From owner-chemistry@ccl.net Thu Nov 30 10:05:01 2006 From: "lior itzhaki lior.itzhaki~~gmail.com" To: CCL Subject: CCL: Comp versus other disciplines Message-Id: <-33123-061129094432-21801-W6SfIE1t5KkPWcDwbiltTA:_:server.ccl.net> X-Original-From: "lior itzhaki" Content-Type: multipart/alternative; boundary="----=_Part_27393_23152221.1164807975404" Date: Wed, 29 Nov 2006 15:46:15 +0200 MIME-Version: 1.0 Sent to CCL by: "lior itzhaki" [lior.itzhaki!^!gmail.com] ------=_Part_27393_23152221.1164807975404 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline hello. Could you please tell me how can I write an e-mail question to all the CCL users? I've sent an e-mail to chemistry-request . ccl.net. is that o.k? thank you lior ------=_Part_27393_23152221.1164807975404 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline
hello.
 
Could you please tell me how can I write an e-mail question to all the CCL users?
I've sent an e-mail to chemistry-request . ccl.net. is that o.k?
thank you
lior

 

------=_Part_27393_23152221.1164807975404-- From owner-chemistry@ccl.net Thu Nov 30 11:28:00 2006 From: "Markus Thut markus.thut%a%iac.unibe.ch" To: CCL Subject: CCL:G: AW: G: NBO visualization with GaussView Message-Id: <-33124-061130044556-24406-Sb7aPnax/jRoiP7aTAhYgg(_)server.ccl.net> X-Original-From: Markus Thut Content-Transfer-Encoding: 8bit Content-type: text/plain; charset="iso-8859-1" Date: Thu, 30 Nov 2006 10:10:09 +0100 MIME-version: 1.0 Sent to CCL by: Markus Thut [markus.thut===iac.unibe.ch] Hi Adil I think you have to turn your .chk file into a .fchk file - check the Gaussian manual for further reference. Regards Markus ---------------------------------------------------------------------------- -------------- Markus Thut University of Bern, Department of Chemistry and Biochemistry Leutwyler Group Freiestrasse 3 3012 Bern Switzerland markus.thut[a]iac.unibe.ch fon: +41316314240 fax: +41316313994 -----Ursprüngliche Nachricht----- Von: owner-chemistry[a]ccl.net [mailto:owner-chemistry[a]ccl.net] Gesendet: Donnerstag, 30. November 2006 04:23 An: Thut, Markus Betreff: CCL:G: NBO visualization with GaussView Sent to CCL by: "Adil Zhugralin" [zhugrali:_:bc.edu] Dear CCL'ers, I don't seem to be able to generate MO's from .chk file in GaussView. I load .log file (calculation done on G03), open MO Editor, check "read checkpoint file" in Guess type and try to load the .chk file after specifying the path to it; however, gv responds with an error message: CConnectionGFCHK::Initialize() Cannot find file What does this error mean? How can I fix it? Any suggestions would be appreciated. Thank you very much in advance. Adil BChttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt From owner-chemistry@ccl.net Thu Nov 30 12:36:01 2006 From: "Ibrahim M. Moustafa I.moustafa##psu.edu" To: CCL Subject: CCL: GROMACS installation on Mac-OS Message-Id: <-33125-061130025552-28691-4GL2XjaGBWuL3N92sQYL4g**server.ccl.net> X-Original-From: "Ibrahim M. Moustafa" Content-Type: text/plain; charset="us-ascii"; format=flowed Date: Wed, 29 Nov 2006 22:40:13 -0500 Mime-Version: 1.0 Sent to CCL by: "Ibrahim M. Moustafa" [I.moustafa**psu.edu] Hi, Thanks Dodson for your reply. Actually, the $GMSBIN was already sitting in the $PATH prentenv confirmed that. As for the Segmentation fault error, it could be as you suggested! but if so, how does this related to the way I type the command? For the time being, I'm aliasing all the commands in $GMSBIN and it is working (at least for the demo)!. As for Amber, unfortunately, I don't have access to download it for the time being....same for Charmm. thanks for your time and help, Ibrahim At 08:53 PM 11/29/2006, you wrote: >Sent to CCL by: "M. L. Dodson" [mldodson^houston.rr.com] >On Wednesday 29 November 2006 16:30, Ibrahim M. Moustafa ria2(a)psu.edu >wrote: > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2|a|psu.edu] > > Hi everybody, > > > > Thanks Rick for your advice; but actually my Xcode was recently > > updated! I'm kind of figured out what was the problem which is: after > > installation and setting the variables I got an error message > > "Segmentation fault" when trying to issue any command...or trying the > > demo tutorial! > > It appeared that my installation was fine (even my compilation was > > alright......in fact compilation resulted in a code that is much > > faster than the provided binary for Mac-OSX). > > I found the whole path is required to issue GROMACS' command to run > > any program; although, the variable $GMXBIN is already set in my > > .cshrc file. This has solved the problem, but still I don't > > understand why I can't run the programs (e.g. pdb2gmx) by just typing > > "pdb2gmx" in the xterm rather than typing "$GMXBIN/pdb2gmx" to make > > it working!! I'm not sure if this is a an OS problem or GROMACS > > problem! or something obvious that I'm missing! Any advice would be > > appreciated. > > > > I haven't heard any comments regarding TINKER vs GROMACS as a > > modelling package! wish I can hear from someone though! > > > > thanks, > > Ibrahim > >Try issuing the command > >printenv PATH > >and look to see if the $GMXBIN directory is in your path. The >behavior you describe indicates it is probably not there. Fix >your .cshrc to include it, and your "full path problem" should go >away. The seg fault problem may indicate that your have some >component(s) compiled for single precision and other components >compiled for double precision. On FreeBSD (at least on Intel >architectures), this can be confusing at times. GROMACS gets a >lot of its speed by running in single precision, whereas most >other MD packages are compiled for double precision. In >particular, make sure the compiled precisions match for your >GROMACS and FFTW installations. > >As to your TINKER vs GROMACS question, I prefer AMBER and namd :-) > >Bud Dodson > > > > > At 12:22 AM 11/28/2006, you wrote: > > >Sent to CCL by: Rick Venable [venabler(_)nhlbi.nih.gov] > > >There can be issues with the gcc and g77 versions under MacOS X; in > > >particular, if you've kept up to date with OS patches, you may need to > > >download a newer version of the Xcode tools from the Apple Developer > > > site. > > > > > >I had some problems compiling CHARMM until I updated my Xcode tools. > > > > > >-- > > >Rick Venable 29/500 > > >Membrane Biophysics Section > > >NIH/NHLBI Lab. of Computational Biology > > >Bethesda, MD 20892-8014 U.S.A. > > >(301) 496-1905 venabler AT nhlbi*nih*gov > > > > > > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2!=!psu.edu] > > > > > > > > I have a limited experience in Molecular modelling of protein > > > > structures; I'd love to expand my experience in that direction. So, I > > > > have few questions for people with more experience in the field. > > > > I wanted to install GROMACS package on my Mac-OS 10.4 according to > > > > the instructions on the GROMACS website. > > > > First, I installed the FFTW source as instructed in the README > > > > file. Then, I installed the GROMACS binary for the MacOS-X 10.4. > > > > However, it did not work! I tried the examples provided with the > > > > package but I got "Segmentation faults"....nothing in the output. > > > > Just a blank pages while the script is running interactively. > > > > I tried to build the package myself from the source code; but again I > > > > got the same error message when running the examples' scripts! > > > > Presumably, my installation was not successful! > > > > even running the command pdb2gmx gives that error message > > > > "Segmentation fault"!!!-- >M. L. Dodson >Email: mldodson-at-houston-dot-rr-dot-com >Phone: eight_three_two-56_three-386_one From owner-chemistry@ccl.net Thu Nov 30 14:16:01 2006 From: "Ian Hovell hovell(0)cetem.gov.br" To: CCL Subject: CCL:G: Summary unusual opt/basis set combination Message-Id: <-33126-061130135614-3904-Iwn/6U+fkBBCqp75C0BFnw_._server.ccl.net> X-Original-From: Ian Hovell Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Thu, 30 Nov 2006 16:43:16 -0200 MIME-Version: 1.0 Sent to CCL by: Ian Hovell [hovell . cetem.gov.br] I wrote: A collegue of mine is trying to optimise a benzene ring with the following command line using G03. # opt=tight b971 gen The gen keyword calls the following TZ2P basis set reproduced at the end of this message. The molcule optimises okay without the tight option and fails with it. Is the TZ2P basis set inappropriate for this one parameter hybrid functional? Any suggestions what I might try to get it to optimise? and the answers I recieved: You may simply need more opt cycles with opt=tight. You can continue the optimization with "opt=readfc geom=check guess=read" or "opt=calcfc geom=check guess=read" ----------------------- (That was my first thought and when this failed was the reason why I consulted CCL) My first thought would be that you need to compute the integrals more accurately to make this work. Try using Int=Grid=Ultrafine and see if that helps it reach Opt=Tight criteria. I would also consider using Opt=(Tight,CalcFC) is I was starting from the optimized benzene ring structure from the smaller basis. -------------------------- (This worked Int=Grid=Ultrafine) Teaches one to read the manual more carefully!) Thanks to David and Douglas for their helpful replies. Ian From owner-chemistry@ccl.net Thu Nov 30 14:51:01 2006 From: "M. L. Dodson mldodson-x-houston.rr.com" To: CCL Subject: CCL: GROMACS installation on Mac-OS Message-Id: <-33127-061130132700-28482-tNhtP8lFy+gZ2nGWmXSn7g^^^server.ccl.net> X-Original-From: "M. L. Dodson" Content-Disposition: inline Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="iso-8859-1" Date: Thu, 30 Nov 2006 12:26:49 -0600 MIME-Version: 1.0 Sent to CCL by: "M. L. Dodson" [mldodson{=}houston.rr.com] On Thursday 30 November 2006 11:40, Ibrahim M. Moustafa I.moustafa##psu.edu wrote: > Sent to CCL by: "Ibrahim M. Moustafa" [I.moustafa**psu.edu] > > Hi, > > Thanks Dodson for your reply. Actually, the $GMSBIN was already > sitting in the $PATH prentenv confirmed that. > As for the Segmentation fault error, it could be as you suggested! > but if so, how does this related to the way I type the command? > For the time being, I'm aliasing all the commands in $GMSBIN and it > is working (at least for the demo)!. > > As for Amber, unfortunately, I don't have access to download it > for the time being....same for Charmm. > Hmmm... I Guess I misunderstood you about the segfault. Do you mean it segfaults only when you DO NOT type the whole path name? Or the other way around? In either case, I do not understand why it should make any difference unless the pdb2gmx program is using its argv[0] (executable file name) variable for some purpose. I will shut up now as I know very little about GROMACS itself. I was just answering with general Unix information. Segfaults can happen for many different reasons. My suggestion was based on an earlier installation of GROMACS on (Intel) FreeBSD where I had the precision mismatch problem. If you still can not get it to work after other people put in their 2 cents, let me know, and I will try to install GROMACS on my iBook G4 to see if I can debug the problem. Should be really funny to measure how much longer simulations take on the iBook :-) Both AMBER and CHARMM cost $$ as well (but are well worth it if you qualify for the academic prices). Bud Dodson > thanks for your time and help, > > Ibrahim > > At 08:53 PM 11/29/2006, you wrote: > >Sent to CCL by: "M. L. Dodson" [mldodson^houston.rr.com] > >On Wednesday 29 November 2006 16:30, Ibrahim M. Moustafa ria2(a)psu.edu > > > >wrote: > > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2|a|psu.edu] > > > Hi everybody, > > > > > > Thanks Rick for your advice; but actually my Xcode was recently > > > updated! I'm kind of figured out what was the problem which is: after > > > installation and setting the variables I got an error message > > > "Segmentation fault" when trying to issue any command...or trying the > > > demo tutorial! > > > It appeared that my installation was fine (even my compilation was > > > alright......in fact compilation resulted in a code that is much > > > faster than the provided binary for Mac-OSX). > > > I found the whole path is required to issue GROMACS' command to run > > > any program; although, the variable $GMXBIN is already set in my > > > .cshrc file. This has solved the problem, but still I don't > > > understand why I can't run the programs (e.g. pdb2gmx) by just typing > > > "pdb2gmx" in the xterm rather than typing "$GMXBIN/pdb2gmx" to make > > > it working!! I'm not sure if this is a an OS problem or GROMACS > > > problem! or something obvious that I'm missing! Any advice would be > > > appreciated. > > > > > > I haven't heard any comments regarding TINKER vs GROMACS as a > > > modelling package! wish I can hear from someone though! > > > > > > thanks, > > > Ibrahim > > > >Try issuing the command > > > >printenv PATH > > > >and look to see if the $GMXBIN directory is in your path. The > >behavior you describe indicates it is probably not there. Fix > >your .cshrc to include it, and your "full path problem" should go > >away. The seg fault problem may indicate that your have some > >component(s) compiled for single precision and other components > >compiled for double precision. On FreeBSD (at least on Intel > >architectures), this can be confusing at times. GROMACS gets a > >lot of its speed by running in single precision, whereas most > >other MD packages are compiled for double precision. In > >particular, make sure the compiled precisions match for your > >GROMACS and FFTW installations. > > > >As to your TINKER vs GROMACS question, I prefer AMBER and namd :-) > > > >Bud Dodson > > > > > At 12:22 AM 11/28/2006, you wrote: > > > >Sent to CCL by: Rick Venable [venabler(_)nhlbi.nih.gov] > > > >There can be issues with the gcc and g77 versions under MacOS X; in > > > >particular, if you've kept up to date with OS patches, you may need > > > > to download a newer version of the Xcode tools from the Apple > > > > Developer site. > > > > > > > >I had some problems compiling CHARMM until I updated my Xcode tools. > > > > > > > >-- > > > >Rick Venable 29/500 > > > >Membrane Biophysics Section > > > >NIH/NHLBI Lab. of Computational Biology > > > >Bethesda, MD 20892-8014 U.S.A. > > > >(301) 496-1905 venabler AT nhlbi*nih*gov > > > > > > > > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2!=!psu.edu] > > > > > > > > > > I have a limited experience in Molecular modelling of protein > > > > > structures; I'd love to expand my experience in that direction. > > > > > So, I have few questions for people with more experience in the > > > > > field. I wanted to install GROMACS package on my Mac-OS 10.4 > > > > > according to the instructions on the GROMACS website. > > > > > First, I installed the FFTW source as instructed in the README > > > > > file. Then, I installed the GROMACS binary for the MacOS-X 10.4. > > > > > However, it did not work! I tried the examples provided with the > > > > > package but I got "Segmentation faults"....nothing in the output. > > > > > Just a blank pages while the script is running interactively. > > > > > I tried to build the package myself from the source code; but > > > > > again I got the same error message when running the examples' > > > > > scripts! Presumably, my installation was not successful! > > > > > even running the command pdb2gmx gives that error message > > > > > "Segmentation fault"!!!-- > > > >M. L. Dodson > >Email: mldodson-at-houston-dot-rr-dot-com > >Phone: eight_three_two-56_three-386_one-- M. L. Dodson Email: mldodson-at-houston-dot-rr-dot-com Phone: eight_three_two-56_three-386_one From owner-chemistry@ccl.net Thu Nov 30 15:26:01 2006 From: "John H Block John.Block=oregonstate.edu" To: CCL Subject: CCL: Preliminary Announcement: QSAR Reborn Symposium, Boston ACS Meeting Message-Id: <-33128-061130151703-16962-VuoLzDvpiTG3C5WXY7inMQ===server.ccl.net> X-Original-From: "John H Block" Date: Thu, 30 Nov 2006 15:17:01 -0500 Sent to CCL by: "John H Block" [John.Block,+,oregonstate.edu] QSAR Reborn: A Symposium in Memory of Dr. Phillip Magee August 2007 American Chemical Society, Boston, MA Dr. Robert Clark (Tripos) and Dr. John H. Block (Oregon State University) are organizing a symposium honoring the late Dr. Phillip Magee, one of the pioneers in utilizing QSAR. Phil was one of the organizers of the International QSAR and Modeling Society and was its first president. While it is common to apply QSAR to bioactivities in humans, he showed its application to agrochemicals. After retiring from the Ortho Chemical Division of Standard Oil, Phil used QSAR to study transdermal properties of molecules. Until his stroke, Phil was utilizing both his knowledge of physical organic chemistry and QSAR descriptors to model bioactivities, and this symposium is designed to honor all work in the QSAR field. Speakers who have developed QSAR methodologies and who apply them to solve problems are encouraged to contact symposium organizers. Topics include QSAR Descriptors (Physico-chemical, Topological, Quantum Mechanics, Geometrical/Stereochemical); QSAR Techniques (2D-QSAR, 3D-QSAR, Topological QSAR, Neural networks, Classification Methods) and QSAR Applications (Drug design, Agrochemical design, Toxicities, Environmental properties). For additional information, please contact John Block at John.Block _ oregonstate.edu. John H. Block Phone: 541-737-5779 College of Pharmacy Fax: 541-737-3999 Oregon State University Corvallis, OR 97331 John.Block _ oregonstate.edu blockj _ onid.orst.edu From owner-chemistry@ccl.net Thu Nov 30 17:19:00 2006 From: "Daniel Jana dfjana]=[gmail.com" To: CCL Subject: CCL:G: AW: G: NBO visualization with GaussView Message-Id: <-33129-061130164526-25826-4FLbPSpVsD1BMYpLsWbdUQ###server.ccl.net> X-Original-From: "Daniel Jana" Content-Disposition: inline Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Thu, 30 Nov 2006 20:00:28 +0000 MIME-Version: 1.0 Sent to CCL by: "Daniel Jana" [dfjana||gmail.com] Hello, On 30/11/06, Markus Thut markus.thut%a%iac.unibe.ch wrote: > I think you have to turn your .chk file into a .fchk file - check the > Gaussian manual for further reference. No, GaussView also works with the check file directly. Why don't you just open the chk file directly instead of going through the log file? I think you can find what you need here. I did... http://educ.gaussian.com/visual/ESP/html/ESPGaussView.htm Cheers, Daniel Jana From owner-chemistry@ccl.net Thu Nov 30 18:49:00 2006 From: "Sengen Sun sengensun(0)yahoo.com" To: CCL Subject: CCL: A philosophical question for CCL'ers Message-Id: <-33130-061130184520-28821-hG58VXazQpuDwbExVCaUvw_._server.ccl.net> X-Original-From: Sengen Sun Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=iso-8859-1 Date: Thu, 30 Nov 2006 15:45:15 -0800 (PST) MIME-Version: 1.0 Sent to CCL by: Sengen Sun [sengensun^-^yahoo.com] > It is seems to me that science and philosophy are heading to a right direction if science is to question but not to worship what we already have. > I like David Shobe's comments. I would say it is hard to give the correct terms for what a simulation-based prediction and an understanding-based prediction. I don't think the terms already exist. The problem is that the former contains some of the latter, and the latter contains some of the former. That is, some quantitative simulation predictions do have understandable meaning while others do not, a case-by-case issue. > Eric Scerri does have valid questions that have not been addressed if he were not influenced by commercials or by any other non-scientific reasons. > "what extent can the resulting properties of compounds be predicted ab initio from current physical theory > from a knowledge of the properties of the contributing elements?" > It is also my question. Most people know such an existing philosophy that the properties of a molecule can be predicted from individual atoms. > In what extend does this philosophy is correct? Is this also a mathematical simulation process without physical meaning? Or it has a general or some limited physical meaning? > To me, there is an issue here how some people cannot get what others try to say (a gap between connotations and denotations?). This is a reality that theoretical chemists have problems to understand each other. > If we have a sincere attitude towards science, free of commercials (excluding those beneficial to run CCL and those governed by CCL rules) and politics, this List provides a very convenient platform to openly address this kind of issues, which cannot be otherwise addressed by creating a few more journals.... > In science, nothing should be free from analyzing and questioning, especially my (our) own sayings and behaviors in the public. So, please criticize me (us)-:)! > > Sengen -------------------------------------------------------------------------------- Shobe, David Wrote: Eric Scerri wrote: > Sodium a poisonous metal combines with chlorine, a greenish poisonous gas, > to yield something that is not only non-poisonous but essential for life - NaCl. > or that an explosive element like hydrogen combines with oxygen, the element that > is essential for combustion, to produce a compound that puts out fires - H2O. Eric, In both these cases, we have reactive elements combining to form products which are less reactive...and there is the thermodynamic argument of why this should normally be the case. (OTOH there are some counterexamples, like the mercury azides HgN6 and Hg2N6!) As to predicting properties of compounds, we may have to make a distinction. The predictions of computational chemistry are based on simulations, i.e. solving complicated sets of equations; these predictions are can be made to an arbitrary level of accuracy, but don't in themselves provide any insight as to why a compound has a given property. Then there are predictions based on causes and effects we can understand. It's one thing to be able to predict the geometries and cohesion energies of a water cluster by running a simulation, and another thing to predict them using one's knowledge of hydrogen bonds. I've come across the term epiphenomenon, meaning a phenomenon of a system which consists of a collection of subsystems, but which is not easily predicted from a (however detailed it may be) knowledge of the subsystems. The properties of solids and molecules are epiphenomena of the properties of atoms, which are in turn epiphenomena of the properties of positive integers. The acknowledgement that epiphenomena are common in natural systems provides a middle ground in the holism-reductionism debate: one can make a simulation-based prediction by considering the system by reducing it to a sum of its parts, but one cannot always make an understanding-based prediction of the system. Regards, --David Shobe Süd-Chemie, Inc. P.S. I would be interested in knowing the correct terms for what I called a simulation-based prediction and an understanding-based prediction. ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com From owner-chemistry@ccl.net Thu Nov 30 19:23:00 2006 From: "Vanessa Oklejas voklejas()stanford.edu" To: CCL Subject: CCL: makefile for installing delphi on RedHat v 7.3 Message-Id: <-33131-061130185638-10495-8q9ITy1fTy7pnOdSkMNtgg]~[server.ccl.net> X-Original-From: "Vanessa Oklejas" Date: Thu, 30 Nov 2006 18:56:38 -0500 Sent to CCL by: "Vanessa Oklejas" [voklejas(!)stanford.edu] Hello all, I would like to compile DelPhi (the electrostatics solver) on a system running RedHat v 7.3. The default makefile doesn't have the correct options, so when I run 'make' the compilation fails. Does anyone have a makefile for Delphi to compile on RedHat v 7.3? If not, can someone tell me how I need to modify the options of the existing makefile that I have (I've included it below). Muchos Gracias, Vanessa _____________________________________________________________________________ Now here's the makefile. (Note: the other makefile (called genericLinux.mk) specified in the current makefile contains the target, source, etc and is in the path. VPATH=/home/voklejas/DELPHI_2004_LINUX/src FC=/usr/bin/f77 CC=/usr/bin/cc RECFLAGS = -g -Kieee -Mrecursive -DLINUX FLAGS = -g -Kieee -DLINUX CFLAGS = -g -Kieee -DLINUX LFLAGS = -g -Kieee RECFLAGS = -O3 -Kieee -Mrecursive -DLINUX FLAGS = -O3 -Kieee -DLINUX -byteswapio FLAGS = -O3 -Kieee -DLINUX CFLAGS = -O3 -Kieee -DLINUX CFLAGS = -O3 -Kieee -DLINUX -byteswapio LFLAGS = -O3 include (VPATH)/genericLinux.mk From owner-chemistry@ccl.net Thu Nov 30 19:58:00 2006 From: "Daria Khvostichenko dkhvosti : uiuc.edu" To: CCL Subject: CCL:G: AW: G: NBO visualization with GaussView Message-Id: <-33132-061130181801-25913-hdmZkoUux/sNt3uqlXp7mw---server.ccl.net> X-Original-From: "Daria Khvostichenko" Date: Thu, 30 Nov 2006 18:18:01 -0500 Sent to CCL by: "Daria Khvostichenko" [dkhvosti(a)uiuc.edu] GaussView only reads chk files created by G03W. So if the checkpoint file was created by any other installation of Gaussian, you'll need to convert it to formatted checkpoint file in order to visualize anything: http://www.gaussian.com/g_tech/gv_diff.htm Also, if you would like to visualize NBO orbitals, remember to include pop=(nbo,savenbo) keyword in the route section. Daria From owner-chemistry@ccl.net Thu Nov 30 20:34:00 2006 From: "Ibrahim M. Moustafa I.moustafa*o*psu.edu" To: CCL Subject: CCL: GROMACS installation on Mac-OS Message-Id: <-33133-061130192958-8421-AACo/mpdprKz8QgjfYIiOA---server.ccl.net> X-Original-From: "Ibrahim M. Moustafa" Content-Type: text/plain; charset="us-ascii"; format=flowed Date: Thu, 30 Nov 2006 19:29:46 -0500 Mime-Version: 1.0 Sent to CCL by: "Ibrahim M. Moustafa" [I.moustafa^-^psu.edu] Yea, it segfaults only when I don't type the whole path name and this is what is puzzling me! Whether is it is a bug or something else faulty....I can't tell! If you get the chance to install GROMOCS on MacOS and managed to get around this, plz let me know! Kind regards, Ibrahim At 02:59 PM 11/30/2006, you wrote: >Sent to CCL by: "M. L. Dodson" [mldodson{=}houston.rr.com] >On Thursday 30 November 2006 11:40, Ibrahim M. Moustafa I.moustafa##psu.edu >wrote: > > Sent to CCL by: "Ibrahim M. Moustafa" [I.moustafa**psu.edu] > > > > Hi, > > > > Thanks Dodson for your reply. Actually, the $GMSBIN was already > > sitting in the $PATH prentenv confirmed that. > > As for the Segmentation fault error, it could be as you suggested! > > but if so, how does this related to the way I type the command? > > For the time being, I'm aliasing all the commands in $GMSBIN and it > > is working (at least for the demo)!. > > > > As for Amber, unfortunately, I don't have access to download it > > for the time being....same for Charmm. > > > >Hmmm... I Guess I misunderstood you about the segfault. Do you >mean it segfaults only when you DO NOT type the whole path name? >Or the other way around? In either case, I do not understand why >it should make any difference unless the pdb2gmx program is using >its argv[0] (executable file name) variable for some purpose. I >will shut up now as I know very little about GROMACS itself. I >was just answering with general Unix information. Segfaults can >happen for many different reasons. My suggestion was based on an >earlier installation of GROMACS on (Intel) FreeBSD where I had the >precision mismatch problem. > >If you still can not get it to work after other people put in >their 2 cents, let me know, and I will try to install GROMACS on >my iBook G4 to see if I can debug the problem. Should be really >funny to measure how much longer simulations take on the iBook :-) > >Both AMBER and CHARMM cost $$ as well (but are well worth it if >you qualify for the academic prices). > >Bud Dodson > > > thanks for your time and help, > > > > Ibrahim > > > > At 08:53 PM 11/29/2006, you wrote: > > >Sent to CCL by: "M. L. Dodson" [mldodson^houston.rr.com] > > >On Wednesday 29 November 2006 16:30, Ibrahim M. Moustafa ria2(a)psu.edu > > > > > >wrote: > > > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2|a|psu.edu] > > > > Hi everybody, > > > > > > > > Thanks Rick for your advice; but actually my Xcode was recently > > > > updated! I'm kind of figured out what was the problem which is: after > > > > installation and setting the variables I got an error message > > > > "Segmentation fault" when trying to issue any command...or trying the > > > > demo tutorial! > > > > It appeared that my installation was fine (even my compilation was > > > > alright......in fact compilation resulted in a code that is much > > > > faster than the provided binary for Mac-OSX). > > > > I found the whole path is required to issue GROMACS' command to run > > > > any program; although, the variable $GMXBIN is already set in my > > > > .cshrc file. This has solved the problem, but still I don't > > > > understand why I can't run the programs (e.g. pdb2gmx) by just typing > > > > "pdb2gmx" in the xterm rather than typing "$GMXBIN/pdb2gmx" to make > > > > it working!! I'm not sure if this is a an OS problem or GROMACS > > > > problem! or something obvious that I'm missing! Any advice would be > > > > appreciated. > > > > > > > > I haven't heard any comments regarding TINKER vs GROMACS as a > > > > modelling package! wish I can hear from someone though! > > > > > > > > thanks, > > > > Ibrahim > > > > > >Try issuing the command > > > > > >printenv PATH > > > > > >and look to see if the $GMXBIN directory is in your path. The > > >behavior you describe indicates it is probably not there. Fix > > >your .cshrc to include it, and your "full path problem" should go > > >away. The seg fault problem may indicate that your have some > > >component(s) compiled for single precision and other components > > >compiled for double precision. On FreeBSD (at least on Intel > > >architectures), this can be confusing at times. GROMACS gets a > > >lot of its speed by running in single precision, whereas most > > >other MD packages are compiled for double precision. In > > >particular, make sure the compiled precisions match for your > > >GROMACS and FFTW installations. > > > > > >As to your TINKER vs GROMACS question, I prefer AMBER and namd :-) > > > > > >Bud Dodson > > > > > > > At 12:22 AM 11/28/2006, you wrote: > > > > >Sent to CCL by: Rick Venable [venabler(_)nhlbi.nih.gov] > > > > >There can be issues with the gcc and g77 versions under MacOS X; in > > > > >particular, if you've kept up to date with OS patches, you may need > > > > > to download a newer version of the Xcode tools from the Apple > > > > > Developer site. > > > > > > > > > >I had some problems compiling CHARMM until I updated my Xcode tools. > > > > > > > > > >-- > > > > >Rick Venable 29/500 > > > > >Membrane Biophysics Section > > > > >NIH/NHLBI Lab. of Computational Biology > > > > >Bethesda, MD 20892-8014 U.S.A. > > > > >(301) 496-1905 venabler AT nhlbi*nih*gov > > > > > > > > > > > Sent to CCL by: "Ibrahim M. Moustafa" [ria2!=!psu.edu] > > > > > > > > > > > > I have a limited experience in Molecular modelling of protein > > > > > > structures; I'd love to expand my experience in that direction. > > > > > > So, I have few questions for people with more experience in the > > > > > > field. I wanted to install GROMACS package on my Mac-OS 10.4 > > > > > > according to the instructions on the GROMACS website. > > > > > > First, I installed the FFTW source as instructed in the README > > > > > > file. Then, I installed the GROMACS binary for the MacOS-X 10.4. > > > > > > However, it did not work! I tried the examples provided with the > > > > > > package but I got "Segmentation faults"....nothing in the output. > > > > > > Just a blank pages while the script is running interactively. > > > > > > I tried to build the package myself from the source code; but > > > > > > again I got the same error message when running the examples' > > > > > > scripts! Presumably, my installation was not successful! > > > > > > even running the command pdb2gmx gives that error message > > > > > > "Segmentation fault"!!!-- > > > > > >M. L. Dodson > > >Email: mldodson-at-houston-dot-rr-dot-com > > >Phone: eight_three_two-56_three-386_one-- >M. L. Dodson >Email: mldodson-at-houston-dot-rr-dot-com >Phone: eight_three_two-56_three-386_one From owner-chemistry@ccl.net Thu Nov 30 21:09:00 2006 From: "Adil Zhugralin zhugrali{:}bc.edu" To: CCL Subject: CCL:G: Summary: NBO visualization with GaussView Message-Id: <-33134-061130203216-6718-VYyAPpYh18JHVmoLl/4ALA\a/server.ccl.net> X-Original-From: "Adil Zhugralin" Date: Thu, 30 Nov 2006 20:32:16 -0500 Sent to CCL by: "Adil Zhugralin" [zhugrali||bc.edu] Dear CCLers, Thank you all. Basically, the problem was that the job was run on a remote cluster and I was trying to open the .chk file copied from the cluster without first turning it into .fchk file. After I ran the job locally I could open the .chk file directly. Alternatively, I had to run "formchk 1.chk 1.fchk" on the remote cluster, transfer 1.fchk to my computer and visualize it with GaussView. To summarize everybody's responses: ---- No, GaussView also works with the check file directly. Why don't you just open the chk file directly instead of going through the log file? I think you can find what you need here. I did... http://educ.gaussian.com/visual/ESP/html/ESPGaussView.htm Cheers, Daniel Jana ---- I would suggest you read the chk file directly into gv. The other way to do is to change chk to fchk, then use the "cubegen" utility to make .cube files which can be opened in a variety of progs (molekel, molden). regards James ---- Hi Adil I think you have to turn your .chk file into a .fchk file - check the Gaussian manual for further reference. Regards Markus ---- Try opening the checkpoint file directly, instead of through the MO editor. The checkpoint file has all of the structure and will load the MO's directly. Does this work? If not view the Calculate/Current Jobs and and select the Log button to see if there was an error message running formchk. Just to confirm, did you generate all of the data on the machine where you are running GaussView? Or was G03 run on a different machine and you moved the files to this machine? If so you should run formchk on the machine where G03 ran and move the .fchk file to be opened by GaussView. Douglas Fox help#%#gaussian.com ---- Adil From owner-chemistry@ccl.net Thu Nov 30 21:43:00 2006 From: "X Yang xyang[*]chem.siu.edu" To: CCL Subject: CCL:G: AW: G: NBO visualization with GaussView Message-Id: <-33135-061130204954-21801-INQ7zZNhrbIAsRuWdxZAvA++server.ccl.net> X-Original-From: "X Yang" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="US-ASCII" Date: Thu, 30 Nov 2006 19:48:29 -0600 MIME-Version: 1.0 Sent to CCL by: "X Yang" [xyang _ chem.siu.edu] Yes, GView can load .chk file directly and let you see MOs from there. That's what I found out after wasting at least 2 days (the total amount of time I spent) to figure out why the FORMCHK of G03W does not work. Don't know what is required to see NBO though. This CCL is so great! There are always very nice ppl here answering questions for beginners. MY thanks go to those kind persons who helped me on a very simple question a few days ago. Jaimee =========================================== Xiaomei Yang, Ph.D. The Laboratory for Nanoscale Exploration and Single-Molecule Imaging (NESMI) Department of Chemistry and Biochemistry Southern Illinois University, MC 4409 Carbondale, IL 62901 =========================================== -----Original Message----- > From: owner-chemistry*|*ccl.net [mailto:owner-chemistry*|*ccl.net] Sent: Thursday, November 30, 2006 4:22 PM To: Jiang, Jaimee-Ann Subject: CCL:G: AW: G: NBO visualization with GaussView Sent to CCL by: "Daniel Jana" [dfjana||gmail.com] Hello, On 30/11/06, Markus Thut markus.thut%a%iac.unibe.ch wrote: > I think you have to turn your .chk file into a .fchk file - check the > Gaussian manual for further reference. No, GaussView also works with the check file directly. Why don't you just open the chk file directly instead of going through the log file? I think you can find what you need here. I did... http://educ.gaussian.com/visual/ESP/html/ESPGaussView.htm Cheers, Daniel Janahttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt