From ch172954@student.uq.edu.au Tue Apr 4 00:11:27 1995 Received: from student.uq.edu.au for ch172954@student.uq.edu.au by www.ccl.net (8.6.10/930601.1506) id AAA27231; Tue, 4 Apr 1995 00:02:55 -0400 Received: from student.uq.edu.au (ch172954@localhost [127.0.0.1]) by student.uq.edu.au (8.6.10/8.6.9) with SMTP id OAA26429; Tue, 4 Apr 1995 14:02:51 +1000 Date: Tue, 4 Apr 1995 14:02:45 +1000 (GMT+1000) From: ch172954 To: chemistry@ccl.net cc: Mansoor Namazian Subject: citation- one more In-Reply-To: <199504032021.QAA22365@www.ccl.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi and sorry for any inconvenience I have made recently just a short note to those who have been interested in article " AM1 vs. PM3" [1] you can get it by ftp-anonymous from ftp www.ccl.net It is located in: pub/chemistry/archived-messages/94/04 and try the file "12", lines 631-685. By the way, what does this sound to you ? [1] A. Holder, ftp-www.ccl.net, pub/chemistry/ archived-messages/94/04/12, 631-685. Thanks, Mansoor From jkl@ccl.net Tue Apr 4 00:41:28 1995 Received: for jkl@ccl.net by www.ccl.net (8.6.10/930601.1506) id AAA27539; Tue, 4 Apr 1995 00:40:05 -0400 Date: Tue, 4 Apr 1995 00:40:05 -0400 From: Jan Labanowski Message-Id: <199504040440.AAA27539@www.ccl.net> To: chemistry@ccl.net Subject: Quoting CCL messages in papers In-Reply-To: Mail from 'ch172954 ' dated: Tue, 4 Apr 1995 14:02:45 +1000 (GMT+1000) Cc: jkl@ccl.net I am afraid, I need to enter the discussion on quoting the messages appearing on Computational Chemistry List. I will try to be brief, and I do it under duress. The messages posted to CCL should be quoted as "private communications" for a variety of reasons, though obviously one may add, when quoting it, that the message was posted on this date to CCL. I see it like: John Doe, private communication (Computational Chemistry List, March 10, 1993). But there is probably no standard (for a good reason...). One reason to quote them as private communication is, as justly stated by Henry Rzepa, the fact that they are volatile. At this moment, there is no organizational guarantee that they will be available for a specified period of time. Obviously, if the archives are indexed, catalogued, and issued on a CD-ROM, they may become available in the Libraries in this form, and can be therefore quoted by title/publisher/and some number (we still do not have LC numbers for CD-ROMs, do we?). Other scenario is a depository -- like University Microfilms for Ph.D. thesis, or QCPE or CPA for software. But such thing does not exist yet, though I am (as you all know) envision it. The other, probably quite politically flamable aspect is that many people on this list need to obtain formal relase from their employers before they can officially publish something related to their work. And if they officially publish something without the consent of their employer using employer's computer/address, they may loose their jobs. I am sure this is not a problem in academia (though who knows -- in my old days, when I was in Poland which was then called People's Republic, we had to get a "consent" for publishing from the special body -- it was a simple formality in Sciences, but it was not that trivial for historians, political, social sciences and other humanities, or in places where they were obsessed with secrets). So please, take all reasons into consideration before you call private communication a publication. Jan Labanowski jkl@ccl.net -- Dr. Jan K. Labanowski, Senior Research/Supercomputer Scientist/Specialist, etc. Ohio Supercomputer Center, 1224 Kinnear Rd, Columbus, OH 43212-1163 ph:(614)-292-9279, FAX:(614)-292-7168, E-mail: jkl@ccl.net JKL@OHSTPY.BITNET From grechtst@pepvax.pepperdine.edu Tue Apr 4 00:56:28 1995 Received: from pepvax.pepperdine.edu for grechtst@pepvax.pepperdine.edu by www.ccl.net (8.6.10/930601.1506) id AAA27631; Tue, 4 Apr 1995 00:47:16 -0400 Received: by pepvax.pepperdine.edu (5.65c/IDA-1.4.4/Pep-5.1) id AA26143 for chemistry@ccl.net; Mon, 3 Apr 1995 21:47:48 -0700 From: Gregory Rechtsteiner Message-Id: <199504040447.AA26143@pepvax.pepperdine.edu> Subject: FCC Graphic To: chemistry@ccl.net Date: Mon, 3 Apr 95 21:47:47 PDT X-Mailer: ELM [version 2.3 PL11] Hello: I apologize for the slight "digression" posed by this question, but I have been searching for the folowing item without much luck. I am looking for a 3D drawing of a face centered cubic crystal structure (any atoms). I would prefer black on a white background, but I am not "picky." If anyone knows of some, or a location (ftp, etc), please let me know as soon as possible. Thank you for your time, Gregory A. Rechtsteiner -- Gregory A. Rechtsteiner Natural Science Division Pepperdine University grechtst@pepvax.pepperdine.edu garex@primenet.com From mayer@cric.chemres.hu Tue Apr 4 08:11:35 1995 Received: from cric.chemres.hu for mayer@cric.chemres.hu by www.ccl.net (8.6.10/930601.1506) id HAA01002; Tue, 4 Apr 1995 07:59:01 -0400 From: Received: by cric.chemres.hu (AIX 3.2/UCB 5.64/4.03) id AA08025; Tue, 4 Apr 1995 14:58:40 +0200 Date: Tue, 4 Apr 1995 14:58:40 +0200 Message-Id: <9504041258.AA08025@cric.chemres.hu> To: chemistry@ccl.net Subject: Free radicals Jeremy Greenwood Department of Pharmacology University of Sydney NSW 2006 Australia Dear Dr. Greenwood, I think, you need the "free valence indices" for visualizing free radicals. These were introduced in my paper I. Mayer, Chem. Phys. Lett. 27, 290 (1983) with some correction in Addendum to that paper: CPL 117, 396 (1985); also see IJQC 26, 151 (1984) [addendum 28, 419 (1985)]; and papers TCA 67, 315 (1985); IJQC 29, 73, 477 (1986) , etc. You will get for each atom a number which is the measure of the free spin on it - something as a "dot" on a chemical formula of a radical. If you are interested, I can send you routines which then you can adapt to your programs. Best regards, Prof. Istvan Mayer Central Research Institute for Chemistry of the Hungarian Academy of Sciences E-mails: mayer@cric.chemres.hu and IB13LVIB@HUEARN.sztaki.hu (or IB13LVIB@HUEARN.bitnet); H1376May@ella.hu (or H1376May@HUELLA.bitnet) From SB@CHEM.UNI.WROC.PL Tue Apr 4 10:11:44 1995 Received: from indigo2 for SB@CHEM.UNI.WROC.PL by www.ccl.net (8.6.10/930601.1506) id KAA04695; Tue, 4 Apr 1995 10:06:06 -0400 Received: from [156.17.86.161] by indigo2 via SMTP (931110.SGI/930416.SGI) for chemistry@ccl.net id AA07973; Tue, 4 Apr 95 15:25:47 +0200 Received: from ICHUWR/SMTPQueue by chem.uni.wroc.pl (Mercury 1.11); Tue, 4 Apr 95 15:26:19 +0100 Received: from Mailqueue by ICHUWR (Mercury 1.11); Tue, 4 Apr 95 15:25:49 +0100 From: "Slawomir Berski" To: chemistry@ccl.net Date: Tue, 4 Apr 1995 15:25:45 GMT+1 Subject: Li-problems Priority: normal X-Mailer: Pegasus Mail v3.1 (R1) Message-Id: <5A021CD514F@chem.uni.wroc.pl> Hi, I would like to perform some calculations on organolithium compounds using HyperChem software. Unfortunately there is a lack of lithium parameters for PM3 method in this program ver. 4.0. I tried to adopt parameters prepared by E.Anders , R.Koch and P.Freunschaft (J.Comp.Chem 14(1993)1301) designed for MOPAC 6/PC program but HyperChem requires additional parameters for monopole- monopole, dipole-dipole and quadrupole-quadrupole interactions with symbols: AM,AD and AQ at positions: 12,13,14 in pm3_2.apb file. Could anyone give me advise how to evaluate these parameters or maybe where can I find such parameters for lithium atom but designed for HyperChem program. Sincerely Slawek From pirard@couperin.scf.fundp.ac.be Tue Apr 4 10:56:39 1995 Received: from hermes.fundp.ac.be for pirard@couperin.scf.fundp.ac.be by www.ccl.net (8.6.10/930601.1506) id KAA05438; Tue, 4 Apr 1995 10:44:17 -0400 Received: by hermes.fundp.ac.be; id AA03917; Tue, 4 Apr 1995 16:48:55 +0200 Received: from couperin.scf.fundp.ac.be by hermes.fundp.ac.be; id AA03917; Tue, 4 Apr 1995 16:48:55 +0200 Received: by couperin.scf.fundp.ac.be (AIX 3.2/UCB 5.64/4.03) id AA27180; Tue, 4 Apr 1995 16:24:57 +0200 From: Bernard Pirard Message-Id: <9504041424.AA27180@couperin.scf.fundp.ac.be> To: chemistry@ccl.net Subject: volume computations Date: Tue, 04 Apr 95 16:24:56 +0100 Dear colleagues, Last monday, we asked a question dealing with softwares available for volume computations. Please find enclosed a summary of the replies. Yours sincerely, Benrard Pirard Florence Lebon. Hi both, I asked a simmilar question about 6 months ago and got an excellent code from: Larry Dodd, dodd@roebling.poly.edu or Doros Theodorus doros@pylos.cchem.berkeley.edu Good luck, John Waite I suggest that you also consider SAVOL. SAVOL2 is available directly from Robert Pearlman, University of Texas. SAVOL, from QCPE. Both are very fast, and as far as I know, accurate. The user can set van der Waals radii, radius of the solvent sphere, and separate the atoms of the ligand into groups for which the volume and surface area will be calculated separately. Yvonne Martin Computer Assisted Molecular Design Project Abbott Laboratories Hello There is a free code developed by a student at our site, it is in ftp.iqm.unicamp.br/pub/chemistry/molecular_surface/surf, there you'll find: surf-ps.doc.z the paper published in Computers & Chemistry in Postscript format surf.doc.z The same paper but (I'm not sure) in troff format surf.f.z the program itself the .z sufix means it was compressed using GNU gzip Regards Pedro > Let me suggest the use of GEPOL available from power.ci.uv.es last time I looked for it. In the source you would find several references explaining the methods used in the program. Also there is a recent article in J. Comp. Chem. (look for Pascual-Ahuir et. al in the CCL periodicals directories) Hope this help. Dr. Rafael R. Pappalardo Dept. of Physical Chemistry, Univ. of Seville (SPAIN) e-mail: rafapa@obelix.cica.es We used MSEED from QCPE for surface area claculations the advantage of this was that derivatives could be calculated Roger. From tsp@oxygen.chem.nthu.edu.tw Tue Apr 4 11:11:40 1995 Received: from nthu.edu.tw for tsp@oxygen.chem.nthu.edu.tw by www.ccl.net (8.6.10/930601.1506) id LAA05874; Tue, 4 Apr 1995 11:11:18 -0400 Received: from oxygen.chem.nthu.edu.tw ([140.114.45.223]) by nthu.edu.tw with SMTP id <181770(1)>; Tue, 4 Apr 1995 23:18:32 +0800 Received: by oxygen.chem.nthu.edu.tw (th3.3w) id AA19073; Tue, 4 Apr 95 23:03:39 +0800 From: tsp@oxygen.chem.nthu.edu.tw Subject: Semiempirical parameters for Na and K. To: CHEMISTRY@ccl.net Date: Wed, 5 Apr 1995 07:03:38 +0800 Mailer: Elm [revision: 70.85] Message-Id: <95Apr4.231832cst.181770(1)@nthu.edu.tw> Dear netters, I need the sodium and potassium parameters for MOPAC. Could anyone help me with these values? Thanks in advance. Shiuh-ping Tseng From U54632@UICVM.CC.UIC.EDU Tue Apr 4 11:41:43 1995 Received: from phem3 for U54632@UICVM.CC.UIC.EDU by www.ccl.net (8.6.10/930601.1506) id LAA06712; Tue, 4 Apr 1995 11:35:52 -0400 Received: from UICVM.CC.UIC.EDU (MAILER@UICVM) by phem3.acs.ohio-state.edu (PMDF V4.2-13 #5888) id <01HOXU5ZVDTC98GK3R@phem3.acs.ohio-state.edu>; Tue, 4 Apr 1995 11:35:39 EDT Received: from UICVM (NJE origin U54632@UICVM) by UICVM.CC.UIC.EDU (LMail V1.2a/1.8a) with BSMTP id 7466; Tue, 4 Apr 1995 10:35:01 -0500 Date: Tue, 04 Apr 1995 10:26:58 -0500 (CDT) From: U54632%UICVM.BITNET@phem3.acs.ohio-state.edu Subject: Chicago ACS national meeting To: chemistry@ccl.net Message-id: <01HOXU5ZVDTE98GK3R@phem3.acs.ohio-state.edu> Content-transfer-encoding: 7BIT I'm a student member of the American Chemical Society and its division of Computers in Chemistry. I'd like to present a poster at the American Chemical Society's national meeting to be held in Chicago this August, in the division of Computers in Chemistry. However, I don't know if the deadline to mail in the abstract is over. Could somebody kindly tell me the date by which I should mail in my abstract and to whom? Also, I'd like to procure the official ACS abstract form. From whom can I request the same? Thanking you in advance, Chaya Duraiswami (U54632@uicvm.uic.edu) From ferenc@rchsg8.chemie.uni-regensburg.de Tue Apr 4 12:11:41 1995 Received: from comsun.rz.uni-regensburg.de for ferenc@rchsg8.chemie.uni-regensburg.de by www.ccl.net (8.6.10/930601.1506) id MAA08951; Tue, 4 Apr 1995 12:11:26 -0400 Received: from rchsg8.chemie.uni-regensburg.de by comsun.rz.uni-regensburg.de with SMTP id AA28930 (5.65c/IDA-1.4.4 for ); Tue, 4 Apr 1995 18:11:04 +0200 Received: by rchsg8.chemie.uni-regensburg.de (931110.SGI/URRZ-Sub-1.5-irix) (for CHEMISTRY@ccl.net) id AA05179; Tue, 4 Apr 95 18:10:25 +0200 Date: Tue, 4 Apr 95 18:10:25 +0200 From: Ferenc.Molnar@chemie.uni-regensburg.de (Ferenc Molnar) Message-Id: <9504041610.AA05179@rchsg8.chemie.uni-regensburg.de> To: CHEMISTRY@ccl.net Subject: G92: electrostatic prop Dear Netters! Can anybody help me with an example input file for Gaussian 92 in which electrostatic properties at some additional points "outside" a molecule are calculated. And more important: What is your experience with this kind of calculation in terms of accuracy of the field and the gradients? Is there any real change in going from a SCF wavefunction to a MP2 wavefunction? Thank you very much in advance, Ferenc Ferenc Molnar --------------------------------------------------------------------------- Institut fuer Physikalische und Theoretische Chemie - Lehrstuhl Prof. Dick - Tel.: (+49) 941 943-4466 /-4486 Universitaet Regensburg Fax.: (+49) 941 943-4488 Universitaetsstrasse 31 D-93053 Regensburg Deutschland / Germany --------------------------------------------------------------------------- EMail (SMTP): Ferenc.Molnar@chemie.uni-regensburg.de --------------------------------------------------------------------------- :-) ...THE PHYSICISTS HAVE MADE THEIR UNIVERSE, AND IF YOU DO NOT LIKE IT, YOU MUST MAKE YOUR OWN. -- E. R. HARRISON IN "COSMOLOGY" (1980) --------------------------------------------------------------------------- From rs0thp@rohmhaas.com Tue Apr 4 14:26:46 1995 Received: from monte.br.rohmhaas.com for rs0thp@rohmhaas.com by www.ccl.net (8.6.10/930601.1506) id OAA16185; Tue, 4 Apr 1995 14:13:32 -0400 Received: by monte.br.rohmhaas.com (AIX 3.2/UCB 5.64/4.03) id AA20022; Tue, 4 Apr 1995 14:13:40 -0400 From: rs0thp@rohmhaas.com (Dr. Tom Pierce) Message-Id: <9504041813.AA20022@monte.br.rohmhaas.com> Subject: Electronic Session at August ACS MTG. To: chemistry@ccl.net, CHMINF-L@IUBVM.UCS.INDIANA.EDU Date: Tue, 4 Apr 1995 14:13:39 -0400 (EDT) Cc: rs0thp@monte.br.rohmhaas.com (Dr. Thomas Pierce) X-Mailer: ELM [version 2.4 PL24] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2384 Dear Folks, Call for Abstracts: If you use the Internet to: Communicate, Collaborate, or Educate - Then you should consider an electronic poster here to share your learnings. The Computers and Chemistry Division of the ACS is sponsoring a "Chemistry on the InfoBahn" symposium at the August '95 ACS Meeting in Chicago. We are planning for one day of oral presentations and and electronic presentation session for multimedia articles and posters. The focus of the symposium is on using the Internet to teach/publish/and probe chemical topics and issues. I KNOW you are out there, so use the new ELECTRONIC ACS FORM to submit an abstract (http://www.ssd.intel.com/ACS/acs.html ). The ACS meeting will be connected to the Internet during the conference. We plan on both a multimedia poster session (See you can even stay home and participate!) and if possible we will broadcast (MBONE) a few of the talks "live" over the Internet. To participate, send in an abstract Real Soon Now (by April 19th). Examples of electronic presentations are at: http://www.ch.ic.ac.uk/talks/ and at http://hackberry.chem.niu.edu:70/0/ECCC/homepage.html Send an ACS Paper Abstract to a session chair, either: Dr. Thomas Pierce Rohm and Haas Co. P.O. Box 219 Bristol, PA 19007 U.S.A Prof. Steven Bachrach Department of Chemistry Northern Illinois University DeKalb, Il 60115 U.S.A Prof. Henry Rzepa, Dept. of Chemistry Imperial College, LONDON SW7 2AY Great Britain Reprise--------------------------------------------------------- Chemistry on the InfoBahn at the ACS Chicago Meeting, August 20-24 1995 |*| A 150-word abstract on ACS Abstract |*| Form are due by April 19, 1995 ** Chemistry on the Infobahn - Dr. Thomas Pierce, Rohm and Haas Co. P.O. Box 219 Bristol, PA 19007; Voice: (215) 785-8989; FAX (215) 781-4204; email: thpierce@rohmhaas.com, Steven Bachrach, Department of Chemistry, Northern Illinois University, DeKalb, Il 60115;voice: (815)753-6863; Fax: (815)753-4802; email: smb@smb.chem.niu.edu, Dr Henry Rzepa, Dept. Chemistry, Imperial College, LONDON SW7 2AY; voice: (44) 171 594 5774 or 594 5809. Fax: (44) 171 594 5804; email: rzepa@ic.ac.uk, World-Wide-Web URL: http://www.ch.ic.ac.uk/rzepa.html. -- Sincerely, Thomas Pierce - THPierce@RohmHaas.Com - Computational Chemist "These opinions are those of the writer and not the Rohm and Haas Company." From prabha@laue.biochem.ubc.ca Tue Apr 4 20:11:46 1995 Received: from laue.biochem.ubc.ca for prabha@laue.biochem.ubc.ca by www.ccl.net (8.6.10/930601.1506) id UAA08960; Tue, 4 Apr 1995 20:05:23 -0400 Received: by laue.biochem.ubc.ca (920330.SGI/920502.SGI.AUTO) for chemistry@ccl.net id AA12000; Tue, 4 Apr 95 17:04:19 -0700 Date: Tue, 4 Apr 1995 17:04:18 -0700 (PDT) From: Prabha Siddarth To: chemistry@ccl.net Cc: hinsen@ere.umontreal.ca Subject: CCL:How to get FTNCHEK In-Reply-To: <9503201712.AA28587@cyclone.ERE.UMontreal.CA> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 20 Mar 1995, Hinsen Konrad wrote: > Since many people have asked me where ftnchek can be found, > I thought it might be helpul to tell everyone: > > You can get ftnchek by ftp from netlib.att.com, directory > /netlib/fortran, file ftnchek.tar.gz. It comes complete > with a manual and test cases. Installation is straightforward > and using it is even easier. There is really *no* excuse for > not using it. > I ftp'd the file but apparently it is a compressed binary file - how can I uncompress it and start to use it? The usual uncompress command in a UNIX machine doesn't work. Thanks for any help Prabha From mailh@comp.bioz.unibas.ch Tue Apr 4 22:04:32 1995 Received: from comp.bioz.unibas.ch for mailh@comp.bioz.unibas.ch by www.ccl.net (8.6.10/930601.1506) id WAA15075; Tue, 4 Apr 1995 22:04:30 -0400 Received: from moldyn.com by netcomsv.netcom.com with UUCP (8.6.9/SMI-4.1) id QAA20589; Tue, 4 Apr 1995 16:02:10 -0700 Received: by bart.moldyn.com (940816.SGI.8.6.9/921111.SGI.AUTO) for DIBUG@comp.bioz.unibas.ch id TAA17803; Tue, 4 Apr 1995 19:08:43 -0400 From: alper@moldyn.com (Howard Alper) Message-Id: <199504042308.TAA17803@bart.moldyn.com> Subject: Ewald sums To: DIBUG@comp.bioz.unibas.ch Date: Tue, 4 Apr 1995 19:08:38 -0500 (DT ) X-Original-Reply-To: moldyn.com!moldyn.com!alper X-Mailer: ELM [version 2.4 PL22] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 6707 Errors-To: DIBUG-ERRORS@comp.bioz.unibas.ch Reply-To: DIBUG@comp.bioz.unibas.ch Precedence: bulk Status: RO Dear Dr. Brown, I saw your post concerning the use of Ewald sums in liquid simulations, as a response to the question of Pieter Stouten (sorry if I have misspelled the name). Since the issue of long-range electrostatic forces is of central importance in simulations, and I have spent much time studying this issue, I thought I would add my own responses to your dialog. The Ewald sum is one of several commonly used methods to include long-range electrostatic interactions in a computationally tractable manner (the other main approach is the continuum-based Reaction Field model, besides, of course, using VERY large cutoffs.) The reason for even considering such methods is that the truncation of molecular interactions commonly employed in simulations gives rise to problems. Many in the simulation community, some for lack of understanding but most for lack of infinite computational resources, tend to downplay or outright ignore the effects of truncating potential and forces. Briefly, for polyatomic systems with regions of significant charge or polarity, use of abrupt truncation (meaning the potential and forces between entities are instantly set to zero when some representative distance between the entities crosses the preset cutoff distance) leads to the development of "hot spots" at those places in the system where truncation of strongly polar/charged interactions is being done. This means that some parts of the system have a temperature much higher than the target temp. (say, 300K) while other regions may have a temp. much lower than the target (since it is common to force the temp. of the whole system to equal the target temp.). This maldistribution of temperature will significantly affect dynamics properties (such as the diffusion coefficient) but will also affect structural quantities as well. Also briefly (well, sort of), the use of so-called switching functions also leads to an unnatural distribution of kinetic energy (ie, temperature), but of a different kind. Switching functions lead to a decrease in the kinetic energy in regions where they are applied, or to "cold spots" of a sort. The reasons for this behavior are discussed more fully in K. F. Lau, H. E. Alper, T. Thacher, and T. R. Stouch, J. Phys. Chem. 98, 8785 (1994). (I and Terry Stouch are from Bristol-Myers Squibb, by the way (or I was until 1994), so the paper is a joint effort between 2 researchers from Biosym and 2 from BMS, and does not necessarily represent the contention of Biosym as a company). Given the deleterious effects of the 2 most common methods for truncating interactions (and similar effects have been observed for a variety of systems (I will provide some references below)), it seems necessary to consider methods for including long-range electrostatic interactions more explicitly. I do not mean to say that for all simulations of any system that failure to use very large cutoffs, or methods such as Ewald sums for a Reaction Field, will lead to totally wrong results. It is a matter of how large a perturbation in the results (from that which would obtain if all interactions were included) one is willing to tolerate, or can tolerate, before one cannot obtain chemically meaningful results. But in the case of systems with highly polar or charged regions, the effects of truncation will be more significant than in more nonpolar systems. This situation will obtain for many biochemical systems of importance. Since the use of infinite cutoffs is impossible, much effort has gone into methods that closely approximate the effect of including all interactions, but in an affective manner. It is true that the Ewald sum was originally developed for use in crystalline/periodic systems, and that its use in liquid systems represents an approximation. The accuracy of that assumption must be tested empirically through simulations on relevant systems. It turns out that the use of Ewald sums on water and many biochemical systems (solvated proteins/nucleic acids) is a very good approximation. Simulations of water have shown that the Ewald sum does not introduce significant "crystallinity" into the behavior of liquid water, and that many properties (such as the orientational structure/dielectric constant and the diffusion coefficient) are produced most accurately/meaningfully only when Ewald sums (or Reaction Fields) are used. For simulations of solvated biomolecules, some care must taken, but when that is done the Ewald sum produced results that are more physically consistent and in better agreement with experiment than results obtained with abrupt truncation or switching functions. Generally similar conclusions are seen when the Reaction Field is used. The Ewald sum can be expensive to use, though how much more depends on how well your code is written, how large your system is, and the values you choose for the relevant Ewald summation parameters. Again, one has to decide how much perturbation in one's results is tolerable. Some references: H. E. Alper and R. M. Levy, J. Chem. Phys. 91, 1242 (1989) - shows the effects of switching functions on the orientational/dielectric properties of water. P. E. Smith and B. M. Pettitt, J. Chem. Phys. 95, 8430 (1991) - simulations of a pentapeptide in water with ions, which shows clearly the differences between using an Ewald sum and truncating with switching functions. H. E. Alper, D. Bassolino-Klimas, and T. R. Stouch, J. Chem. Phys. 98, 9798 (1993) - a simulation of water above a phosphlipid monolayer which demonstrates the necessity of using large cutoffs in simulations of charged or polar systems, if artifacts are to be avoided. H. E. Alper, D. Bassolino-Klimas, and T. R. Stouch, J. Chem. Phys. 99, 5547 (1993) - simulations of water above the monolayer which describe the properties of this system when the artifacts due to truncation are essentially eliminated, but also showing in more detail the effects of truncation. This paper complements the previous one. H. Schreiber and O. Steinhauser, Biochemistry 31, 5856 (1992) - this paper shows how the treatment of the long-range electrostatic forces affects simulations of protein stability in water. Many more references could be listed, but this should serve as a good start. Please forgive the length of this note; I hope it is informative. Please feel free to share it with anyone you consider relevant . Also feel free to contact me (alper@moldyn.com) if you want to discuss this further. Sincerely, Howard Alper -- Dr. Howard Alper Moldyn Inc. 1033 Massachusetts Avenue Cambridge, MA 02138 617-354-3124 x19 - helping molecules find happiness for over a 10th of a century. ____________________________________________________________________________ The following information is automatically appended by the mailer. This list is called DIBUG@COMP.BIOZ.UNIBAS.CH and covers the Discover, Insight and Biosym products User group. Mail (un)subscribe messages to DIBUG-REQUEST@COMP.BIOZ.UNIBAS.CH, other to DIBUG-ADMIN@COMP.BIOZ.UNIBAS.CH The archive is available on bioftp.unibas.ch as FTP and GOPHER/WWW service. From mailh@comp.bioz.unibas.ch Tue Apr 4 22:33:16 1995 Received: from comp.bioz.unibas.ch for mailh@comp.bioz.unibas.ch by www.ccl.net (8.6.10/930601.1506) id WAA16487; Tue, 4 Apr 1995 22:33:14 -0400 Received: from eole.ERE.UMontreal.CA by condor.CC.UMontreal.CA with SMTP id AA09903 (5.65c/IDA-1.4.4 for DIBUG@comp.bioz.unibas.ch); Tue, 4 Apr 1995 19:39:09 -0400 Received: from cyclone.ERE.UMontreal.CA by eole.ERE.UMontreal.CA (940406.SGI/5.17) id AA15953; Tue, 4 Apr 95 19:39:07 -0400 Received: by cyclone.ERE.UMontreal.CA (940406.SGI/5.17) id AA26429; Tue, 4 Apr 95 19:39:07 -0400 Date: Tue, 4 Apr 95 19:39:07 -0400 From: hinsenk@ERE.UMontreal.CA (Hinsen Konrad) Message-Id: <9504042339.AA26429@cyclone.ERE.UMontreal.CA> To: DIBUG@comp.bioz.unibas.ch In-Reply-To: <9504041009.ZM26352@sage.syntex.com> (dan@sage.syntex.com) Subject: Re: Discover 94.0 for solvent dynamics and ewald summation.. Errors-To: DIBUG-ERRORS@comp.bioz.unibas.ch Reply-To: DIBUG@comp.bioz.unibas.ch Precedence: bulk Status: RO > If you look at the right structural properties of your solution (angular > correlation functions, Kirkwoods G-Faktor, and NOT radial distribution > functions) such treatment of the Coulomb-interaction gives impressingly better > results! The question arises, do you get "better" behavior here for the right reasons? What happens if I have a strongly polarizing solute with a large When using Ewald sums (or any other method that calculates the full lattice sum of the electrostatic interactions), you calculate the properties of the thermodynamic limit of a sphere made up from periodic replications of your elementary cells, usually assuming that the sphere is embedded into a perfect conductor. This is of course not exactly the system that you want to have, but at least it is a physically possible system with real physical interactions. When using electrostatic interactions with cutoffs, you calculate the properties of a physically impossible system - electrostatic interactions with cutoffs do not exist in nature. You loose an important feature of real electrostatics - its long-range nature. The effects of this long-range nature can be of macroscopic magnitude, for some quantities of the same order of magnitude as the quantity itself. The quantities affected mostly are electrostatic ones (e.g. the dielectric constant) or those strongly dependent on the electric field (like the orientation of polar molecules). Calculating such quantities with cutoffs simply doesn't make sense - but it does make sense to calculate them with Ewald sums (or any other... (see above)), even if that means studying the "wrong" system. dipole moment. I know that the waters out at the edge of the box are not as highly oriented by the solute as the ones close in. Once you apply the ewald summation, however, the edge waters see the backend of the solute in the current box, and every solute ad infinitum in that direction is oriented in the same direction resulting in a huge dipole in the infinite solution Of course, ideally you would want the box to be much larger than your solute (I know that's too expensive!). Obviously, neither periodic boundary conditions (i.e. Ewald summation) nor cutoffs are the right solution for such problems. But I don't know of any better one, although I can imagine some methods that could be tried. However, it seems that no one is really working on this topic. ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsenk@ere.umontreal.ca Departement de chimie | Tel.: +1-514-343-6111 ext. 3953 Universite de Montreal | Fax: +1-514-343-7586 C.P. 6128, succ. A | Deutsch/Esperanto/English/Nederlands/ Montreal (QC) H3C 3J7 | Francais (phase experimentale) ------------------------------------------------------------------------------- ____________________________________________________________________________ The following information is automatically appended by the mailer. This list is called DIBUG@COMP.BIOZ.UNIBAS.CH and covers the Discover, Insight and Biosym products User group. Mail (un)subscribe messages to DIBUG-REQUEST@COMP.BIOZ.UNIBAS.CH, other to DIBUG-ADMIN@COMP.BIOZ.UNIBAS.CH The archive is available on bioftp.unibas.ch as FTP and GOPHER/WWW service.