From dewi@ri.ac.uk Wed Jun 5 05:51:51 1996 Received: from mail-d.bcc.ac.uk for dewi@ri.ac.uk by www.ccl.net (8.7.5/950822.1) id FAA15605; Wed, 5 Jun 1996 05:38:03 -0400 (EDT) Received: from ri.ac.uk (actually davy.ri.ac.uk) by mail-d.bcc.ac.uk with SMTP (PP); Wed, 5 Jun 1996 10:37:51 +0100 Received: by ri.ac.uk (5.x/SMI-SVR4) id AA03784; Wed, 5 Jun 1996 10:37:51 +0100 From: dewi@ri.ac.uk (Dewi Lewis) Message-Id: <9606050937.AA03784@ri.ac.uk> Subject: Animation software To: chemistry@www.ccl.net Date: Wed, 5 Jun 1996 10:37:50 +0100 (BST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Dear all, Can any molecular animation package handle an animation where the *number of atoms* and the connectivity of those atoms change from one frame to anther? Jeffrey Gosper's Re_View (thanks for the answer) will certainly handle the change in connectivity. But does anyone handle a change in the number of atoms? There is a good reason for asking I promise! Any suggestions gratefully received and acknowledged. Dewi -- Dewi Lewis | Royal Institution of GB | email: dewi@ri.ac.uk | 21 Albemarle St. | WWW: http://www.ri.ac.uk/DFRL | London W1X 4BS | phone: 0171 409 2992 From peon@medchem.dfh.dk Wed Jun 5 06:51:52 1996 Received: from danpost.uni-c.dk for peon@medchem.dfh.dk by www.ccl.net (8.7.5/950822.1) id GAA15783; Wed, 5 Jun 1996 06:11:18 -0400 (EDT) Received: from medchem.dfh.dk (medchem.dfh.dk [130.225.177.15]) by danpost.uni-c.dk (8.6.4/8.6) with ESMTP id MAA21665 for <@danpost.uni-c.dk:chemistry@www.ccl.net>; Wed, 5 Jun 1996 12:11:17 +0200 Received: from [130.225.177.59] by medchem.dfh.dk via SMTP (940816.SGI.8.6.9/940406.SGI) for id MAA02878; Wed, 5 Jun 1996 12:28:31 +0200 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 5 Jun 1996 12:11:17 +0200 To: chemistry@www.ccl.net From: peon@medchem.dfh.dk (Per-Ola Norrby) Subject: Re: CCL:G:rigid vs relaxed scan in Gaussian Since I can't get through to Victor direct, I'll continue the discussion here instead. Victor M. Rosas Garcia wrote: >Hi there, > I'm trying to calculate a bond stretching constant for molecular >mechanics work. I'm doing a Scan of the potential energy surface (well, curve >in this case) for the stretching of the bond with the 6-31+G* basis set in >Gaussian94. The problem is that the keyword "Scan" in gaussian has two >options: rigid and relaxed. What is the recommendation for this kind of >parametrization work? Should the scan keep the molecule rigid or let it relax >at each point of the stretch? My boss says that it should be relaxed while I'm >going for the rigid. I think that letting the molecule relax at each point >would make the constant too specific for the model compound and, thus, not of >general applicability; not to mention cross-contamination by other stretches >and torsions present in the molecule. >Any comments are appreciated. > >Thanx > >Victor. Hi Victor, You have an interesting problem there. The "standard" way to solve it would be to use a relaxed scan (by the way, the recommended method in g94 is to do this without SCAN, just using OPT and the "S" option in the variable specification, either Z-matrix or IRC). You would then have to make sure that your MM method yields the same curve after parameterization. This is a bit wasteful, and there IS a method to do it with rigid scans. We have discussed the method for torsions (J.Comp.Chem. 16, 620-627, 1995). The trick is to make sure you use the same EXACT geometries in both the ab initio and molecular mechanics when you parameterize. It's easiest with torsions, since the energies aren't that high, and the couplings with other parameters may not even be included in the force field. If you are doing stretches, and are working with a force field that includes a stretch-bend term, you should probably sample the potential energy surface at several points, not only changing the bond length but also angles. If you are able to include not only energies, but also energy derivatives in the parameterization, it improves the results a lot. For a beautiful example of this and a discussion of the method, see Hagler et al., J.Comp.Chem. 15, 162-182, 1994. Note also that if you are staying at the relatively low level HF/6-31+G*, you should use scaling. A scale factor of 0.8929 is commonly used in frequency calculation, Hagler et al. gives some other examples in J.Am.Chem.Soc. 116, 2515-2525, 1994. You could also take a look at Halgrens recent parameterization in, I believe, the March or April number of J.Comp.Chem. this year, he uses similar techniques but higher ab initio levels for his parameterization. Best Regards, Per-Ola Norrby ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ * Per-Ola Norrby * The Royal Danish School of Pharmacy, Dept. of Med. Chem. * Universitetsparken 2, DK 2100 Copenhagen, Denmark * tel. +45-35376777-506, +45-35370850 fax +45-35372209 * Internet: peon@medchem.dfh.dk, http://compchem.dfh.dk/ From heelisp@delta.newi.ac.uk Wed Jun 5 09:52:00 1996 Received: from delta for heelisp@delta.newi.ac.uk by www.ccl.net (8.7.5/950822.1) id JAA16484; Wed, 5 Jun 1996 09:27:14 -0400 (EDT) From: Date: Wed, 5 Jun 1996 13:38:33 +0100 Message-Id: <96060513383345@delta.newi.ac.uk> To: CHEMISTRY@www.ccl.net Subject: intermolecular complexes summary X-VMS-To: SMTP%"CHEMISTRY@www.ccl.net" Recently I posted a question on modelling intermolecular complexes. I enclose a summary. Thanks to all who replied. Paul Heelis Original Question Dear Ccl people, >Has anyone tried to model intermolecular complex formation. I am interested in nitrogen heterocyclics, which appear to form a spectroscopically distinct complex when both oxidised and reduced form are present in equal amounts. Is a valid approach to ask a semi-empirical programm to optimise a file containing both molecules in close proximity? and how close to start from? Any answers will be summarised for the group. REPLIES 1. We routinely model various complexes in MOPAC (and its brethren). We usually will start with a trial geometry from a molecular mechanics calculation. Our major problem is that there may be several structures which are consistent with whatever experimental evidence you have to compare. Getting a unique answer to a question may thus be impossible, but it is instructive to see which types of interactions may be operative. =================================================== Stephen B. Bowlus, Ph.D. Computer-Aided Molecular Design Research Division e-mail: bowlus@sandoz.com Sandoz Agro, Inc. Phone: + 1 415 354 3904 975 California Ave. Fax: + 1 415 857 1125 Palo Alto, CA 94304 2. Regarding your question, a few comments: (1) Depending on the nature of your bonding, you could select a value between normal covalent bond length and vdw distance. Try to run a constrained optimization first with constraints of such bond(s) which allow the rest of degrees of freedom settled, and then release the constraint and carry out a full geometry optimization. (2) Alternative is to do a coordinate driving, driving a distance from covalent distance to vdw distance with N step. Watch the energy of these N sub structure and see there is any minimum there. (3) SE methods are not parameterized for weakly bonded complex with small interaction energy (bond energy). High level theories might have to used to calibrate such. Try a model system for a control test. Hope these help! --Wayne -- +---------------------------------------------------------------+ | Wayne Huang, Ph.D. | 18401 Von Karman, Suite 370 | | Computational Chemist | Irvine, California 92715 | | Wavefunction, Inc. | 714-955-2120 <> 955-2118(fax)| | huang@wavefun.com | Web: http://www.wavefun.com 3. I presume that you have some kind of charge-transfer complex. Calculations on complexes should be possible just like with regular molecules, as the "complex" is a molecule, just with some bonds weaker than normal covalent bonds. Your systems are reminiscent of hydroquinone/quinone (quinhydrone) complexes--look these up in the lit. If you have a GUI with spacefilling models, dock the two species till they are in contact (van der Waals separation, then merge the files and optimize (if you can use Spartan you are lucky). Yes, it should be OK to treat the "supermolecule" as a single file. For deltaE, watch out for BSSE---well-known in ab initio, but I'm not sure how one treats it in semiemp. Good Luck Errol Lewars From rosas@irisdav.chem.vt.edu Wed Jun 5 10:51:55 1996 Received: from irisdav.chem.vt.edu for rosas@irisdav.chem.vt.edu by www.ccl.net (8.7.5/950822.1) id KAA17018; Wed, 5 Jun 1996 10:39:38 -0400 (EDT) Received: by irisdav.chem.vt.edu (940816.SGI.8.6.9/921111.SGI.AUTO) for chemistry@www.ccl.net id KAA07651; Wed, 5 Jun 1996 10:37:58 -0700 From: "Victor M. Rosas Garcia" Message-Id: <9606051037.ZM7649@irisdav.chem.vt.edu> Date: Wed, 5 Jun 1996 10:37:54 -0400 In-Reply-To: peon@medchem.dfh.dk (Per-Ola Norrby) "CCL:G:rigid vs relaxed scan in Gaussian" (Jun 5, 12:11pm) References: X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: chemistry@www.ccl.net Subject: REPOST: rigid vs relaxed scan in Gaussian Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hi people, Some days ago I posted the following message: >Hi there, > > I'm trying to calculate a bond stretching constant for molecular >mechanics work. I'm doing a Scan of the potential energy surface (well, curve >in this case) for the stretching of the bond with the 6-31+G* basis set in >Gaussian94. The problem is that the keyword "Scan" in gaussian has two >options: rigid and relaxed. What is the recommendation for this kind of >parametrization work? Should the scan keep the molecule rigid or let it relax >at each point of the stretch? My boss says that it should be relaxed while I'm >going for the rigid. I think that letting the molecule relax at each point >would make the constant too specific for the model compound and, thus, not of >general applicability; not to mention cross-contamination by other stretches >and torsions present in the molecule. >Any comments are appreciated. > >Thanx > >Victor. But my machine went down the day after (hard drive crash) and stayed like that for a few days (fridays are bad days for maintenance). So if some people tried to reply directly to me, is must have bounced back. Could those people kindly resend me their replies? I'll post a summary of the answers. So far I've got the answers from Per-Ola Norrby, Artem Masunov and Andy Dustman (thanx). Any others? TIA Victor. -- ----------------------------------------------------------------------- Victor M. Rosas Garcia * "How can we contrive to be rosas@irisdav.chem.vt.edu * at once astonished at the Virginia Tech doesn't necessarily share * world and yet at home in it?" the opinions you just read. * G. K. Chesterton ----------------------------------------------------------------------- From owner-chemistry@ccl.net Wed Jun 5 11:51:55 1996 Received: from bedrock.ccl.net for owner-chemistry@ccl.net by www.ccl.net (8.7.5/950822.1) id LAA17163; Wed, 5 Jun 1996 11:05:57 -0400 (EDT) Received: from cat.syr.edu for cmartin@cat.syr.edu by bedrock.ccl.net (8.7.5/950822.1) id LAA13283; Wed, 5 Jun 1996 11:05:32 -0400 (EDT) Received: from serval.cat.syr.edu.syr.edu by cat.syr.edu (4.1/1.0-6/5/90) id AA13197; Wed, 5 Jun 96 11:05:12 EDT Received: by serval.cat.syr.edu.syr.edu (SMI-8.6/SMI-SVR4) id LAA19421; Wed, 5 Jun 1996 11:04:33 -0400 Date: Wed, 5 Jun 1996 11:04:33 -0400 From: cmartin@cat.syr.edu (Charles H. Martin) Message-Id: <199606051504.LAA19421@serval.cat.syr.edu.syr.edu> To: amt@ftn.net Cc: chemistry@ccl.net In-Reply-To: (amt@ftn.net) Subject: Re: CCL:Compu Chem Club in Toronto? -------------------- Hi, I saw your message on the COmp Chem club. I am not in the TORONTO area, but I am wodering if you could tell me bout the Albany one Thanks Chuck Martin Syracuse University ------------------------- Date: Mon, 3 Jun 96 10:10 EDT X-Sender: amt@172.27.99.1 (Unverified) X-Mailer: Windows Eudora Version 1.4.3 Mime-Version: 1.0 From: amt@ftn.net (Adi Treasurywala) Content-Type: text/plain; charset="us-ascii" Sender: chemistry-request@www.ccl.net Errors-To: jkl@ccl.net Precedence: bulk Folks, I am writing to ask if anyone knows of a regularly convened meeting of computational chemists/biologists in the Toronto CANADA area. I have some experience with these gatherings having particcipated in the founding of the ones at Albany NY. and Philadelphia PA. If one exists already I would like to participate. If one does not yet exist I would be willing to help start one. Please let me know directly if you know of one or if you are interested in participating in founding one. ----------------------------- From page@plains.nodak.edu Wed Jun 5 17:51:58 1996 Received: from plains.nodak.edu for page@plains.nodak.edu by www.ccl.net (8.7.5/950822.1) id QAA18817; Wed, 5 Jun 1996 16:59:09 -0400 (EDT) Received: from p.chem.ndsu.nodak.edu (p.chem.ndsu.NoDak.edu [134.129.127.25]) by plains.nodak.edu (8.7.1/8.7.1) with SMTP id PAA11387 for ; Wed, 5 Jun 1996 15:58:59 -0500 (CDT) Message-Id: <199606052058.PAA11387@plains.nodak.edu> Date: Wed, 05 Jun 96 22:00:41 0600 From: michael page X-Mailer: Mozilla 1.1N (Macintosh; I; 68K) MIME-Version: 1.0 To: chemistry@www.ccl.net Subject: comp.theory award/call for applicants X-URL: http://www.ccl.net/ccl/welcome.html#3 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii >from Subdivision of Theoretical Chemistry, American Chemical Society --Call for Applications IBM/Cornell Theory Center Graduate Student Awards in Computational Chemistry Below is an announcement for two Awards in Computational Chemistry that are open to current graduate students. We are grateful to IBM and the Cornell Theory Center for their support of these awards. Let me urge you to encourage your graduate students to apply. The competition is open to any graduate student (regardless of citizenship) who is an ACS member (or whose advisor is an ACS member). These awards are designed to encourage graduate work in computational chemistry, to recognize research accomplishments, and to stimulate interest in the Subdivision of Theoretical Chemistry and the Physical Chemistry Division of the ACS. An Awards Committee will consider all the applicants. The awards applications should be sent to Mark Gordon at Iowa State University. Note that the deadline for applications is August 9. These two awards, supported by IBM and the Cornell Theory Center, will provide one-time cash stipends of $2500 and $1000 as supplements to normal financial aid to doctoral candidates in the research-dissertation stage in the 1996-1997 academic year. The Theory Center will provide each awardee up to 1000 service units on its SP2 parallel computer for the awardees to actually carry out a portion of the awarded research. The awardees will have access to the consulting services of the Theory Center normally available to all users. Awardee selection will be made on a competitive basis. Applicants should be working on new and innovative computational chemistry methods or applications in theoretical chemistry. Applicants should prepare a written description of a computational chemistry research project that requires high performance computing, with an explanation of the scientific importance of the project. Proposals need to include an estimate of the computing resources required in SP2 cpu-hours. Applicants should explain how they plan to use the grant funds. Two letters of recommendation, including one from the student's advisor, along with a vita and transcript, are required. In addition, a PI (other than the applicant) responsible for the applicant's use of the Cornell Center resources must be identified. Forward applications by August 9, 1996 to Prof. Mark Gordon Department of Chemistry Iowa State University Ames, Iowa 50011-3111 mark@si.fi.ameslab.gov The awardees will be chosen and announced at the Fall 1996 ACS National Meeting. From owner-chemistry@ccl.net Wed Jun 5 19:51:59 1996 Received: from bedrock.ccl.net for owner-chemistry@ccl.net by www.ccl.net (8.7.5/950822.1) id TAA19259; Wed, 5 Jun 1996 19:19:58 -0400 (EDT) Received: from gatekeeper.homecare.com for Karl.F.Moschner@unilever.com by bedrock.ccl.net (8.7.5/950822.1) id TAA18699; Wed, 5 Jun 1996 19:19:57 -0400 (EDT) Received: from x500mhub.lo.u1388.unilever.com by gatekeeper.homecare.com; (5.65v3.2/1.1.8.2/24Jan95-0439PM) id AA24489; Thu, 6 Jun 1996 00:19:24 +0100 Received: by x500mhub.lo.u1388.unilever.com; Thu, 6 Jun 96 00:18:04 Z X400-Received: by /c=GB/admd=telemail/prmd=Unilever/; Relayed; 06 Jun 96 00:18:01 Z X400-Received: by mta MTAunilever in /c=GB/admd=telemail/prmd=Unilever/; Relayed; 06 Jun 96 00:18:01 Z X400-Mts-Identifier: [/c=GB/admd=telemail/prmd=Unilever/; 006F131B615A9015-MTAunilever] Content-Identifier: 006F131B615A9015 Content-Return: Allowed X400-Content-Type: P2-1988 ( 22 ) Conversion: Allowed Priority: normal Disclose-Recipients: Prohibited Alternate-Recipient: Allowed X400-Originator: Karl.F.Moschner@unilever.com X400-Recipients: non-disclosure; Message-Id: <006F131B615A9015*/c=gb/admd=telemail/prmd=unilever/o=unilever/ou=ccmail/s=Moschner/g=Karl/i=F/@MHS> Date: 06 Jun 96 00:18:01 Z From: "Karl F Moschner" To: "mckelvey@Kodak.COM" (Return requested), "CHEMISTRY@ccl.net" (Return requested) Subject: Re: Compiling G94 on SGI R8000/IRIX 6.1 Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="PART.BOUNDARY.x500mhub.06f2.31b615ac.0001" --PART.BOUNDARY.x500mhub.06f2.31b615ac.0001 Content-Type: text/plain; charset=us-ascii RFC-822-Headers: X-Mailer: Z-Mail (3.0.0 15dec93) Errors-To: ccl@www.ccl.net Precedence: bulk --PART.BOUNDARY.x500mhub.06f2.31b615ac.0001 Content-Type: text/plain; charset=us-ascii Subject: CCL:G:Compiling G94 on SGI R8000/IRIX 6.1 Author: chemistry-request@www.ccl.net at INTERNET Date: 6/3/96 12:30 PM CClers: What G94 rev-level is required for error free compiles and builds for G94 on an SGI-R8000 using IRIX 6.1 ? HELP! Thanks, John -- John M. McKelvey email: mckelvey@Kodak.COM Computational Science Laboratory phone: (716) 477-3335 2nd Floor, Bldg 83, RL Eastman Kodak Company Rochester, NY 14650-2216 -------------------------------- REPLY ----------------------------------------- I have successfully compiled all revisions since June 95 (at least later than G94 Rev. B.1)on my Power Challenge under Irix 6.1 using SGI Power Fortran 6.1. NOTE: The 'sgi.make' file for the latest release, Rev D.3, requires Irix 6.2, [Power] Fortran 6.2 which is currentl;y being rolled out (I'm scheduled to get my upgrade later this month). But it can be compiled successfully under Irix/Fortran 6.1 by removing the compiler flag 'r8000' from the 'sgi.make' file which appears on only 1 line. WARNING: In the off chance you allocate a lot of scratch space, modify 'Default.Route' so 'MAXDISK=15GB' or less. If you don't you may inadvertently get integer overflow in a counter which can cause an infinite loop in fully direct MP2 calculations such as 'test065'. Good luck! Karl _______________________________________________________________________ / \ | Comments are those of the author and not Unilever Research U. S. | | | | Karl F. Moschner, Ph. D. | | | | Unilever Research U. S. e-mail: karl.f.moschner@unilever.com | | 45 River Road Phone: (201) 840-2629 | | Edgewater, NJ 07020 FAX: (201) 840-8299 | \_______________________________________________________________________/ --PART.BOUNDARY.x500mhub.06f2.31b615ac.0001--