: Mozilla 4.05 [en] (X11; I; IRIX 5.3 IP22) MIME-Version: 1.0 To: Comp Chem Mailing List Subject: g94 format / basis set/ ECP problem Content-Type: multipart/alternative; boundary="------------CDC48D4F8CF6406056407DB2" --------------CDC48D4F8CF6406056407DB2 Content-Type: text/plain; charset="us-ascii" Dear Colleagues, I am trying to run some calculations with g94 on some uranium compounds and as a test I am running UO2_2+ with a basis set and ECP from W. Kuechle Diplomarbeit (reformatted by me correctly I hope). But I am getting a error- Isotopes: U-238,O-16,O-16 Standard basis: 6-31+G(d) (6D, 7F) WANTED A FLOATING POINT NUMBER AS INPUT. FOUND AN END-OF-LINE FOR INPUT. 17.25477 0.001398 ? Error termination via Lnk1e in /usr/software/g94/l301.exe. Job cpu time: 0 days 0 hours 0 minutes 1.4 seconds. File lengths (MBytes): RWF= 5 Int= 0 D2E= 0 Chk= 1 Scr= 1 While the input file is- $ RunGauss %Chk=UO2_gaussian # IOP1 = NoAssem MP4/6-31+G* ExtraBasis pseudo=read Test Opt=FP # OptCyc=50 SCF=Direct SCFCyc=500 UO2 trial with relativistic ECPs and basis set 2 3 U O 1 r2 O 1 r3 2 a3 r2 2.100000 r3 2.100000 a3 180.000000 U 0 S 4 1.000 12.125253 0.021921 7.161545 -0.225160 4.774836 0.560299 2.011693 -1.071209 S 1 1.000 0.586852 1.000000 S 1 1.000 0.279115 1.000000 S 1 1.000 0.063372 1.000000 S 1 1.000 0.025611 1.000000 SP 4 1.000 17.25477 0.001398 7.735356 -0.033346 5.155878 0.110578 2.241670 0.032130 SP 1 1.000 0.581858 1.000000 SP 1 1.000 0.267908 1.000000 SP 1 1.000 0.083442 1.000000 SP 1 1.000 0.032130 1.000000 D 3 1.000 4.841070 0.005731 2.160162 -0.057236 0.575630 0.238828 D 1 1.000 0.278136 1.000000 0.124879 1.000000 D 1 1.000 0.051548 1.000000 F 3 1.000 2.436441 0.355011 1.144682 0.400846 0.529693 0.304679 F 1 1.000 0.240596 1.000000 F 1 1.000 0.101867 1.000000 G 1 1.000 1.18 1.000000 G 1 1.000 0.42 1.000000 **** U 0 ECP78 4 78 l = 0 3 0 4.063653 112.920103 0 1.883995 15.647500 0 0.886567 -3.689971 l = 1 3 0 3.986181 118.758016 0 2.000160 15.077228 0 0.960841 0.556720 l = 2 3 0 4.147972 60.855892 0 2.234563 29.280047 0 0.913695 4.998029 l = 3 3 0 3.998938 49.924035 0 1.998840 -24.674042 0 0.995641 1.389480 l = 4 2 0 3.817422 -36.040977 0 0.262501 -0.090511 -- ------------------------------------------------------------------------ Dr. David W. Price, Tel: +44 (0)118 9875123 extn 7415 Department of Chemistry, Fax: +44 (0)118 9316331 University of Reading, mailto:d.w.price@reading.ac.uk Whiteknights, READING http://www.chem.rdg.ac.uk/g50/mmrg/dave/dave.html RG6 6AD U.K. ------------------------------------------------------------------------ --------------CDC48D4F8CF6406056407DB2 Content-Type: text/html; charset="us-ascii" Dear Colleagues,
            I am trying to run some calculations with g94 on some uranium compounds
and as a test I am running UO2_2+ with a basis set and ECP from W. Kuechle
Diplomarbeit (reformatted by me correctly I hope).  But I am getting a error-

 Isotopes: U-238,O-16,O-16
 Standard basis: 6-31+G(d) (6D, 7F)
  WANTED A FLOATING POINT NUMBER AS INPUT.
  FOUND AN END-OF-LINE FOR INPUT.
 17.25477      0.001398
                       ?
 Error termination via Lnk1e in /usr/software/g94/l301.exe.
 Job cpu time:  0 days  0 hours  0 minutes  1.4 seconds.
 File lengths (MBytes):  RWF=    5 Int=    0 D2E=    0 Chk=    1 Scr=    1

While the input file is-

$ RunGauss
%Chk=UO2_gaussian
# IOP1 = NoAssem MP4/6-31+G* ExtraBasis pseudo=read Test Opt=FP
# OptCyc=50 SCF=Direct SCFCyc=500

UO2 trial with relativistic ECPs and basis set

2 3
  U
  O    1      r2
  O    1      r3        2      a3

  r2        2.100000
  r3        2.100000
  a3      180.000000

U 0
S    4   1.000
12.125253     0.021921
7.161545     -0.225160
4.774836      0.560299
2.011693     -1.071209
S    1   1.000
0.586852      1.000000
S    1   1.000
0.279115      1.000000
S    1   1.000
0.063372      1.000000
S    1   1.000
0.025611      1.000000
SP   4   1.000
17.25477      0.001398
7.735356     -0.033346
5.155878      0.110578
2.241670      0.032130
SP   1   1.000
0.581858      1.000000
SP   1   1.000
0.267908      1.000000
SP   1   1.000
0.083442      1.000000
SP   1   1.000
0.032130      1.000000
D   3   1.000
4.841070      0.005731
2.160162     -0.057236
0.575630      0.238828
D   1   1.000
0.278136      1.000000
0.124879      1.000000
D   1   1.000
0.051548      1.000000
F   3   1.000
2.436441      0.355011
1.144682      0.400846
0.529693      0.304679
F   1   1.000
0.240596      1.000000
F   1   1.000
0.101867      1.000000
G   1   1.000
1.18          1.000000
G   1   1.000
0.42          1.000000
****

U 0
ECP78 4 78
l = 0
3
0   4.063653  112.920103
0   1.883995   15.647500
0   0.886567   -3.689971
l = 1
3
0   3.986181  118.758016
0   2.000160   15.077228
0   0.960841    0.556720
l = 2
3
0   4.147972   60.855892
0   2.234563   29.280047
0   0.913695    4.998029
l = 3
3
0   3.998938   49.924035
0   1.998840  -24.674042
0   0.995641    1.389480
l = 4
2
0   3.817422  -36.040977
0   0.262501   -0.090511
 
  

-- 
------------------------------------------------------------------------

Dr. David W. Price,       Tel: +44 (0)118 9875123  extn 7415
Department of Chemistry,  Fax: +44 (0)118 9316331
University of Reading,    mailto:d.w.price@reading.ac.uk
Whiteknights, 
READING                http://www.chem.rdg.ac.uk/g50/mmrg/dave/dave.html
RG6 6AD 
U.K.
 
------------------------------------------------------------------------
  --------------CDC48D4F8CF6406056407DB2-- From daizadeh@kappa.ucdavis.edu Wed Jul 22 11:29:56 1998 Received: from kappa.ucdavis.edu (kappa.ucdavis.edu [128.120.7.55]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id LAA13164; Wed, 22 Jul 1998 11:29:52 -0400 (EDT) Received: (from daizadeh@localhost) by kappa.ucdavis.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) id IAA02898; Wed, 22 Jul 1998 08:29:52 -0700 Resent-From: "Iraj Daizadeh" Resent-Message-Id: <9807220829.ZM2896@kappa.ucdavis.edu> Resent-Date: Wed, 22 Jul 1998 08:29:52 -0700 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) Resent-To: chemistry@www.ccl.net, chemistry-request@www.ccl.net Received: (from daizadeh@localhost) by kappa.ucdavis.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) id KAA05534; Thu, 16 Jul 1998 10:45:09 -0700 From: "Iraj Daizadeh" Message-Id: <9807161045.ZM5532@kappa.ucdavis.edu> Date: Thu, 16 Jul 1998 10:45:09 -0700 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: David van der Spoel , "Daniel E. Macks" , Dayong He , "Dr. Thomas Strassner" , Keith Refson , Kenneth Geisshirt , Nathalie Godbout , William Gropp , amber@cgl.ucsf.edu, chemistry@www.ccl.net, chemistry-request@www.ccl.net, crowley@scripps.edu, daizadeh@kappa.ucdavis.edu, fink@indigo.ucdavis.edu, medvedev@omega.ucdavis.edu, aas@omega.ucdavis.edu, guo@indigo.ucdavis.edu, leee@indigo.ucdavis.edu, antony@indigo.ucdavis.edu, stuchebr@indigo.ucdavis.edu Subject: Summary: Software on mpi or pvm. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hello. The contents of this summary are as follows: 1. Initial question concerning the use of MPI or PVM within the Chemistry community. 2. Replies. 3. I have asked Dr. W. Gropp, who is the Senior Computer Scientist at the Mathematics and and Computer Science Division of Argonne National Laboratory, to contribute to this discussion. Dr. Gropp has been with the MPI project at ANL since its genesis and is one of the authors of the very popular MPICH implementation of MPI. His comments have been appended to the end of this list. I should mention that the commonly used G94 program utilizes Linda instead of the standardized MPI or PVM systems. Those of you who would like to gain a greater understanding of MPI or PVM, please see the following URL sites: MPI: http://www-c.mcs.anl.gov/mpi/ PVM: http://www.netlib.org/pvm3/faq_html/faq.html http://www.epm.ornl.gov/pvm/ I would like to thank Dr. Gropp as well as those who replied to this query. It is my hope that this summary will be of use to others in the future. Iraj. -- Iraj Daizadeh Department of Chemistry University of California One Shields Ave. Davis, CA 95616-5295 Phone: 530.754.8695 Fax: 530.752.8995 email: daizadeh@kappa.ucdavis.edu XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 1. Hello. We have a couple of dec 500 au's (DEC unix 4.0c) and a sgi r10000 (6.2 irix) machine that we are considering parallelizing via mpi or pvm. I would like to know what chemistry software can utilize mpi or pvm. For example, it would be nice if there were dynamics packages such as Amber or Gromos that could use these tools; thus, one could run two classical trajectories simultaneously. If there is enough interest in this question, then I will post a summary of all results as is traditionally done on the CCL. Thank you in advance. Iraj. XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 2. Dear Iraj, AMBER is fully parallel using MPI or other native communications schemes (such as on Cray t3e where shmem is faster than MPI). PVM is no longer supported in AMBER. However, it is not parallel in the sense that one would run two classical traj. at once. That is called an "embarassingly parallel" application by the computer science people. To do that one would most likely write a shell script that would start up several instances of AMBER on separate processors. Rather, AMBER spreads the work of single traj. over many processors getting the work done that much faster. If you want more detail, please ask. Sincerely, Mike Crowley 000000000000000000000000000000000000000000000000000000000000000000000000000 Dear Iraj Daizadeh, daizad>I would like to know what chemistry software can utilize mpi or pvm. I know of a few chemistry software packages that use PVM and MPI. The MD programs Gromacs and DLpoly can use both, and the ab inition package GAMESS uses PVM. I have successfully compiled and runned all three on IBM SP computers, ordinary IBM RS/6k and Linux boxes. Best regards, Kenneth ----------------------------------------------------------- Kenneth Geisshirt, M.Sc., Ph.D. Computational Scientist Structural Bioinformatics Tel: +45 4516 2804 Advanced Technologies Fax: +45 4516 2802 http://www.strubix.com kneth@strubix.dk LOAD "http://www.sslug.dk",8,1 0000000000000000000000000000000000000000000000000000000000000000000000000000 Hello, The ADF (Amsterdam Density Functional) code uses PVM. I have installed it Origin 200 SGI computers (Irix 6.4). Nathalie -------------------------------------------- | Nathalie Godbout | | University of Illinois at Urbana-Champaign | | 600 S. Goodwin Av. Box 23.6 | | Urbana, IL 61801 | -------------------------------------------- 0000000000000000000000000000000000000000000000000000000000000000000000000000000 My MD simulation code Moldy will run in parallel using MPI. If you take a look at my web page http://www.earth.ox.ac.uk/~keith/moldy.html it contains links to other parallel MD codes including AMBER and GROMOS and DL_POLY. One piece of advice though. If your machines are connected to an unswitched ethernet shared by others in your department, forget it. You will at least need a switched network to get good parallel performance. This does depend on the parallelization strategy, but even a replicated-data approach as used in Moldy won't give much of a gain unless you have an essentially dedicated (ie switched) network. Keith Refson -- ------------------------------------------------------------------------------ Email: Keith.Refson@ | Tel: +44 1865 272026 | Dr Keith Refson, | earth.ox.ac.uk | Fax: +44 1865 272072 | Dept of Earth Sciences | Spam: root@cyberpromo.com | Parks Road, Oxford OX1 3PR, UK| ------------------------------------------------------------------------------ 0000000000000000000000000000000000000000000000000000000000000000000000000000000 See http://rugmd4.chem.rug.nl/~gmx/index.html Good luck. Dayong 0000000000000000000000000000000000000000000000000000000000000000000000000000000 We're using CHARMM built over PVM, and it (claims it:) can just as easily be configured to use MPI instead. dan -- Daniel Macks dmacks@a.chem.upenn.edu dmacks@netspace.org http://www.netspace.org/~dmacks 0000000000000000000000000000000000000000000000000000000000000000000000000000000 Hi, check out the GROMACS package for MD and trajectory analysis. It runs with either PVM or MPI on (homogeneous) cluster of workstations. http://rugmd0.chem.rug.nl/~gmx Groeten, David. ________________________________________________________________________ Dr. David van der Spoel Biomedical center, Dept. of Biochemistry s-mail: Husargatan 3, Box 576, 75123 Uppsala, Sweden e-mail: spoel@xray.bmc.uu.se www: http://zorn.bmc.uu.se/~spoel phone: 46 18 471 4205 fax: 46 18 511 755 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 3. At 02:36 PM 7/14/98 -0700, Iraj Daizadeh wrote: >Hello Dr. Gropp. > >I have another question I would like to raise if you have time. > >I have recently submitted a query to the Computational Chemistry List (CCL) >concerning the topic of MPI and PVM (see below) and its use in Chemistry >software. I will submit to them a summary of all responses that I receive >concerning the question. Since you are an authority in this field, I would >like to append onto this summary any information that you could kindly supply >to me concerning MPI. > MPI (particularly MPI-1) is most suited to massively parallel applications that run with a fixed number of processes; it is particular good for applications requiring good communications performance for both point-to-point and collective operations. It isn't appropriate for most task-farm approaches. PVM is more appropriate for applications that use changing numbers of processes. MPI-2 adds a number of features, such as dynamic processes and parallel I/O. To date, only the parallel I/O is widely available. Bill 000000000000000000000000000000000000000000000000000000000000000000000000000000 At 02:25 PM 7/14/98 -0700, Iraj Daizadeh wrote: >Hello. > >We are considering parallizing two DEC 500 a.u.'s (DEC UNIX 4.0c) and an sgi >r10000 (irix 6.2), and were wondering if this was a worth while venture when >these machines are on an ethernet connection that is used by others. > It really depends on your application, but my guess would be that it is not worthwhile. Because of load balancing and the low bandwidth/high latency of Ethernet, any application that would work would be easily written with rsh or ssh directly. Bill XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX From rich@censa.org Wed Jul 22 15:02:47 1998 Received: from venus.star.net (root@venus.star.net [199.232.114.5]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id PAA01538 Wed, 22 Jul 1998 15:02:46 -0400 (EDT) From: rich@censa.org Received: from www.censa.org (censa2.censa.org [199.232.115.57]) by venus.star.net (8.8.5/8.8.5) with SMTP id OAA29599; Wed, 22 Jul 1998 14:24:06 -0400 Received: by www.censa.org(Lotus SMTP MTA SMTP v4.6 (462.2 9-3-1997)) id 85256649.0064E422 ; Wed, 22 Jul 1998 14:22:00 -0400 X-Lotus-FromDomain: CENSA To: World_Archivists_and_Records_Managers_Listservs@www.censa.org, World_Editor_E-Mailing_List@www.censa.org, World_Pharma_Records_Manager_Interest_List@www.censa.org, World_Science_ListServs_Interest_List@www.censa.org, World_SLA_Interest_List@www.censa.org, World_User_Interest_List@www.censa.org, World_Vendor_Interest_List@www.censa.org Message-ID: <85256649.0063969C.00@www.censa.org> Date: Wed, 22 Jul 1998 14:21:55 -0400 Subject: Deadline rapidly approaching for Electronic Lab Notebooks Course offered at the American Chemical Society Meeting in Boston Mime-Version: 1.0 Content-type: text/plain; charset=us-ascii Dear Colleague, The deadline is quickly approaching for course preregistration for the American Chemical Society (ACS) short course covering electronic notebooks, groupware, document management, and related systems. This details of exciting course are given below. It is being held in Boston, one of the world's most exciting cities! Please register directly with the ACS at http://www.acs.org/education/profdevl/sccm01.html. To receive the preregistration discount, please register within the next two weeks. Thank you. =========================================================================== === "Electronic Lab Notebooks and R&D Team Computing Systems" Who Should Attend Scientists, managers, engineers, MIS and computer support personnel, regulatory staff in the chemical, pharmaceutical, environmental, biotech, food and beverage, and related areas. This course will be of interest to anyone needing to know the legal, regulatory, technical, and social aspects of electronic notebooks, recordkeeping, document management, or other collaborative systems used in R&D or testing labs. How You'll Benefit from This Course Stay current with latest technologies, regulations, and legal imperatives for laboratory recordkeeping. Quickly assess your company's infrastructure and cultural readiness to use electronic lab notebooks; learn how to prepare them to use electronic lab notebooks. Learn when to buy versus build electronic notebook systems; how to grow them into full-blown R&D Team Computing Systems. Get practical tips for evaluating and selecting vendors and systems. See key products evaluated and demonstrated without vendor hype. Course Topics Introduction to Collaborative Electronic Lab Notebooks and R&D Team Computing Systems: legal and patent aspects; regulatory requirements and procedures; technical design concepts; social and cultural aspects. Electronic Recordkeeping and Records Management Systems good recordkeeping practices; the 13 fundamental properties of electronic records; electronic records, signatures, notary and security systems for protecting intellectual property; acceptable optical and magnetic storage media for electronic records. Practical Aspects of Applying The Technologies: understanding the costs and benefits of electronic notebooks and team computing systems in your lab; implementing systems to meet legal, regulatory, technical, and social needs; technical requirements: basic and advanced design & usability issues; connecting existing instruments, LIMS, other automation systems; end-user experiences and case studies; addressing social and cultural issues; methodologies to ensure successful pilot projects and system rollouts for small to large organizations; augmenting team intelligence with team computing systems. The Instructors Richard Lysakowski, Ph.D. is Executive Director, Collaborative Electronic Notebook Systems Association (CENSA), Woburn, Massachusetts. James T. Currie Jr., Ph.D. is an instructor, researcher, and engineer for CENSA. Apologies in advance for any duplications in the mailing lists. From rino@gpc.ibc.wustl.edu Wed Jul 22 17:29:12 1998 Received: from wugate.wustl.edu (wugate.wustl.edu [128.252.120.1]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id RAA10925 Wed, 22 Jul 1998 17:29:12 -0400 (EDT) Received: from [128.252.137.150] ([128.252.137.150]) by wugate.wustl.edu (8.8.8/8.8.5) with ESMTP id QAA20461 for ; Wed, 22 Jul 1998 16:21:56 -0500 (CDT) Date: Wed, 22 Jul 1998 16:21:56 -0500 (CDT) Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: chemistry@www.ccl.net From: Rino Ragno Subject: molecule fragmentation code Hi to everyone, I am looking for a code to fragment a given molecule into chunks. The fragments should have the breaking point in common. Is any of you aware of such a piece of code (better if in C)? Thanks to all. -- Rino Ragno ++----------------------------------------------------------------------++ ++----------------------------------------------------------------------++ || || || Dr. Rino Ragno E-mail: ragno@uniroma1.it || || Institute for Biomedical Computing or: rino@gpc.wustl.edu || || Center for Molecular Design or: ragno@axcasp.caspur.it || || Box8036, Washington University Phone : 314-362-2272 || || 700 South Euclid Avenue FAX : 314-362-0234 || || St. Louis, Missouri 63110 || || U. S. A. || || || || WEB PAGE : www.ibc.wustl.edu/cmd/people/rino/rino.html || || || ++----------------------------------------------------------------------++ ++----------------------------------------------------------------------++ From mckelvey@ncsa.uiuc.edu Wed Jul 22 18:33:52 1998 Received: from ex1.ncsa.uiuc.edu (ex1.ncsa.uiuc.edu [141.142.3.16]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id SAA11597 Wed, 22 Jul 1998 18:33:52 -0400 (EDT) Received: from mckelvey (tollfree1.ncsa.uiuc.edu [141.142.8.48]) by ex1.ncsa.uiuc.edu (8.9.1/8.8.8) with SMTP id RAA10584; Wed, 22 Jul 1998 17:33:46 -0500 (CDT) Message-Id: <3.0.5.32.19980722173515.009b3b60@pop.ncsa.uiuc.edu> X-Sender: mckelvey@pop.ncsa.uiuc.edu X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.5 (32) Date: Wed, 22 Jul 1998 17:35:15 -0700 To: ep7@dent.okayama-u.ac.jp From: John McKelvey Subject: Re: CCL:G:Summary: X,Yx,Z vs. internal coordinates Cc: CHEMISTRY@www.ccl.net In-Reply-To: <199807220011.JAA13768@deews1.dent.okayama-u.ac.jp> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" X-NCSA-md5: e041c7a3c804a06c814eed2596bbffce Content-MD5: 4EHHo8gEoGyBTu0llrv/zgAA Content-MD5-Origin: ex1.ncsa.uiuc.edu Folks, Could I throw in my 2-cents worth at this point. I agree with some of the submitted points, and all of my experience is covered already, in pieces, in the submitted discussion. I would like to render the following: I have done many calculations containing both low barrier modes (floppy phenyl groups) and polycyclic rings. I have developed a perspective that I found almost bullet-proof (submit the job and go on vacation). ... Using z-matrices 1)Optimize the geometry in Gaussian using PM3 2)Then optimize the geometry with HF/3-21G OPT=mndofc iop(4/8=20) [gets the PM3 Hessian at the PM3 geometry(clearly positive definite) and optimizes with 3-21G basis set] 3)Then optimize at next theory/basis set level... 4) etc.... I haven't done much with redundant internals along this line.. I do, however, _highly_ recommend steps similar to 1 and 2 above. Good initial guesses for initial torsin angles is quite important, and I have found PM3 to be far superior to Am1 and MNDO or MINDO/3 or CNDO/INDO after about 15 years of doing this sort of thing. Hope this is useful.. John McKelvey NCSA At 09:29 AM 7/22/98 +0900, ep7@dent.okayama-u.ac.jp wrote: >Dear Sir: > > I summarized x,y,z vs. internal coordinates. > >Masao Masamura >Preventive Dentistry >Okayama University Dental School >Shikata-cho, 2-5-1 >Okayama 700 >Japan >FAX: 81-86-235-6714 >e-mail: ep7@dent.okayama-u.ac.jp > > > > I ask about the following. > >(1) The optimization using x,y,z coordinates is faster than that using >internal coordinates. > >(2) Do you experience that the optimization using x,y,z coordinates >converges unexpected optimized structure ? The optimization using >internal coordinates is more safe than the optimization using x,y,z >coordinates ? > > Thank you. > > > >> (1) The optimization using x,y,z coordinates is faster than that using >> internal coordinates. > >No, generally not. Because in internal coord. you optimize 3N-6 >variables, whereas in xyz you have 3N. > >> >> (2) Do you experience that the optimization using x,y,z coordinates >> converges unexpected optimized structure ? The optimization using >> internal coordinates is more safe than the optimization using x,y,z >> coordinates ? > >Yes and no. For cyclic systems or bigger molecules, a little deviation >of one variable like an angle can cause big distortions of your >molecule. For example, since C1 and C6 in a 6-membered ring are not >"connected", changing the dihedral 2-3-4-5 by 50 causes a relativily >great change in bond length for C1-C6. > >There are more types of coordinates like the redundant internal >coordinates used in Gaussian. There is also a neverending debate about >the "right" ones. >See their manual at www.gaussian.com for references. > >-- > Andreas Goeller > >--------------------------------------------------------------- > Dr. Andreas Goeller Institut fuer Physikalische Chemie > Friedrich-Schiller-Universitaet > Lessingstr. 10 >phone: +49(0)-3641-948352 D-07743 Jena > fax: +49(0)-3641-948302 (secretary) Germany >email: goeller@pc04.chemie.uni-jena.de >http://www.uni-jena.de/chemie/photo/goeller/goeller.html >--------------------------------------------------------------- > Dr. Andreas Goeller ehemals Computer Chemie Centrum > >email: goeller@organik.uni-erlangen.de >http://www.organik.uni-erlangen.de/clark/goeller/goeller.html >--------------------------------------------------------------- > > > > I would like to ask about the following. > >(1) The redundant internal coordinates is more reliable ? >(2) What is www.gaussian.com ? >(3) The optimization using the redundant internal coordinates is faster >than the optimization using x,y,z coordinates ? > > > >> (1) The redundant internal coordinates is more reliable ? > >They say so. These coord. have the advantage of being redundant, i.e., >there are achieved additional coord. from your input., so the system can >change to other coord. during opt. >> (2) What is www.gaussian.comwww.gaussian.com ? >http://www.gaussian.com >> (3) The optimization using the redundant internal coordinates is faster >> than the optimization using x,y,z coordinates ? >More likely leeding to convergence I would think >> > > > >About the second item I have no experience, although it sounds a bit strange. >However, for the first one, I can say that generally it is the opposite and >optimization in the internal coordinates are much faster than cartesian >coordinates. For some special cases, however, it is not straightforward to set >up a suitable internal coordinate set which can be properly used during the >optimization and one has to use the XYZ coordinates (or redundant internal >coordinates, if implemented). > > > >I have heard arguments on both sides. One the one hand, >optimization steps taken in torsion space are thought by >many to follow more closely the natural low energy pathways >in molecules, or at least in proteins. This argument seems to >be backed up by recent numerical experiments in Ruben Abagyan's group: > >Abagyan R.A. and Totrov, M.M. J. Mol. Bio 235 (1994) 983 > >(which admittedly I have not read), in which they measure how >far one can move away from a local minimum and still return via >optimization. > >Others argue that by restricting the conformational search >to torsion space, the energy barriers are made higher >and the search space becomes "topologically" more complex >as a result. Note that the ECEPP forcefield used by Abagyan >has adjusted torsional barrier heights, supposedly to >compensate for this. > >Personally, I think this is comparing apples with oranges until >someone can perform an optimization in ALL internal coordinates >(including bends, stretches etc. but excluding redundant coordinates) >so that the number of degrees of freedom are the same. I suspect, >however, that internal coords will win for the reasons stated above. >Anybody know of a progam that can do FULL internal coordinate >optimization? > > >Richard Gillilan >Cornell Theory Center > > > >Cartesian <-> internal coordinates: >This question has to be answered with is clear >"It depends". > >Usually cartesian coordinates introduces a great >deal of undesired couplings. One may get rid of them >by choosing the right internal coordinates. >My experience is that nothing can beat a well-chosen >Z-matrix. Other internal coordinate systems, e.g. >Pulays natural internal coordinates compare also very >well in the literature. >However, the choice of the coordinate system depends >-at least in my view- entirely on two things: > >1. The nature of the molecule one wishes to optimize. > Establishing a Z-matrix for polycyclic systems is a > nightmare that nobody really wants to go through. > Some people claim that it is impossible to compose > a good Z-matrix for such a system. So if the only two > options available are cartesians and Z-matrix AND the > molecule is unpleasantly complex, cartesians > should be preferred. > In agreement with this, many papers mention that cartesians > are superior to anything else for polycyclic molecules. > However, these statements are NOT independent of 2.) > >2. The quality of the initial Hessian that is available. > Because of the couplings they introduce, cartesians do > in general not have the best reputation. However, > Baker and Hehre showed that if a good approximation to the Hessian > is available, cartesians perform comparably to other > sets of internal coordinates. They also stress, that for > FLEXIBLE polycyclic systems (whatever that may be) they found > the performance of cartesian coordinates to be below internal > coordinates. > >A few of the relevant references are: >J.Baker and W.J. Hehre, J.Comp.Chem. 12, 606 (1991) >J.Baker, J.Comp.Chem. 14, 1085 (1993) >J.Baker and D.Bergeron, J.Comp.Chem. 14, 1339 (1993) >P.Pulay and G.Fogarasi, J.Chem.Phys. 96, 2856 (1991) > > > Hope this helps, > Folke > > >Folke Noertemann >Lehrstuhl f. Theoret. Chemie >TU Muenchen >noertema@theochem.tu-muenchen.de > > >-------This is added Automatically by the Software-------- >-- Original Sender Envelope Address: ep7@dent.okayama-u.ac.jp >-- Original Sender From: Address: ep7@dent.okayama-u.ac.jp >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 > >