From chemistry-request@server.ccl.net Fri May 21 04:27:31 1999 Received: from avalon.ciam.unibo.it (root@avalon.ciam.unibo.it [137.204.214.167]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA28331 for ; Fri, 21 May 1999 04:27:30 -0400 Received: from gandalf.ciam.unibo.it (giampi@localhost [127.0.0.1]) by avalon.ciam.unibo.it (8.8.7/8.8.7) with ESMTP id JAA23097 for ; Fri, 21 May 1999 09:13:51 +0200 Sender: giampi@avalon.ciam.unibo.it Message-ID: <374507AF.AF0A0930@gandalf.ciam.unibo.it> Date: Fri, 21 May 1999 09:13:51 +0200 From: "Dr. Gian Pietro Miscione" X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.0.36 i686) MIME-Version: 1.0 To: chemistry@server.ccl.net Subject: Oniom and UFF (G98) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi all, I'm performing some test jobs with the oniom and uff methods, available on Gaussian '98. Two main problems/questions: 1) UFF: After I've got an optimized structure, I perform another optimization, starting from that optimized geometry. What I get is a very much different geometry and a very much lower energy. Sounds pretty strange! What happened? 2) Oniom: I perform an optimization on a molecule divided in 2 layers. What does it happen, in the optimization alghoritm, to the atoms in the low layer? I mean: are they treated as hydrogen atoms or what? Thanks in advance! Cheers, Gian Pietro From chemistry-request@server.ccl.net Fri May 21 06:01:39 1999 Received: from Salicet.ucd.ie (salicet.ucd.ie [137.43.1.23]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA29125 for ; Fri, 21 May 1999 06:01:38 -0400 Received: from melanie.ucd.ie ([137.43.12.56]) by Salicet.ucd.ie (PMDF V5.2-31 #32510) with SMTP id <0FC2000ALUA2YC@Salicet.ucd.ie> for CHEMISTRY@ccl.net; Fri, 21 May 1999 10:56:26 +0100 (BST) Received: by melanie.ucd.ie (1.37.109.4/0.0) id AA01878; Fri, 21 May 1999 10:45:05 +0100 Date: Fri, 21 May 1999 10:45:05 +0100 (BST) From: larry@melanie.ucd.ie (Larry Cuffe) Subject: Solvation and TS's To: CHEMISTRY@ccl.net Message-id: <9905210945.AA01878@melanie.ucd.ie> Mailer: Elm [revision: 70.85] Dear CCL I'd apreciate it if anyone could give me pointers to the relevent literature or their thoughts on the following: I have calculated Transistion states for some simple molecules using the Onsager and SCIPCM models as implemented in G94, using a dielectric constant of 80.25 to simulate solvation in water. Now my question: presumably when a molecule makes a transition this hapens on a very short time scale. The high dielectric constant of water is a DC value and coresponds to realignments of molecules with a high dipole moment and also to some polarization effects within the electronic structure of individual molecules. The variation of dielectric constant with frequency has been measured so we could choose a suitable lower dielectric constant with which to optimize the transition state geometry, coresponding to the time scale of the reaction. The question is how would we relate the results of such a calculation to the stable molecular geometries at either side of the TS for which a DC dielectric constant seems to be the most apropriate choice. The Transistion state of interest coresponds to an E-Z rotation about a partial Double bond. Any help would be apreciated Larry Cuffe (larry@melanie.ucd.ie) From chemistry-request@server.ccl.net Fri May 21 07:56:53 1999 Received: from aleph.net.uniovi.es (aleph.net.uniovi.es [156.35.11.1]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id HAA30306 for ; Fri, 21 May 1999 07:56:51 -0400 Received: from pinon.ccu.uniovi.es by aleph.net.uniovi.es (PMDF V5.1-11 #30226) with ESMTP id <01JBGM37IDCO00K4FE@aleph.net.uniovi.es> for chemistry@ccl.net; Fri, 21 May 1999 13:43:26 +0200 (MET) Received: (from pueyo@localhost) by pinon.ccu.uniovi.es (8.9.2/8.9.2) id NAA00784; Fri, 21 May 1999 13:43:23 +0200 (METDST) Date: Fri, 21 May 1999 13:43:23 +0200 (METDST) From: Victor Lua~na Subject: Re: CCL:Significant figures and single point calculations To: lavelle@MBI.UCLA.EDU, owner-chemistry@server.ccl.net Cc: chemistry@ccl.net Message-id: <199905211143.NAA00784@pinon.ccu.uniovi.es> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7bit > Date: Thu, 20 May 1999 17:35:09 +0000 > From: Richard Wood > > But 0.001/1 corresponds to 0.1%. So any propagation of this gives maybe > 0.2%. Is this important? Probably not. > > On Thu, 20 May 1999, Laurence Lavelle wrote: > > Agreed. > > > > In fact my original posting makes it clear that this is a moot point for MD > > and ab initio simulations because the change in atomic positions > 3sf. > > > > But single point calculations on what should be the identical file will > > give different answers when bond lengths change by +/- 0.001A. > > > > At 01:07 PM 5/20/99 +0000, Richard Wood wrote: > > >Based on my experience, which is limited, in calculations derived from MD > > >simulations, that only the first place afer the decimal is significant. > > >I other words, if a calculation gives a bond length of 1.510867 Angstroms, > > >you should report it as 1.51 Angstroms. > > > > > >On Thu, 20 May 1999 whuber@almaden.ibm.com wrote: > > >> >> Many software packages round off significant figures. > > >> What do you mean by "significant"? Do you believe the results of your > > >> calculations (with all the approximations that go in there), or any > > >> experimental > > >> data, is that accurate? The point is not that calculations are more or less accurate. Nobody disputes that you should report results after carefully considering its significance. The point is that the output of a program, or the files designed to interface several programs together SHOULD provide REPRODUCIBLE results. Thats one of the essencial points of science. By producing a gross round-off many programs preclude reproducibility or further usage of their data. Regards, Victor Lua~na -- HomePage %%http://www3.uniovi.es/~quimica.fisica/qcg/vlc/luana.html%% +----------------------------------------------+ +---^---/ / ! Victor Lua~na ! | ~ / Just in case ! Departamento de Quimica Fisica y Analitica ! | | you don't ! Universidad de Oviedo, 33006-Oviedo, Spain ! < / remember ! e-mail: %%victor@carbono.quimica.uniovi.es%% ! | / where Oviedo ! phone: (34)-8-5103491 fax: (34)-8-5103125 ! |____ ___/ is ;-) +----------------------------------------------+ \/ From chemistry-request@server.ccl.net Fri May 21 08:58:56 1999 Received: from tigris.klte.hu (tigris.klte.hu [193.6.138.33]) by server.ccl.net (8.8.7/8.8.7) with SMTP id IAA30988 for ; Fri, 21 May 1999 08:58:47 -0400 Message-Id: <199905211258.IAA30988@server.ccl.net> Received: from anti02 (anti02.chem.klte.hu) by tigris.klte.hu (MX V4.1 VAX) with SMTP; Fri, 21 May 1999 14:52:14 +0100 Comments: Authenticated sender is From: "Tamas Gunda" To: CHEMISTRY@ccl.net Date: Fri, 21 May 1999 14:57:20 +1 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re:CCL:Significant figures... Priority: normal X-mailer: Pegasus Mail for Windows (v2.42a) Laurence Lavelle wrote: > I was hoping that the various software companies that I contacted would > make an announcement to CCL. Since this has not occurred you should be > aware of the following. > > Many software packages round off significant figures. > > For example, an original file with Cartesian coordinates to 6 sig fig. > > When saved as .pdb in HyperChem Professional 5 has 3 sig fig. > When saved as Z-matrix, 4 sig fig. > > In Chem 3D ULTRA 5, bond lengths changed by 0.001 A. > Again sig fig are different for different formats. > > Babel 1.3 converted a 6 sig fig file to 3 sig fig when going from .hin to .pdb > > Hopefully Gaussian will post significant figures for all file formats in > NewZMat. > > These round off errors will affect single point calculations. > > Laurence > > > > > -= 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 Hi, 1) From programmers point of view: Its true, that different programs output coordinates rounded off differently and significant figures vary. When I programmed my program Mol2mol, which interfaces between different softwares by transforming the file formats, the number of decimals was choosen according to the output files of original programs. If the new file format contains less decimals than the original, the numbers are rounded. If more, the numbers are right-padded with zeroes. 2) From chemist's point of view: The results of a calculation, let they be coordinates or bond lengths, depends on the input/starting data. For example, when making an MM force field calculation, the force constants have usually 1-2 significant decimals. Therefore the 3rd and next digits of the result have no physical importance, I think. Even if a software outputs coordinates or bond length with 6 decimals, the user should use only max 3. It is similar to laboratory practice: if you make a reaction from 1.22 g of starting material, it is nonsense to write into the note book that 0.9876 g was the product and the yield was 88.122418 % (althogh your calculator may show this result). Tamas Gunda ************************************************************************ Dr Tamas E. Gunda phone: (+36-52) 316666-2479 Research Group of Antibiotics fax : (+36-52) 310936 L. Kossuth University e-mail: tamasgunda@tigris.klte.hu POBox 36 http://dragon.klte.hu/~gundat H-4010 Debrecen Hungary ************************************************************************ From chemistry-request@server.ccl.net Fri May 21 11:15:12 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 LAA00574 for ; Fri, 21 May 1999 11:15:12 -0400 Received: from mserv.rug.ac.be (mserv.rug.ac.be [157.193.40.37]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id LAA03009 for ; Fri, 21 May 1999 11:09:54 -0400 (EDT) Received: from elptrs5.rug.ac.be (elptrs5.rug.ac.be [157.193.113.3]) by mserv.rug.ac.be (8.9.0/8.9.0) with SMTP id RAA23258 for ; Fri, 21 May 1999 17:09:39 +0200 (MET DST) Received: by elptrs5.rug.ac.be (AIX 4.1/UCB 5.64/4.03) id AA07504; Fri, 21 May 1999 17:09:21 +0200 Date: Fri, 21 May 1999 17:09:21 +0200 From: Sandra.Wauters@rug.ac.be (Sandra Wauters) Message-Id: <9905211509.AA07504@elptrs5.rug.ac.be> To: CHEMISTRY@ccl.net Subject: Discrepancy between G94 and G98 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Md5: SkNIqjYCUJ6BvaOebqzj9g== Dear all, we are doing some volume calculations with GAUSSIAN. We have done the same calculations on the same structure (for instance methyl radical) with as well GAUSSIAN 94 and GAUSSIAN 98. We have found the following values : for methyl radical : molar volume = 24.538 cm**3/mol GAUSSIAN 94 molar volume = 22.721 cm**3/mol GAUSSIAN 98 also for tetraline : molar volume = 123.722 cm**3/mol GAUSSIAN 94 molar volume = 108.403 cm**3/mol GAUSSIAN 98 Does anyone know the reason for this discrepancy between both packages. The calculation method itself was B3LYP. Thanks in advance, Sandra Wauters PhD Student Laboratorium voor Petrochemische Techniek Krijgslaan 281, S5, 9000 Gent (Belgium) email : Sandra.Wauters@rug.ac.be From chemistry-request@server.ccl.net Fri May 21 12:00:03 1999 Received: from macgw.chem.iupui.edu (macgw.chem.iupui.edu [134.68.137.71]) by server.ccl.net (8.8.7/8.8.7) with SMTP id MAA01305 for ; Fri, 21 May 1999 12:00:03 -0400 Message-ID: Date: 21 May 1999 10:59:33 -0500 From: "Boyd" Subject: Journal of Molecular Graphics and Modelling To: "OSC CCL" X-Mailer: Mail*Link SMTP for Quarterdeck Mail; Version 4.1.0 Dear CCLers, We have some exciting new and recent computational chemistry books and would like to get them reviewed for the Journal of Molecular Graphics and Modelling. JMGM is the official publication of the Molecular Graphics and Modelling Society and of the ACS Computers in Chemistry Division. The books cover these topics: (1) relativistic effects, (2) molecular dynamics and kinetics, (3) modeling on a PC, (4) linear algebra for quantum mechanics, (5) modeling colloids, and (6) spin dynamics in chemical reactions. If you would like to be considered for reviewing one of these books, please send me your name and address, and a sentence about your area of expertise. Thanks, Don Donald B. Boyd, Ph.D. Editor, Journal of Molecular Graphics and Modelling Department of Chemistry Indiana University-Purdue University at Indianapolis 402 North Blackford Street Indianapolis, Indiana 46202-3274, U.S.A. Telephone 317-274-6891, FAX 317-274-4701 E-mail boyd@chem.iupui.edu Web http://chem.iupui.edu/rcc/rcc.html From chemistry-request@server.ccl.net Fri May 21 03:02:13 1999 Received: from metro.ucc.usyd.edu.au (metro.ucc.usyd.edu.au [129.78.64.2]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id DAA27516 for ; Fri, 21 May 1999 03:02:11 -0400 Received: from pharmacy.pharm.su.oz.au (pharmacy.pharm.su.OZ.AU [129.78.154.1]) by metro.ucc.usyd.edu.au (8.9.3/8.9.3) with SMTP id QAA04381 for ; Fri, 21 May 1999 16:56:59 +1000 (EST) Received: from pc146 (pc146.pharm.su.OZ.AU) by pharmacy.pharm.su.oz.au (4.1/5.17) id AA10481; Fri, 21 May 99 16:15:17 EST Message-Id: <000701bea356$a252c180$d29a4e81@pc146> From: "Michael Ivery" To: Subject: Compiling Autodock 3.0 under Linux Date: Fri, 21 May 1999 16:53:29 +1000 Dear all, I would like to talk to anyone who has successfully compiled Autodock 3.0 under Linux on a PC. I need help. Thanks very much Michael Ivery michaeli@pharm.usyd.edu.au From chemistry-request@server.ccl.net Fri May 21 09:12:54 1999 Received: from bacchus.pc1.uni-duesseldorf.de (root@bacchus.pc1.uni-duesseldorf.de [134.99.152.11]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id JAA31165; Fri, 21 May 1999 09:12:40 -0400 Received: from bacchus.pc1.uni-duesseldorf.de (jochen@localhost [127.0.0.1]) by bacchus.pc1.uni-duesseldorf.de (8.8.7/8.8.7) with SMTP id PAA23349; Fri, 21 May 1999 15:06:20 +0200 From: Jochen Kuepper Organization: Heinrich-Heine-Universitaet To: Victor Lua~na , lavelle@MBI.UCLA.edu, owner-chemistry@server.ccl.net Subject: Re: CCL:Significant figures and single point calculations Date: Mon, 27 Jun 2135 21:30:41 +0200 X-Mailer: KMail [version 1.0.21] Content-Type: text/plain Cc: chemistry@ccl.net References: <199905211143.NAA00784@pinon.ccu.uniovi.es> MIME-Version: 1.0 Message-Id: <99052115061900.23272@bacchus.pc1.uni-duesseldorf.de> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id JAA31166 On Fre, 21 Mai 1999 Victor Lua~na wrote: >> > >> >> Many software packages round off significant figures. >> > >> What do you mean by "significant"? Do you believe the results of your >> > >> calculations (with all the approximations that go in there), or any >> > >> experimental >> > >> data, is that accurate? > > The point is not that calculations are more or less accurate. Nobody >disputes that you should report results after carefully considering its >significance. The point is that the output of a program, or the files >designed to interface several programs together SHOULD provide REPRODUCIBLE >results. Thats one of the essencial points of science. By producing a >gross round-off many programs preclude reproducibility or further usage >of their data. Well, if your number is accurate to some figure, you don't need to specify all this ten numbers you calculated _behind_ it, since all results _are_ reproducible within their error by the given numbers. (You probably need one more or so, but you get the idea here.) Greetings, Jochen ----------------------------------------------------------------------- Jochen Küpper Heinrich-Heine-Universität Düsseldorf jochen@uni-duesseldorf.de Institut für Physikalische Chemie I Universitätsstr. 1, Geb 26.43 Raum 02.29 phone ++49-211-8113681 40225 Düsseldorf fax ++49-211-8115195 Germany http://www-public.rz.uni-duesseldorf.de/~jochen ----------------------------------------------------------------------- From chemistry-request@server.ccl.net Fri May 21 14:46:15 1999 Received: from macgw.chem.iupui.edu (macgw.chem.iupui.edu [134.68.137.71]) by server.ccl.net (8.8.7/8.8.7) with SMTP id OAA03585 for ; Fri, 21 May 1999 14:46:15 -0400 Message-ID: Date: 21 May 1999 13:46:17 -0500 From: "Boyd" Subject: re: book reviews To: "OSC CCL" X-Mailer: Mail*Link SMTP for Quarterdeck Mail; Version 4.1.0 Hi All, Thanks for your overwhelming response to review the books I mentioned for the Journal of Molecular Graphics and Modelling. Please NO more requests for the book on PC modeling! I only had 1 copy! The only book not yet committed is #6: "Dynamic Spin Chemistry: Magnetic Controls and Spin Dynamics of Chemical Reactions." Thanks, Don Donald B. Boyd, Ph.D. Editor, Journal of Molecular Graphics and Modelling From chemistry-request@server.ccl.net Fri May 21 20:16:55 1999 Received: from darwin.ntu.edu.au (darwin.ntu.edu.au [138.80.128.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id UAA09889 for ; Fri, 21 May 1999 20:16:53 -0400 Received: from quoll.ntu.edu.au (quoll.ntu.edu.au [138.80.63.13]) by darwin.ntu.edu.au (8.8.8/8.8.8) with SMTP id JAA08233; Sat, 22 May 1999 09:43:17 +0930 (CST) Received: by quoll.ntu.edu.au; (5.65/1.1.8.2/09Aug94-1121AM) id AA14032; Sat, 22 May 1999 09:57:12 +0930 Message-Id: <19990522095711.A9380@quoll.ntu.edu.au> Date: Sat, 22 May 1999 09:57:11 +0930 From: Brian Salter-Duke To: Jochen Kuepper Cc: CHEMISTRY@ccl.net Subject: Re: CCL:Significant figures and single point calculations Mail-Followup-To: Jochen Kuepper , CHEMISTRY@ccl.net References: <199905211143.NAA00784@pinon.ccu.uniovi.es> <99052115061900.23272@bacchus.pc1.uni-duesseldorf.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2i In-Reply-To: <99052115061900.23272@bacchus.pc1.uni-duesseldorf.de>; from Jochen Kuepper on Sat, May 22, 1999 at 03:02:25PM +0200 On Sat, May 22, 1999 at 03:02:25PM +0200, Jochen Kuepper wrote: > On Fre, 21 Mai 1999 Victor Lua~na wrote: > > >> > >> >> Many software packages round off significant figures. > >> > >> What do you mean by "significant"? Do you believe the results of your > >> > >> calculations (with all the approximations that go in there), or any > >> > >> experimental > >> > >> data, is that accurate? > > > > The point is not that calculations are more or less accurate. Nobody > >disputes that you should report results after carefully considering its > >significance. The point is that the output of a program, or the files > >designed to interface several programs together SHOULD provide REPRODUCIBLE > >results. Thats one of the essencial points of science. By producing a > >gross round-off many programs preclude reproducibility or further usage > >of their data. > > Well, if your number is accurate to some figure, you don't need to specify all > this ten numbers you calculated _behind_ it, since all results _are_ > reproducible within their error by the given numbers. > (You probably need one more or so, but you get the idea here.) > There are two meanings to significant here and I think we are getting them confused. One is the agreement with experiment which is often very low. The other is the number of figures that are significant for the model. Let us take ab initio. We decide on a method say MP2, and we decide on a basis set, say 6-311G(d,p). If we do this calculation on different machines and/or with different code we expect to get the same answer, but of course we do not. We get result that are the same to maybe 12 figures if the geometries are identical. If the geometries have been rounded we get much poorer agreement and we can not check whether one code is really doing the same as another code. We often need to make these checks. No code should round input data unless of course it contains more figures than the machine can handle. Cheers, Brian. -- Associate Professor Brian Salter-Duke (Brian Duke) School of Mathematical and Physical Sciences, Northern Territory University, Darwin, NT 0909, Australia. Phone (61) (0)8-89466702 e-mail: b_duke@lacebark.ntu.edu.au or b_duke@quoll.ntu.edu.au