From chemistry-request@server.ccl.net Thu Apr 11 06:18:16 2002 Received: from igor.urz.unibas.ch ([131.152.1.3]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BAIFj11421 for ; Thu, 11 Apr 2002 06:18:15 -0400 Received: from proinformatix.com (yutang.pz.unibas.ch [131.152.84.72]) by igor.urz.unibas.ch (PMDF V6.1-1 #40761) with ESMTP id <0GUE008IQFA7U1@igor.urz.unibas.ch> for chemistry@ccl.net; Thu, 11 Apr 2002 12:18:07 +0200 (MET DST) Date: Thu, 11 Apr 2002 12:18:18 +0200 From: Sergio Manzetti Subject: ZN ions To: CCL Reply-to: sergio@proinformatix.com Message-id: <3CB562EA.6010703@proinformatix.com> MIME-version: 1.0 Content-type: text/plain; format=flowed; charset=us-ascii Content-transfer-encoding: 7bit X-Accept-Language: en-us User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:0.9.4) Gecko/20011128 Netscape6/6.2.1 Dear CCLrs I am trying to simulate a ZN peptidase. The problem is that the ZN ion, which is directed by an anterior sulphur atom to donate certain orbitals to one carboxylate oxygen and one carbonyl oxygen for polarization effect to carry out catalysis, is presented in the GROMOS96 force fields as a structural ion. Indeed, you could say that I am expecting an EVB/QM model of the ZN ion in a MD program, however I was thinking it could be possible to reduce the current hexacoordinated state to tetra coordinated state somehow. I have been advised to create a topology of the tetra coordinated state manually, but this is rather tedious because of the state of the protein-inhibitor complex. Are there options to modify the parameters in the GROMOS force field to your knowledge? Sergio From chemistry-request@server.ccl.net Thu Apr 11 06:03:08 2002 Received: from chinon.cnrs-orleans.fr ([163.9.6.107]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BA38j11178 for ; Thu, 11 Apr 2002 06:03:08 -0400 Received: (from hinsen@localhost) by chinon.cnrs-orleans.fr (8.11.2/8.11.2) id g3BA1wS09791; Thu, 11 Apr 2002 12:01:58 +0200 X-Authentication-Warning: chinon.cnrs-orleans.fr: hinsen set sender to hinsen@cnrs-orleans.fr using -f Sender: hinsen@chinon.cnrs-orleans.fr To: "J. Zheng" Cc: chemistry@ccl.net Subject: Re: CCL:reduced protein models References: From: Konrad Hinsen Date: 11 Apr 2002 12:01:58 +0200 In-Reply-To: Message-ID: Lines: 30 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii "J. Zheng" writes: > I am curious that does anyone has develop the general force field > (epslo, sigma, charge) for only 20 acid amio residues if we treat > each residue as one atom. If yes, could you please give me a > reference paper? A simple harmonic forcefield for a one-particle-per-residue model is described in the following paper: K. Hinsen, A.J. Petrescu, S. Dellerue, M.C. Bellissent-Funel, G.R. Kneller Harmonicity in slow protein dynamics Chem. Phys. 261, 25-37 (2000) The paper also shows that this forcefield plus harmonic Brownian dynamics (Brownian modes) gives excellent results for dynamic quantities (correlation functions) over a large time range. An implementation of this technique is available in the Molecular Modelling Toolkit, to be found at http://dirac.cnrs-orleans.fr/MMTK/ -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsen@cnrs-orleans.fr Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24 Rue Charles Sadron | Fax: +33-2.38.63.15.17 45071 Orleans Cedex 2 | Deutsch/Esperanto/English/ France | Nederlands/Francais ------------------------------------------------------------------------------- From chemistry-request@server.ccl.net Thu Apr 11 07:29:53 2002 Received: from mx1.ustc.edu.cn ([61.132.182.1]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BBTqj12590 for ; Thu, 11 Apr 2002 07:29:52 -0400 Received: from James ([202.38.66.112]) by mx1.ustc.edu.cn (8.8.7/8.8.6) with SMTP id TAA12048 for ; Thu, 11 Apr 2002 19:23:42 +0800 Message-Id: <200204111123.TAA12048@mx1.ustc.edu.cn> Date: Thu, 11 Apr 2002 19:29:10 +0800 From: Zhen Xie To: "chemistry@ccl.net" Subject: about PES scan X-mailer: FoxMail 4.0 beta 2 [cn] Mime-Version: 1.0 Content-Type: text/plain; charset="GB2312" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by server.ccl.net id g3BBTrj12591 Dear all the CCLers £¬ when I use gaussian98 to perform the job with route section "#p hf/6-31g opt=z-matrix test" to scan one parameter of a molecular,It always aborted with l502.exe. How can I solve this probelm? ¡¡¡¡ Zhen Xie zxie3@mail.ustc.edu.cn 2002-04-11 From chemistry-request@server.ccl.net Thu Apr 11 04:28:28 2002 Received: from mailer.gwdg.de ([134.76.10.26]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3B8SSj09919 for ; Thu, 11 Apr 2002 04:28:28 -0400 Received: from vrlab02.mpibpc.gwdg.de ([134.76.211.88] helo=mpi-bpc.mpg.de) by mailer.gwdg.de with esmtp (Exim 3.33 #11) id 16vZwM-00057g-00 for chemistry@ccl.net; Thu, 11 Apr 2002 10:28:22 +0200 Message-ID: <3CB54921.3060909@mpi-bpc.mpg.de> Date: Thu, 11 Apr 2002 10:28:17 +0200 From: Vlad Cojocaru User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.4.1) Gecko/20020314 Netscape6/6.2.2 X-Accept-Language: en-us MIME-Version: 1.0 To: chemistry@ccl.net Subject: new RNA residues Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Hi! I am trying to introduce some fluorescent base analogues (2aminopurine and pteridine analogues) into an RNA first on the computer and than exp. I created the new residues for AMBER but there are several problems: 1. the force field parameters are all taken from analogies with other structures already existent in AMBER 2. I calculated the charges using Gaussian (STO-3G and 6-31G*) first for the base alone and than I tried with the whole residue. There are significant differences. I did not do a RESP fitting yet. 3. I did geometry optimization with Gaussian but the amino group of 2AP was not planar anymore. 4. I put all the residues in a simple AMBER minimizer but the energy landscape looks very funny in the sense that it fluctuates a lot before convergence. 5. I tried also some MD. I have a question: Can somebody give a hint which would be the correct procedure for building these new residues and how can I obtain a reliabile residue structure that I can use for introducing in the RNA later on. Thanks a lot, vlad -- Vlad Cojocaru Max Planck Institut for Biophysical Chemistry Deparment: 060 Am Fassberg 11, 37077 Goettingen, Germany tel: ++49-551-201.1389 e-mail: Vlad.Cojocaru@mpi-bpc.mpg.de Home Address: Gutenbergstrasse 26/8 37075 Goettingen, Germany Home tel. number: ++49-551-9963204 Mobile: ++49-179-6851586 alternative email: johhnny_ar@yahoo.com From chemistry-request@server.ccl.net Thu Apr 11 05:57:12 2002 Received: from parasceve.cartuja.csic.es (root@[161.111.152.30]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3B9vBj11096 for ; Thu, 11 Apr 2002 05:57:12 -0400 Received: from us.es (ptolomeus.cartuja.csic.es [161.111.152.119]) by parasceve.cartuja.csic.es (8.9.3/8.9.3) with ESMTP id KAA02265 for ; Thu, 11 Apr 2002 10:31:37 GMT Message-ID: <3CB55E28.A2A719B6@us.es> Date: Thu, 11 Apr 2002 11:58:00 +0200 From: "Antonio J. =?iso-8859-1?Q?D=EDaz?= Quintana" Organization: IBVF (US-CSIC) X-Mailer: Mozilla 4.5 [es] (Win98; I) X-Accept-Language: es MIME-Version: 1.0 To: CCL Subject: SHAKE troubles X-Priority: 1 (Highest) Content-Type: multipart/mixed; boundary="------------49C0A052B2A3D6840044619D" Este es un mensaje multipartes en formato MIME. --------------49C0A052B2A3D6840044619D Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Dear Sirs, I´m trying to perform Mol. Dynamics calculations with periodic boundary and constant NTP (T=345K and 295K) conditions in AMBER. After c.a. 100 ps, I have problems with coordinate reseting by the SHAKE algorithm. I tried to solve this by reducing integration steps and changing bath and gear coupling constants. I should be very grateful if any of you could help me. Thanks in advance Antonio Díaz --------------49C0A052B2A3D6840044619D Content-Type: text/x-vcard; charset=us-ascii; name="qzaida.vcf" Content-Transfer-Encoding: 7bit Content-Description: Tarjeta de Antonio J. Díaz Quintana Content-Disposition: attachment; filename="qzaida.vcf" begin:vcard n:Díaz-Quintana;Antonio J. x-mozilla-html:TRUE org:University of Sevilla;Plant Biochemistry and Molecular Biology version:2.1 email;internet:qzaida@us.es title:Assitant Proffesor adr;quoted-printable:;;IBVF (US-CSIC)=0D=0AAvda Am=E9rico Vespucio s/n=0D=0A=0D=0A;Sevilla;;41092;SPAIN fn:Antonio Díaz Quintana end:vcard --------------49C0A052B2A3D6840044619D-- From chemistry-request@server.ccl.net Thu Apr 11 05:20:18 2002 Received: from mail.fqspl.com.pl ([212.244.147.20]) by server.ccl.net (8.11.6/8.11.0) with SMTP id g3B9KGj10669 for ; Thu, 11 Apr 2002 05:20:17 -0400 Received: (qmail 30284 invoked by uid 505); 11 Apr 2002 09:18:51 -0000 Received: from victor@fqspl.com.pl by mail by uid 502 with qmail-scanner-1.10 (F-PROT: 3.11. Clear:0. Processed in 0.225544 secs); 11 Apr 2002 09:18:51 -0000 Received: from test.fqspl.com.pl (HELO test) (10.10.10.50) by mail.fqspl.com.pl with SMTP; 11 Apr 2002 09:18:50 -0000 Message-ID: <004a01c1e13a$3b18ac30$320a0a0a@test> From: "Victor Anisimov" To: References: <00e901c1dca9$3f922290$320a0a0a@test> Subject: Re: MESP visualization Date: Thu, 11 Apr 2002 11:21:15 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="koi8-r" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Dear Colleagues, In my latest question to the list I guess I wanted too much with 8 million MESP points on 120 000 atom's protein as I could collect no responses. Actually complexity of the task could be in principle reduced by switching to visualization of only some part of the large MESP map. Having let's say only part of the whole Connolly surface near to a centre of my interest would it be possible to visualize this partial surface? As before I have coordinates of dots corresponding to this partial Connolly surface of my interest with MESP potentials calculated elsewhere. What visualization tool one can use to draw this partial MESP surface? The question is actually in special format of MESP surface, when GUI obtains a ready surface for visualization rather trying to reconstruct it from a cube grid. Is it implemented in some package? Perhaps, displaying a ready surface would be an extremely economic way when MESP of 120 000 atoms protein is a subject of study, i.e. if there is no way to display the whole surface at once, it would be good to display at least a part. If it is necessary any density of dots (0.1-0.001A) can be prepared in the MESP file to eliminate extra spline interpolation work within GUI. Is there any idea how this partial surface data could be feeded to a visualization program? Thank you for your input. Sincerely, Victor. ----- Original Message ----- From: "Victor Anisimov" To: Sent: Friday, April 05, 2002 3:53 PM Subject: CCL:MESP visualization > Dear CCLers, > > I'd like to make nice 3D visualization of a molecular electrostatic > potential calculated > for 120 000 atoms protein system on Connolly surface. My MESP file is a > textual file > with three columns corresponding to Cartesian coordinates of a point and > fourth > column containing MESP value, i.e. each line corresponds to a separate point > on > the Connolly surface. Number of points on one Connolly surface is about one > million. > No cube format is possible I guess by size of the molecule and number of the > mesh > points one would need to store in the cube file. > > Can someone suggest me a free software that can read this file and draw nice > 3D > map with colours corresponding to MESP value? My hardware limitation is 1GB > RAM and Windows/Linux single Athlon CPU machine. I'd greatly appreciate any > hints. > > Thank you in advance, > Victor. > From chemistry-request@server.ccl.net Thu Apr 11 02:09:49 2002 Received: from messi.uku.fi ([193.167.224.8]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3B69mj05879 for ; Thu, 11 Apr 2002 02:09:49 -0400 Received: from localhost (thassine@localhost) by messi.uku.fi (8.12.3/8.12.3) with ESMTP id g3B69XZa307174 for ; Thu, 11 Apr 2002 09:09:37 +0300 Date: Thu, 11 Apr 2002 09:09:33 +0300 (WET) From: Tommi Hassinen To: chemistry@ccl.net Subject: Re: Fwd: CCL:reduced protein models Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by server.ccl.net id g3B69nj05880 Hello. I have done something quite similar; please see Hassinen T.; Peräkylä M. J Comput Chem 22, 1229-1242 (2001) I have used 1-3 mass points / amino acid residue, and a solvent-accessible surface based solvation model. The program is also available (with source code) at http://www.bioinformatics.org/ghemical If you have any further questions, I would be glad to answer. Regards, Tommi Hassinen University of Kuopio Dept. of Chemistry > Begin forwarded message: > > > From: "J. Zheng" > > Date: Wed Apr 10, 2002 06:49:41 PM US/Central > > To: chemistry@ccl.net > > Subject: CCL:reduced protein models > > > > > > Dear CCLer: > > > > I have a general question about the protein simulation. For protein > > simulation, force field is important issue. As I know as far, most of > > persons use different force fields such as Charmm, Gromas, Amber etc to > > describe different protein systems with all-atoms model. > > > > Sometimes protein is so huge that we have to consider another way to > > deal with it. For example, we introduce new algorithm (Cell multipole > > method, PPPM) to reduce computation time, or we resort to parallel > > program. I am curious > > that does anyone has develop the general force field (epslo, sigma, > > charge) for only 20 acid amio residues if we treat each residue as > > one atom. If yes, could you please give me a reference > > paper? > > > > Now I am working on one protein simulation. I was just thinking about > > which way I should go. > > > > I will appreciate for any suggestion. > > > > good day. > > > > jie > > ----------------------------------------------- > > | JIE ZHENG | > > | Department of Chemical Engineering | > > | University of Washington | > > | Seattle, WA 98105, USA | > > ----------------------------------------------- > > | Tel: (206) 616-6510 (o) | > > | Email: jzheng73@u.washington.edu | > > | Webpg: students.washington.edu/jzheng73 | > > ----------------------------------------------- From chemistry-request@server.ccl.net Thu Apr 11 15:10:36 2002 Received: from mxout1.cac.washington.edu ([140.142.32.5]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BJAaj21674 for ; Thu, 11 Apr 2002 15:10:36 -0400 Received: from mailscan-out3.cac.washington.edu (mailscan-out3.cac.washington.edu [140.142.32.18]) by mxout1.cac.washington.edu (8.12.1+UW01.12/8.12.1+UW02.01) with SMTP id g3BJANYB018926 for ; Thu, 11 Apr 2002 12:10:27 -0700 Received: FROM mailhost2.u.washington.edu BY mailscan-out3.cac.washington.edu ; Thu Apr 11 12:10:22 2002 -0700 Received: from donald (borg.chem.washington.edu [128.95.128.134]) by mailhost2.u.washington.edu (8.12.1+UW01.12/8.12.1+UW02.01) with SMTP id g3BJAMww017634 for ; Thu, 11 Apr 2002 12:10:22 -0700 From: "Carsten Detering" To: "ChemistryList" Subject: strange output Date: Thu, 11 Apr 2002 20:54:24 +0200 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Importance: Normal Dear CClers, I discovered a strange output from a couple of my AutoDock runs: INITIALIZING AUTOMATED DOCKING SIMULATION _________________________________________ RUN 3... Time-dependent Seed: 1018551687 Sorry, too many retries (10000). Continuing... Will use the state with the lowest energy found, 123.04 INITIAL STATE: ?Á°¡+ƒ+_<-q¦a+-^Æyåñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ ______________ INITIAL STATE: ?Á°¡+ƒ+_<-q¦a+-^Æyåñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶USER Transformed Initial Coordinates INITIAL STATE: ?Á°¡+ƒ+_<-q¦a+-^Æyåñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 1 C23 ROS 1 102.878 -75.105 -82.340 +1.00 +0.00 +0.211 INITIAL STATE: ?Á°¡+ƒC23 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 2 n2 ROS 1 103.064 -74.820 -80.873 +1.00 +0.00 -0.504 INITIAL STATE: ?Á°¡+ƒn2 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 3 C22 ROS 1 104.009 -75.780 -80.259 +1.00 +0.00 +0.211 INITIAL STATE: ?Á°¡+ƒC22 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 4 C6 ROS 1 102.337 -73.748 -80.329 +1.00 +0.00 +0.083 INITIAL STATE: ?Á°¡+ƒC6 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 5 C5 ROS 1 102.591 -73.551 -78.925 +1.00 +0.00 +0.000 INITIAL STATE: ?Á°¡+ƒC5 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 6 C8 ROS 1 100.694 -71.877 -80.408 +1.00 +0.00 +0.000 INITIAL STATE: ?Á°¡+ƒC8 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 7 C11 ROS 1 100.499 -70.520 -77.045 +1.00 +0.00 +0.000 INITIAL STATE: ?Á°¡+ƒC11 ROS 1åñ"?´^ÑÙJ@Hm;5@!¹TD-@YÑ?o¶ATOM 8 C3 ROS 1 101.427 -71.325 -76.396 +1.00 +0.00 +0.026 .... Does any of you have an idea what this could be caused by?? My guess is that it is related to the random number generator, but I am not sure. Appreciate helpful suggestions. Thanks, Carsten ------------------------ Carsten Detering University of Washington Seattle, WA 98195 Phone: 206 543 5081 email: detering@u.washington.edu From chemistry-request@server.ccl.net Thu Apr 11 07:54:58 2002 Received: from mail1.rrz.Uni-Koeln.DE ([134.95.100.208]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BBsvj13396 for ; Thu, 11 Apr 2002 07:54:58 -0400 Received: from campfire.rrz.Uni-Koeln.DE (IDENT:2897@campfire.rrz.Uni-Koeln.DE [134.95.19.27]) by mail1.rrz.Uni-Koeln.DE (8.12.3/8.12.2) with ESMTP id g3BBsc14003136 for ; Thu, 11 Apr 2002 13:54:42 +0200 (MEST) Received: from localhost (acp64@localhost) by campfire.rrz.Uni-Koeln.DE (8.9.1a/8.9.1) with SMTP id NAA08888 for ; Thu, 11 Apr 2002 13:54:36 +0200 (MET DST) X-Authentication-Warning: campfire.rrz.Uni-Koeln.DE: acp64 owned process doing -bs Date: Thu, 11 Apr 2002 13:54:35 +0200 (MET DST) From: Johannes Weber Reply-To: Johannes Weber To: chemistry@ccl.net Subject: SUMMARY: g98: TD-DFT geometry optimizations ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 X-Virus-Scanned: by amavisd-milter (http://amavis.org/) Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by server.ccl.net id g3BBswj13397 Dear CCL members, Two weeks ago I posted a g98 related question concerning geometry optimizations and frequency calculations with the time-dependent DFT method. Here is the original question: ------------------------------------------------------------- Dear CCL members, I have a Gaussian 98 related question: According to the g98 users manual (p. 155, TD keyword) it should be possible to do Time Dependent DFT geometry optimizations of the N-th excited state using the Root=N option. However, I can´t find a proper input. The easiest input line #P TD=(Root=1) B3LYP/STO-3G Test DENSITY OPT=Z-Matrix leads to an error in link 1002: (Enter /vol/chemie/libexec/g98.A9/g98/l1002.exe) Minotr: Closed-shell wavefunction. Cannot mix post-SCF and DFT. Error termination via Lnk1e in /vol/chemie/libexec/g98.A9/g98/l1002.exe. I played around with several other options, tried to do the job in a two step procedure, like it is described for the CIS method in Forsman/Frisch´s "Exploring Chemistry with Electronic Structure Methods", all without success. Is it possible to do these calculations? Anybody out there with experience in this topic? The ultimate goal would be to perform a normal mode frequency calculation for excited states within the TD-DFT formalism. Regards, :-)ohannes. --------------------------------------------------------------------- I got several answers, my special thanks go to Dr. Doug Fox and Prof. Galina Orlova for the fruitful discussion. Other people which I´m indebted to are Wolfgang Roth and Rainer Remenyi. To put it in a nutshell: Yes, you can do TDDFT geometry optimizations and frequency calculations. However, since no gradients and higher derivatives are available for this method in g98, you have to resort to numerical optimizations, namely using the EF-algorithm with the opt=(Enonly,Z-Matrix) and freq=(Enonly) keyword. The optimization has to be performed in Z-Matrix coordinates and you need an initial guess for the force constants, which Dr. Fox recommended to take from a CIS calculation (optimization and frequency calculation). You can also try an take the force constants out of a different calculation with the RCFC option in the opt keyword. Opt and freq calculation have to be done in subsequent steps, a one-step procedure with opt and freq in the same input card will fail. To my (very limited) experience the usability of TDDFT optimizations is restricted by the fact that the numerical derivatives suffer from long run times (especially the freq calculations are really expensive for polyatomic molecules!) and a quite low accuracy (this is at least what I found from in the calculated frequencies). Until now, I wasn´t able to use force constants from other than CIS calculations for the initial guess. Best regards and thanks again to all people who answered. The relevant original answers are appended below. :-)ohannes. ====================================================================== answer 1: From: Doug Fox Subject: Re: CCL:g98: TD-DFT geometry optimizations ? Dr Weber, The manual is in error and there is not an implementation of analytic gradients for TD-DFT nor is there for TD-HF (aka RPA). You can get various roots and if you use OPT=(RCFC,EnOnly), taking the force constants perhaps from a single point CIS FREQ calculation, you could do a geometry optimization. Note that this will be more practical for smaller systems. ====================================================================== answer 2: Prof Orlova forwarded an email to me that she had received earlier from Jim Hess, help@gaussian.com. From: Galina Orlova Subject: Fwd: Re: TDDFT (from WWW) ----- Forwarded message from gaussian.com!help@gaussian.com ----- Date: Mon, 25 Mar 2002 11:49:06 -0500 From: gaussian.com!help@gaussian.com Reply-To: gaussian.com!help@gaussian.com Subject: Re: TDDFT (from WWW) To: gorlova@uoguelph.ca Dear Galina, Since there aren't analytic gradients available for TD methods, normal geometry optimizations are not possible with this keyword. The error you get with TD-DFT occurs when Gaussian actually tries to calculate the analytic gradient, but fails to find the necessary data. The only way to do a td-dft optimization is with numerical gradients. To do this, use "opt=(enonly,rcfc)". This option requires that the optimization be done in symbolic z-matrix coordinates and that initial force constants be read from the checkpoint file or from cards. The force constants would typically come from an analytic frequency. In this case, CIS frequencies seem to be the logical choice. As for your TD-HF optimizations, I'm afraid to say that they are not truly TD-HF optimizations. The code should give an error message, but instead it does the optimization using something very close to a CIS gradient. Since the optimizations you already completed were done with gradients based upon CIS, if you choose to go ahead with the numerical TD-HF geometry optimizations, these final geometries should be fairly close to the TD-HF minimum so I'd recommend using them as initial structures. Regards, Jim Hess --------------------------------------------------------------------- ... Dr. Galina Orlova Vis. Res. Professor Department of Chemistry and Biochemistry University of Guelph Guelph, ON, Canada N1G 2W1 tel. work (Int+1)(519)824-4120 ext 3102 home (Int+1)(519)829-3119 fax (519)766-1499 e-mail gorlova@uoguelph.ca ===================================================================== answer 3: From: Wolfgang Roth Subject: Re: CCL:g98: TD-DFT geometry optimizations ? Hallo Johannes, ich habe zwar schon eine ganze Weile nichts mehr mit Gaussian gerechnet und kenne auch die neuesten Features nicht, ABER: Zu dem von Dir beschriebenen Problem >According to the g98 users manual (p. 155, TD keyword) it should be >possible to do Time Dependent DFT geometry optimizations of the N-th >excited state using the Root=N option. kann ich Dir folgendes sagen: Der TD-DFT output von G98 Rev a.7 und a.9 sieht genauso aus wie die AUsgabe, die man bei einer CIS-Rechnung bekommt. Deswegen gehe ich mal davon aus, daß die Programmierer die Algorithmen wiederverwendet haben. Das "Root" Keyword beschreibt nur den "state of interest" d.h. den gewünchten elektronischen Zustand, für den die gewählte Rechnung durchzuführen ist. Ob man oder ob man nicht Geometrieoptimierungen durchführen kann ist bei der Berechnung der Wellenfunktion nicht von belang. Meines Wissens gibt es in keinem Programm einen Algorithmus zur Geometrieoptimierung mit Time dependent Methoden, auch nicht für TD-HF, das ja eigentlich der CIS-Methode sehr ähnlich/vergleichbar sein sollte. Das Problem ist, daß es wohl keine Gradienten für den TD (EOM??) Algorithmus gibt. > Is it possible to do these calculations? Ich denke: NEIN ich habs auch mal über die numerische Gradienten versucht, da tut Gaussian dann ywar ne ganze Weile (3N TD-DFT single points lang) was, steigt dann aber mit Fehlermeldung/ohne Ergebnis aus. >Anybody out there with experience in this topic? The ultimate goal would be to perform >a normal mode frequency calculation for excited states within the TD-DFT >formalism. Ja, das wär wirklich nett, v.a. weil CASSCF Freqs mit Gaussian mühsam sind.. Schöne Grüße Wolfgang ==W=R===========================W=o=l=f=g=a=n=g==R=o=t=h== TU Brauschweig Institut f. Physikalische und Theoretische Chemie Hans-Sommer-Strasse 10 Tel ++49 (0)531 391 5346 38106 Braunschweig Fax ++49 (0)531 391 5396 ======== http://www-public.tu-bs.de:8080/~wroth/ ========= ===================================================================== answer 4: From: Rainer Remenyi Subject: Re: CCL:g98: TD-DFT geometry optimizations ? Hi Johannes, Try UB3LYP instead of B3LYP. Regards, Rainer Remenyi Universität Heidelberg ===================================================================== answer 5: The second answer of Prof. Orlova, after I asked her if she also calculated the frequencies. Hi, Johannes I guess, if G98 does not have analitic gradients for TD then it does not have analitic second deriv. either. So you must run frequencies numerically (freq=numer). If you did do so and it still doesn't work you better ask the gaussian helpers. Unfortunately, my systems are so large that even numerical optimization is a great problem. Thus I don't have experience with numerical frequencies within TDDFT. Jim means the TD-hartree-Fock (not TDDFT!) optimization I've run "successfully" analitically. In fact, G98 must abort this job. For some reason it was running but Jim's guess is that in fact G98 performed the CIS-HF optimization instead of TD-HF. Actually,the CIS-HF results might be very close to those from TD-HF optimization with numerical derivatives. best regards, Galina Dr. Galina Orlova Vis. Res. Professor Department of Chemistry and Biochemistry University of Guelph Guelph, ON, Canada N1G 2W1 tel. work (Int+1)(519)824-4120 ext 3102 home (Int+1)(519)829-3119 fax (519)766-1499 e-mail gorlova@uoguelph.ca ======================================================================== answer 6: The second answer of Dr. Fox, after I tried out a numerical frequency optimization -- with peculiar results. From: "Cust. Service Doug" Subject: Re: CCL:g98: TD-DFT geometry optimizations ? Dr. Weber, There is no option to do numerical or enonly frequencies as a part of an optimization. ... As to your first point clearly the attempt a multi-step job, ie. OPT FREQ, is generating a bad keyword. As this is still present in the current revision I can only suggest you do this as a two step job, OPT=(RCFC,Enonly) as the first and then FREQ=Enonly for the second. I will have the combination examined for a future release but we don't plan any further bug fixes for G98 at this point. Note that it has created the bad combination RTD-B3LYP. It may also have a problem with correctly generating FREQ=EnOnly but I can't confirm that because it fails before actually generating that route. Numerical frequencies do suffer from much lower precision. The step size is a compromise between the infinitesmal step which would be the closes approximation to the analytic derivative but which has such a small energy difference that it is numerically inaccurate or a larger step which gives a numerically stable energy difference but might step outside of the harmonic region and introduce errors that way. In the case of very low frequencies you are probably better off using a somewhat larger step as the low frequencies will require a larger step to get a significant energy difference. Smaller steps might help the higher frequencies but that is less important for the optimization. ... > > Dear Dr. Fox,=20 > > thanks again for your quick response. After the easter vacation I now > come back to the TDDFT geometry optimization and frequency calculation > problem and pester you with further questions. Please, be patient with me!= > =20 > :-) I tested several cases with a minimal basis on para-benzoquinone. My > results are as follows:=20 > > (1) I managed to do a numerical frequency calculation after the > optimization job. However, it seems not possible to do a geometry > optimization and a frequency calculation in a single step! Having done a > CIS-optimization and freq calculation in advance the input card > > > # TD=3D(Root=3D1,Read) RB3LYP/STO-3G INTEGRAL(GRID=3DULTRAFINE) Test > > DENSITY=3Dcurrent Geom=3DCheck Guess=3DRead OPt=3D(EnOnly,RCFC) Freq=3DEN= > Only =20 > > leads - after a sucessful TD-DFT geometry optimization - to the following > error message in the freq run: > > >... > > ------------------------------------------------------------ > > #P Geom=3DAllCheck Guess=3DTCheck Test RTD-B3LYP(FC)/STO-3G Freq > > ------------------------------------------------------------ > > QPERR --- A SYNTAX ERROR WAS DETECTED IN THE INPUT LINE. > > GUESS=3DTCHECK TEST RTD-B3LYP(FC)/STO-3G=20 > > ' > > Last state=3D"GCL" > > TCursr=3D 820 LCursr=3D 44 > > Error termination via Lnk1e in /vol/chemie/libexec/g98.A9/g98/l1.exe. > > Job cpu time: 0 days 0 hours 0 minutes 0.1 seconds. > > File lengths (MBytes): RWF=3D 105 Int=3D 0 D2E=3D 0 Chk=3D 12 S= > cr=3D1 > > Segmentation Fault > >... > > In contrast, the following two step procedure seems to work: > > > %chk=3Dpbq.chk > > # TD=3D(Root=3D1,Read) RB3LYP/STO-3G INTEGRAL(GRID=3DULTRAFINE) Test > > DENSITY=3Dall Geom=3DAllcheck Guess=3DRead OPt=3D(EnOnly,RCFC)=20 > > > > > Why is this the case? Does the single step procedure always assume > analytical frequencies, even though Freq=3DENOnly was specified originally?= > =20 ... ====================================================================== ---------------------------------------------------------------------- | Johannes Weber | Email: | | Universitaet zu Koeln | Johannes.Weber@uni-koeln.de | | Institut fuer Physikalische Chemie | | | Lehrstuhl Prof. Dr. G. Hohlneicher | tel: 0049-(0)221-470-4812 | | Luxemburger Str. 116 | fax: 0049-(0)221-470-5144 | | 50939 Koeln | | ---------------------------------------------------------------------- From chemistry-request@server.ccl.net Thu Apr 11 11:11:16 2002 Received: from mx1.ustc.edu.cn ([61.132.182.1]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BFBFj17408 for ; Thu, 11 Apr 2002 11:11:15 -0400 Received: from James ([202.38.66.112]) by mx1.ustc.edu.cn (8.8.7/8.8.6) with SMTP id WAA24867 for ; Thu, 11 Apr 2002 22:58:08 +0800 Message-Id: <200204111458.WAA24867@mx1.ustc.edu.cn> Date: Thu, 11 Apr 2002 23:3:37 +0800 From: Zhen Xie To: "chemistry@ccl.net" Subject: Re: Re: CCL:about PES scan X-mailer: FoxMail 4.0 beta 2 [cn] Mime-Version: 1.0 Content-Type: text/plain; charset="GB2312" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from base64 to 8bit by server.ccl.net id g3BFBGj17409 //here is the error report >>>>>>>>>> Convergence criterion not met. SCF Done: E(UHF) = -76.5306831474 A.U. after 65 cycles Convg = 0.4020D-03 -V/T = 2.0125 S**2 = 0.0000 KE= 7.558483419521D+01 PE=-2.156388261837D+02 EE= 4.428698909758D+01 Annihilation of the first spin contaminant: S**2 before annihilation 0.0000, after 0.0000 Convergence failure -- run terminated. Error termination via Lnk1e in /usr/g98/l502.exe. //here is the input file #p uhf/6-311g** opt=z-matrix nosymm pes scan yique 0.2-0.002 0 1 X X 1 xx C 1 cx1 2 acx1 C 2 cx2 1 acx2 3 da1 H 3 ch1 1 ach1 2 da2 H 4 ch2 2 ach2 1 da3 xx 0.5 S 600 0.002 cx1=2.0 cx2=2.0 ch1=1.0 ch2=1.0 acx1=90. acx2=90. ach1=90. ach2=90. da1=0. da2=180. da3=180. ¡¡¡¡ >======= 2002-04-11 22:39:00 you write£º======= > >>You had better put on your input file and the error message >> >>**************************** >>* Linhong Chen * >>* Center for Astrophysics * >>* Tsinghua University * >>* Beijing, P.R.China * >>* 8610-62792126(phone) * >>* 8610-62792125(fax) * >>**************************** >>Email:dilys98@mails.tsinghua.edu.cn >> >>********You wrote********* >>>Dear all the CCLers £¬ >>> >>>when I use gaussian98 to perform the job with route section "#p hf/6-31g opt=z-matrix test" >>>to scan one parameter of a molecular,It always aborted with l502.exe. How can I solve this probelm? >>> >>>¡¡¡¡ >>> >>> >>>Zhen Xie >>>zxie3@mail.ustc.edu.cn >>>2002-04-11 >>> >>> >>>-= 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 >> >> >> >>. > >= = = = = = = = = = = = = = = = = = = = > > >Best regards£¡ > > >Zhen Xie >zxie3@mail.ustc.edu.cn >2002-04-11 > >===================================================== > >===================================================== = = = = = = = = = = = = = = = = = = = = Ö Àñ£¡ Zhen Xie zxie3@mail.ustc.edu.cn 2002-04-11 From chemistry-request@server.ccl.net Thu Apr 11 15:08:20 2002 Received: from web21409.mail.yahoo.com ([216.136.232.79]) by server.ccl.net (8.11.6/8.11.0) with SMTP id g3BJ8Kj21608 for ; Thu, 11 Apr 2002 15:08:20 -0400 Message-ID: <20020411190810.7224.qmail@web21409.mail.yahoo.com> Received: from [64.3.125.99] by web21409.mail.yahoo.com via HTTP; Thu, 11 Apr 2002 12:08:10 PDT Date: Thu, 11 Apr 2002 12:08:10 -0700 (PDT) From: Masakatsu Watanabe Subject: Re: CCL:reduced protein models To: "J. Zheng" , chemistry@ccl.net Cc: masa_w@pacbell.net In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Dear Jie, Tremendous efforts have recently been poured into developing a coarse-grain potential for simulating and investigating large biological molecular motions characterized by the long time dynamics. Here are some references that can help you: 1) Troyer et.al. utilize the simple point-per-residue potential developed by Miyazawa & Jernigan to perform the Langevin dynamics. (Check http://www.cmpharm.ucsf.edu/~troyer/eccc/toc.html). 2) The other approaches are to simplify amino acid residues by more than one interaction point; - The Head-Gordon & Brooks approach are described as the Virtual body Dynamics (See Biopolymers, 31, p 77, 1991). In their approach, however, the development of electrostatic potential is not so simple. Therefore, we took their representations of residues and developed our unique potential by defining the pair distribution function, described by Bahar et. al. (See PROTEINS, 29, p292, 1997), from the selected protein molecules. This develop potential was used in the simulation. - Haliloglu and Bahar (See Proteins, 31, p271, 1998) had also utilized their developed a coarse-grain potential for their MD simulations. - The coarser representation of residue is done by Herzyk and Hubbard (See PROTEINS, 17, p310, 1993). They used their residue representations in the content of NMR refinement, but their representation can be utilized to develop unique potential for the dynamics simulation. I myself involved some researches to utilize coarse-grain potential. In my opinion, there are not particularly outstanding coarse-grain potentials yet. Which coarse-grain potential be used in your study truly depends on what level of detail information you want to gain from the simulation. You can also expand the search from the reference presented here. Hopefully, this will help you a bit. Masa Watanabe --- "J. Zheng" wrote: > > Dear CCLer: > > I have a general question about the protein > simulation. For protein > simulation, force field is important issue. As I > know as far, most of > persons use different force fields such as Charmm, > Gromas, Amber etc to > describe different protein systems with all-atoms > model. > > Sometimes protein is so huge that we have to > consider another way to > deal with it. For example, we introduce new > algorithm (Cell multipole > method, PPPM) to reduce computation time, or we > resort to parallel > program. I am curious > that does anyone has develop the general force field > (epslo, sigma, > charge) for only 20 acid amio residues if we treat > each residue as > one atom. If yes, could you please give me a > reference > paper? > > Now I am working on one protein simulation. I was > just thinking about > which way I should go. > > I will appreciate for any suggestion. > > good day. > > jie ===== ========================================================= Masakatsu Watanabe, Ph.D. Wavefunction, Inc. Scientific Developer 18401 Von Karman, Suite 370 E-mail: masa@wavefun.com Irvine, CA 92612, USA Phone: 949-955-2120 FAX: 949-955-2118 Web: http://www.wavefun.com =========================================================== __________________________________________________ Do You Yahoo!? Yahoo! Tax Center - online filing with TurboTax http://taxes.yahoo.com/ From chemistry-request@server.ccl.net Thu Apr 11 17:42:07 2002 Received: from mxout3.cac.washington.edu ([140.142.32.19]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3BLg7j25211 for ; Thu, 11 Apr 2002 17:42:07 -0400 Received: from mailscan-out2.cac.washington.edu (mailscan-out2.cac.washington.edu [140.142.33.17]) by mxout3.cac.washington.edu (8.12.1+UW01.12/8.12.1+UW02.01) with SMTP id g3BLfw8J003944 for ; Thu, 11 Apr 2002 14:41:59 -0700 Received: FROM homer38.u.washington.edu BY mailscan-out2.cac.washington.edu ; Thu Apr 11 14:41:57 2002 -0700 Received: from localhost (xjtan@localhost) by homer38.u.washington.edu (8.12.1+UW01.12/8.12.1+UW02.01) with ESMTP id g3BLfrfI113216 for ; Thu, 11 Apr 2002 14:41:57 -0700 Date: Thu, 11 Apr 2002 14:41:53 -0700 (PDT) From: "X. Tan" To: chemistry@ccl.net Subject: torsion constraints in autodock Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear all, I'd like to ask you a question about torsion constraints. I have a set of small molecules with the peptide bonds(CO-NH) attaching to a benzene ring. In principle, all the atoms in peptide bonds and benzene should be almost coplanar. In another word, the bond between peptide bond and benzene ring should has the torsion angel either 0 or 180 degree. When I did autodock computation without torsion constraint on this kind of bonds, it gave me the answers with the torsion angel about 90 degree. Later, I used "hardtorcon" on these bonds. The conformations I got were still not good, with the torsion angel about 60 degree. Here is one of my pdbq file, ---------------------- REMARK 12 active torsions (identified by new id numbers): REMARK status: ('A' for Active; 'I' for Inactive) REMARK I 1- 14 between atoms: C1 and O30 REMARK 1 A 2- 16 between atoms: C2 and O3 REMARK 2 A 16- 17 between atoms: O3 and C4 REMARK 3 A 19- 20 between atoms: C10 and C11 REMARK 4 A 22- 24 between atoms: N13 and C15 REMARK 5 A 24- 25 between atoms: C15 and C16 REMARK 6 A 25- 26 between atoms: C16 and C17 REMARK 7 A 26- 27 between atoms: C17 and N32 REMARK 8 A 36- 37 between atoms: C6 and N7 REMARK 9 A 39- 41 between atoms: C40 and C41 REMARK 10 A 41- 42 between atoms: C41 and O42 REMARK 11 A 42- 43 between atoms: O42 and C43 REMARK 12 A 43- 44 between atoms: C43 and C45 REMARK I 4- 5 between atoms: C20 and C21 REMARK I 5- 6 between atoms: C21 and O22 REMARK I 8- 9 between atoms: C24 and O25 REMARK I 11- 12 between atoms: C27 and O28 ROOT ... -------------------- For this molecule, I defined the torsion constaints like this: hardtorcon 3 0. 10. # constrain torsion, num., angle(deg), range(deg) hardtorcon 8 0. 10. # constrain torsion, num., angle(deg), range(deg) I wondered whether I gave the wrong constraint. For a torsion A-B-C-D, in autodock program, the rotatable bond is defined by two atoms(B-C), then where the A and D come from, and is it possible to assign the torsion angle unambiguously in the constraints? How to do that? BTW, I have another question about autodock. Atom F is a stronger H-Bond acceptor, but why there is no H-Bond parameters for F in autodock(I used version 3.03)? Any help would be greatly appreciated! Xiaojian Tan xjtan@u.washington.edu From chemistry-request@server.ccl.net Thu Apr 11 23:16:19 2002 Received: from hkusua.hku.hk ([147.8.2.91]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g3C3GIj31061 for ; Thu, 11 Apr 2002 23:16:19 -0400 Received: from localhost (h9531257@localhost) by hkusua.hku.hk (8.11.4/8.11.4) with ESMTP id g3C3G2J05485 for ; Fri, 12 Apr 2002 11:16:02 +0800 (HKT) Date: Fri, 12 Apr 2002 11:16:02 +0800 (HKT) From: Ng Man Fai X-Sender: h9531257@hkusua To: CHEMISTRY@ccl.net Subject: Re: LH2 question In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I write to ask a question about the LH2 system in Rs. Molischianum. I find the structure from a protein data bank with Protein ID 1LGH, I find that the Mg to Mg distance for intra-dimer is 9.36A while 8.78A for inter-dimer. However, I find that a paper published by Sturcture, it states that the distances are 9.2A and 8.9A, respectively. So which one is correct? Or both are correct? Thanks for anyone who can help. Best regards, Andy