From owner-chemistry@ccl.net Mon Apr 27 04:32:01 2009 From: "Chaofu Wu xiaowu759%hotmail.com" To: CCL Subject: CCL: which structure in NPT MD trjectory to start another NVT MD? Message-Id: <-39172-090427043019-13524-HGhDA9FU0WDpUE+4hnELzg:+:server.ccl.net> X-Original-From: "Chaofu Wu" Date: Mon, 27 Apr 2009 04:30:15 -0400 Sent to CCL by: "Chaofu Wu" [xiaowu759++hotmail.com] Dear ccl users, Usually, we perform NPT MD to achieve optimized density of system and start another NVT MD to calculate other properties of interest. I think there are three possible ways to choose the structure to start another NVT MD:(1) the structure of last step in NPT MD trjectory;(2)the structure with the density closest to the averaging density;(3)the structure of last step in NPT MD trjectory but the box size needs to be rescaled so that it is the averaging size. My question is: which way should be chosen to start another NVT MD? Any hints would be thanked very much. Best wishes, xiaowu From owner-chemistry@ccl.net Mon Apr 27 05:15:01 2009 From: "Jens Spanget-Larsen spanget#,#ruc.dk" To: CCL Subject: CCL: Diffuse function as a necessay evil Message-Id: <-39173-090427043841-18242-vdBCH5+07ztIRJy5U6Cy9g:+:server.ccl.net> X-Original-From: Jens Spanget-Larsen Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Mon, 27 Apr 2009 09:46:08 +0200 MIME-Version: 1.0 Sent to CCL by: Jens Spanget-Larsen [spanget|a|ruc.dk] Dear Mi! Try the option SCF=QC. In my experience, this generally solves scf convergence problems with diffuse basis functions. Yours, Jens >--< ------------------------------------------------------ JENS SPANGET-LARSEN Office: +45 4674 2710 Dept. of Science (18.1) Fax: +45 4674 3011 Roskilde University Mobile: +45 2320 6246 P.O.Box 260 E-Mail: spanget{=}ruc.dk DK-4000 Roskilde, Denmark http://www.ruc.dk/~spanget ------------------------------------------------------ mi yang agri_chemist(_)yahoo.com wrote: > Sent to CCL by: "mi yang" [agri_chemist^yahoo.com] > Hello CCL colleagues, > > I want to discuss about basis set with diffuse function. Normally, with anionic or excess electron species during the calculation of CIS (Opt) or TD-DFT the diffuse function is consider essential but often it very difficultly converge the SCF (sometimes not at all) even with systems less than three benzene rings. > > Now My questions are: > > 1- I am dealing with optical properties of halogen ions (medium size systems) and often use 6-31+G* basis set in which diffuse function is often proves for me a necessary evil. Is there other equivalent basis set having diffuse function and also easy to converge in CIS or TD-DFT calculations? > > 2- Now my idea is How it would be If I will use the split basis set (Gen) I mean diffuse function for halogen ions only and without diffuse for other atoms in a system? > > Any related material or comment would be highly apriciated. > > Mi > agri_chemist++yahoo.com> > > From owner-chemistry@ccl.net Mon Apr 27 05:41:00 2009 From: "Lukasz Cwiklik cwiklik|,|gmail.com" To: CCL Subject: CCL: Computational Chemistry Wiki Message-Id: <-39174-090427045009-7693-kkED8Vpic2zePE35rY/hUw..server.ccl.net> X-Original-From: Lukasz Cwiklik Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Date: Mon, 27 Apr 2009 11:19:41 +0300 MIME-Version: 1.0 Sent to CCL by: Lukasz Cwiklik [cwiklik++gmail.com] On Fri, Apr 24, 2009 at 12:22 PM, Sebastian Kozuch kozuchs^yahoo.com wrote: > The idea of a specific computational chemistry wiki is not to undermine o= r > compete with Wikipedia. The idea is to complement Wikipedia, blogs, books > and papers. Wikipedia is extremely helpful, and if I had the knowledge I > would have loved to contribute there. More than that, as far as I=92m > concerned if you look for something basic like DFT in the new wiki there > should be only a reference to the entry in Wikipedia. > > But I cannot expect Wikipedia to have a specific entry on, let=92s say, B= P86 > or UFF. Neither I suppose we can add there a table with the mean error of > the different post-HF geometry bond lengths, or a list of the programs th= at > can run M06. But most of all, I=92d like an up to date encyclopedia where= if I > search for =93MP4=94 it won=92t appear with a multimedia player. > > And, as Brian says, Wikipedia is not for self-publicity, and most of the = new > methods that appear in computational chemistry cannot enter there unless > they are really fundamental methods. This takes years, and leave most > methods outside. > > Again, for those that are real specialists on traditional methods, I > encourage them to write in Wikipedia. But all the information that can be > useful only for computational chemists, I encourage to put it in this > specific wiki. > > Take into account that the same people of Wikipedia support the creation = of > specific encyclopedias for specific communities. > > Again, if you like the idea and want to contribute, the page is > > > > http://computationalchemistry.wikia.com > > Maybe it will work, maybe not. But for sure it's worth to try it. Dear Sebastian, Brian and others, I agree, you convinced me that it would be useful to develop a specialized wiki keeping in mind that we all should also disseminate our knowledge in Wikipedia. People can definitely benefit from the combination of these both sources. Best, Lukasz --=20 Lukasz Cwiklik http://cwiklik.wordpress.com From owner-chemistry@ccl.net Mon Apr 27 06:16:01 2009 From: "mi yang agri_chemist() yahoo.com" To: CCL Subject: CCL:G: Re: Diffuse function as a necessay evil Message-Id: <-39175-090427023704-30935-GqiSxiF+vfF553cNNeK8bw^^server.ccl.net> X-Original-From: "mi yang" Date: Mon, 27 Apr 2009 02:37:01 -0400 Sent to CCL by: "mi yang" [agri_chemist*|*yahoo.com] > Sent to CCL by: "mi yang" [agri_chemist^yahoo.com] Hello CCL colleagues, I want to discuss about basis set with diffuse function. Normally, with anionic or excess electron species during the calculation of CIS (Opt) or TD-DFT the diffuse function is consider essential but often it very difficultly converge the SCF (sometimes not at all) even with systems less than three benzene rings. Now My questions are: 1- I am dealing with optical properties of halogen ions (medium size systems) and often use 6-31+G* basis set in which diffuse function is often proves for me a necessary evil. Is there other equivalent basis set having diffuse function and also easy to converge in CIS or TD-DFT calculations? 2- Now my idea is How it would be If I will use the split basis set (Gen) I mean diffuse function for halogen ions only and without diffuse for other atoms in a system? Any related material or comment would be highly apriciated. Mi agri_chemist[#]yahoo.com **************************************************************** "Dipankar Roy dipankarroy%a%iitb.ac.in" wrote: > > Sent to CCL by: "Dipankar Roy" [dipankarroy{=}iitb.ac.in] > Hi, > You can use Duuning's augmented basis sets. > For Gaussian program, you can use Gen keyword for defining different basis > set for different atoms. Gaussian web sites and manual contains detailed > explanation for the same. > > Hope this may help, > Dipankar Roy > > > *********************************************** > Dipankar Roy > Graduate Student of Prof. R. B. Sunoj > Computational Chemistry Laboratory > Dept. of Chemistry > Indian Institute of Technology, Bombay > India - 400076 > Phone: +91-22-2576-4130(lab) ******************************************************** Dear Colleague! Thank you for reply You mean to say that Dunning is more efficient and economical from computational cost but there are different types of Dunnig basis sets on EMSL e.g. Dunning-Hay Diffuse, "DZ (Dunning)" or "DZP + Diffuse (Dunning)" which can be equivalet (round about) to 6-31+G*. Secondly, I knew the use of Gen basis set but my question was How it would be If I will use the split basis set (Gen) I mean diffuse function for halogen ions only and without diffuse for other atoms in a system? Because I didnt find any work in which Gen is used for separation of diffuse function? agri_chemist###yahoo.com From owner-chemistry@ccl.net Mon Apr 27 06:52:00 2009 From: "kalpana krishnaswami kalpuk]=[gmail.com" To: CCL Subject: CCL: building models for higher order mers Message-Id: <-39176-090423141411-11389-KfWDD30Xx/DNBQNADrfKhA:-:server.ccl.net> X-Original-From: "kalpana krishnaswami" Date: Thu, 23 Apr 2009 14:14:08 -0400 Sent to CCL by: "kalpana krishnaswami" [kalpuk**gmail.com] Hi: I am working with a protein that forms dimers and higher order oligomers. I have a modeled structure for the protein and some idea about the surfaces that might be involved in the oligomerisation process. I would like to build a dimer model. Can any one give me any pointers or references. I would like to be able to use open source software to build this. Any help would be appreciated. Regards Kalpana From owner-chemistry@ccl.net Mon Apr 27 07:27:01 2009 From: "Alberto Sergio Garay sgaray],[fbcb.unl.edu.ar" To: CCL Subject: CCL: Question about ddikick.x error in Gamess-US software Message-Id: <-39177-090426182029-20502-8ioe+jfcGJnTQQtXcplU3w,+,server.ccl.net> X-Original-From: Alberto Sergio Garay Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes"; format="flowed" Date: Sun, 26 Apr 2009 18:44:15 -0300 MIME-Version: 1.0 Sent to CCL by: Alberto Sergio Garay [sgaray[*]fbcb.unl.edu.ar] Hi all I am trying to run a Gamess-US quantum calculation in my PC, but I take a ddikick.x error: ddikick.x : Unable to create thread to asynchronous accept socket connection= s. ddikick.x: Fatal error detected. The error is most likely to be in the application, so check for input errors, disk space, memory needs, application bugs, etc. ddikick.x will now clean up all processes, and exit... ddikick.x: Sending kill signal to DDI processes. ddikick.x: Execution terminated due to error(s). Could anyone give any advice or clue about this problem? --=20 Dr. Sergio Garay Facultad de Bioquimica y Cs. Biol=F3gicas Universidad Nacional del Litoral Santa Fe - Argentina C.C. 242 - Ciudad Universitaria - C.P. S3000ZAA Argentina Ph. +54 (342) 4575-213 Fax. +54 (342) 4575-221 From owner-chemistry@ccl.net Mon Apr 27 08:12:02 2009 From: "Lukasz Cwiklik cwiklik%gmail.com" To: CCL Subject: CCL: which structure in NPT MD trjectory to start another NVT MD? Message-Id: <-39178-090427080947-11705-nguazxSJBA0me9QGDqb06Q**server.ccl.net> X-Original-From: Lukasz Cwiklik Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=ISO-8859-1 Date: Mon, 27 Apr 2009 15:09:33 +0300 MIME-Version: 1.0 Sent to CCL by: Lukasz Cwiklik [cwiklik*_*gmail.com] On Mon, Apr 27, 2009 at 11:30 AM, Chaofu Wu xiaowu759%hotmail.com wrote: > > Sent to CCL by: "Chaofu =A0Wu" [xiaowu759++hotmail.com] > Dear ccl users, > > =A0Usually, we perform NPT MD to achieve optimized density of system and = start another NVT MD to calculate other properties of interest. =A0I think = there are three possible ways to choose the structure to start another NVT = MD:(1) the structure of last step in NPT MD trajectory;(2)the structure wit= h the density closest to the averaging density;(3)the structure of last ste= p in NPT MD trajectory but the box size needs to be rescaled so that it is = the averaging size. My question is: which way should be chosen to start ano= ther NVT MD? Any hints would be thanked very much. Dear Wu, As you pointed, one usually performs NPT only to get a proper density, i.e, box size for a given number of molecules, and an instantaneous configuration of particles in the box is of less importance. So formally one should take an average box size from an equilibrated NPT and start NVT. Now, concerning the ways you described: (1) not formally exact, but in practice it is being used because it avoids possible close-contact problems (2) formally the best way (the structure must be taken from an already equlibrated part of the trajectory) (3) formally exact, but close-contacts can occur if your rescaling would decrease a box size or a long equlibration may be needed if the box was expanded. In practice, for well equlibrated liquid systems, if (2) is too troublesome one can use method (1) if volume fluctuations are not very large. It can cause changes in the temperature but it can be easily controlled. More problematic would be if one wants to go from NPT to NVE but even then the well-behaving temperature would be a good indicator that everything is ok in your system. Best, Lukasz --=20 Lukasz Cwiklik http://cwiklik.wordpress.com From owner-chemistry@ccl.net Mon Apr 27 08:47:00 2009 From: "Toomas Tamm tt-ccl2!A!yki.ttu.ee" To: CCL Subject: CCL:G: Calculation of Redox Potentials Message-Id: <-39179-090427063646-31729-r0BS1TowSSMtKhSvt0EEhA[*]server.ccl.net> X-Original-From: Toomas Tamm Content-Transfer-Encoding: 7bit Content-Type: text/plain Date: Mon, 27 Apr 2009 12:35:16 +0300 Mime-Version: 1.0 Sent to CCL by: Toomas Tamm [tt-ccl2(_)yki.ttu.ee] On Fri, 2009-04-24 at 16:08 +0100, Tobias Kraemer tobiask * chem.gla.ac.uk wrote: > I'm trying to get experience in calculating some redox potentials of > copper complexes to refer these data to experimental values. How confident are you about your experimental values? Our experience (J. Phys. Chem. A, 107, 9997 (2003)) showed that even "well established" E^0 values quoted in many reference books may turn out to be unreliable upon close scrutiny. > I have done a number of initial calculations on a test set of > organometallic molecules, to get used to the method, using B3LYP/6-31G* > and the default PCM model as implemented in Gaussian with methanol as > solvent. Our tests (not published) showed that B3LYP was in no ways better than the BP86 we used. I think that the transition metals are not described well enough by the B3LYP functional. > ... and in a next step increased the > basis set to TZVP for copper and ligating atoms, and SVP for remaining > ligand atoms. Have you tried to introduce additional solvent molecules? In our work this was very beneficial. > After reading some papers about the topic, I feel that the use of Jaguar > for calculations of this type is preferred. This causes some difficulty > in reproducing results, as I'm using > Gaussian. Except for possible nuances of implementation (integration grids, tesselation of the PCM surface, convergence criteria, etc) I do not think the software should make any difference in results. Computation timings may differ and this may be significant. If the timings are important, have you tried the density fitting / resolution of identity approximation? You need a pure DFT functional (eg BP86, not the hybrid B3LYP) to make use of this in Gaussian. > I would especially like to know, if it is advisable to do full > optimisations in solvent, to improve the results, Yes. If your system is conformationally flexible, you may need to check alternative conformations (perhaps separated from the gas-phase minimum by a barrier) as well. -- Toomas Tamm e-mail: tt-ccl2 (at) yki . ttu . ee Chair of Inorganic Chemistry voice: INT+372-620-2810 Tallinn University of Technology fax: INT+372-620-2828 Ehitajate tee 5, EE-19086 Tallinn, Estonia http://www.kk.ttu.ee/toomas/ From owner-chemistry@ccl.net Mon Apr 27 10:32:02 2009 From: "Allan Tucker allan.tucker!A!brunel.ac.uk" To: CCL Subject: CCL: Systems Biology Symposium 25 - 26 June at Brunel University London Message-Id: <-39180-090427093734-2759-HSWboc7Wh0V/WCo+cpJP2Q]~[server.ccl.net> X-Original-From: "Allan Tucker" Date: Mon, 27 Apr 2009 09:37:29 -0400 Sent to CCL by: "Allan Tucker" [allan.tucker**brunel.ac.uk] Symposium on Biological and Chemical Informatics for Health: Foundations for Systems Biology Brunel University, West London, UK 25-26 June 2009 Find out more and register for this free event here: http://bcih-sb.brunel.ac.uk Integration of biological and chemical informatics presents a real opportunity to address some of the most exciting and challenging issues posed by systems biology. These diverse disciplines, and their interaction, play a crucial role in addressing key issues in drug design and discovery, environmental sustainability, health and disease processes, and disease prevention, diagnosis and treatment. Join us for a two day symposium sponsored by the BBSRC and Brunel University, where influential leaders in bioinformatics, chemoinformatics, systems biology and health, will present the latest developments that are helping to shape this interdisciplinary field. This symposium is open to those working in the field, including academics, researchers, students and industry and will provide an excellent environment for delegates to participate in networking activities and developing collaborative, interdisciplinary research partnerships. We are delighted that the following have already agreed to speak at the event: Bioinformatics Janet Thornton, Director of the European Bioinformatics Institute Marco Ramoni, Director of Biomedical Cybernetics Laboratory, Harvard Medical School Christine Orengo, Professor of Bioinformatics, University College London Health Sally Davies, Director General of Research and Development and Chief Scientific Adviser for the Department of Health and NHS Chemoinformatics Peter Willett, Professor of Information Science, Sheffield University Robert Glen, Director of Unilever Centre for Molecular Science Informatics, Cambridge University Peter Johnson, Professor of Computational Chemistry, Leeds University Systems Biology Tom Kirkwood, Director of the Newcastle Centre for Integrated Systems Biology of Ageing and Nutrition Simon Gaskell, Vice-President for Research, Manchester University Chris Rawlings, Head of Department of Biomathematics and Bioinformatics, Rothamsted Research, BBSRC David Gilbert, Head of School of Information Systems, Computing and Mathematics, Brunel University Colin Miles, Head of Systems Biology, BBSRC Stephen Muggleton, Director of Modelling, Centre for Integrative Systems Biology at Imperial College Peter Bramley, Head of School of Biological Sciences, Royal Holloway London For more details and free registration see: http://bcih-sb.brunel.ac.uk From owner-chemistry@ccl.net Mon Apr 27 12:54:01 2009 From: "Venable, Richard (NIH/NHLBI) E venabler[a]nhlbi.nih.gov" To: CCL Subject: CCL: which structure in NPT MD trjectory to start another NVT MD? Message-Id: <-39181-090427124954-5684-vZvWhOQx3glZfrQZuA3Gsw .. server.ccl.net> X-Original-From: "Venable, Richard (NIH/NHLBI) [E]" Content-Language: en Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="iso-8859-1" Date: Mon, 27 Apr 2009 12:49:40 -0400 MIME-Version: 1.0 Sent to CCL by: "Venable, Richard (NIH/NHLBI) [E]" [venabler ~ nhlbi.nih.gov] I've generally used method (2), taking care to discard any heating and init= ial equilibration data from the calculation of the average. Also, I usuall= y try to find a match from the last 500 ps or so of the NPT MD. That way, = a coord set near the end of the NPT with a size very close to the average i= s used. -- Rick Venable 5635FL/T906 Membrane Biophysics Section NIH/NHLBI Lab. of Computational Biology Bethesda, MD 20892-9314 U.S.A. (301) 496-1905 venabler AT nhlbi*nih*gov Sent to CCL by: "Chaofu Wu" [xiaowu759++hotmail.com] Dear ccl users, Usually, we perform NPT MD to achieve optimized density of system and sta= rt another NVT MD to calculate other properties of interest. I think there= are three possible ways to choose the structure to start another NVT MD:(1= ) the structure of last step in NPT MD trjectory;(2)the structure with the = density closest to the averaging density;(3)the structure of last step in N= PT MD trjectory but the box size needs to be rescaled so that it is the ave= raging size. My question is: which way should be chosen to start another NV= T MD? Any hints would be thanked very much. Best wishes, xiaowu From owner-chemistry@ccl.net Mon Apr 27 13:29:01 2009 From: "Elaine Meng meng+*+cgl.ucsf.edu" To: CCL Subject: CCL: building models for higher order mers Message-Id: <-39182-090427120250-15319-kkED8Vpic2zePE35rY/hUw*_*server.ccl.net> X-Original-From: "Elaine Meng" Date: Mon, 27 Apr 2009 12:02:44 -0400 Sent to CCL by: "Elaine Meng" [meng]![cgl.ucsf.edu] Hi Kalpana, Chimera includes several ways to build multimers if your PDB file includes matrices or other symmetry information. (a) to use BIOMT information in the PDB file, you could use the Multiscale Models tool or command "sym" http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/multiscale/framemulti.html http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/midas/sym.html (b) to use other symmetry information in the PDB file (SMTRY, CRYST1, MTRIX), you could use Multiscale Models or Unit Cell http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/unitcell/unitcell.html However, since your protein is modeled you might not have this information in the file. If the template structure used for modeling has such information and you believe multimerization is generally the same for that protein, you could try copying the information from the template into your model file. Another possibility is to create the multimer of the template and then fit copies of the model to the copies of the template. http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/superposition.html Actually, depending on the modeling process, perhaps you could create a multimeric model directly from the multimeric template structure. I hope this helps, Elaine P.S. Chimera is free for noncommercial use, download includes all the python code, and the C++ stuff is also available for download. http://www.cgl.ucsf.edu/chimera/index.html ----- Elaine C. Meng, Ph.D. meng(~)cgl.ucsf.edu UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html From owner-chemistry@ccl.net Mon Apr 27 18:07:00 2009 From: "Marc Baaden baaden(!)smplinux.de" To: CCL Subject: CCL: Programs for molecular surface calculations and a question about QCPE Message-Id: <-39183-090427180211-6134-Hn8UzV3N3uiBa2JvLbsOYA||server.ccl.net> X-Original-From: "Marc Baaden" Date: Mon, 27 Apr 2009 18:02:07 -0400 Sent to CCL by: "Marc Baaden" [baaden---smplinux.de] Dear CCL, I am currently looking for programs performing molecular surface calculations. Of course there are lots of them as well as a wealth of algorithms. I do however have a number of restrictions which seem to limit the number of candidates (I haven't found any so far..): - I want to use the software within another (open-source) project, so source should be available, modifiable and redistributable - performance is very important - parallelisation/parallelizability would be great - approximation is acceptable - the code should determine a mesh, ie vertices, faces and normals I'd be very grateful for any suggestions. Thank you in advance! Marc Baaden By the way, I was also testing programs referred to in publications etc. and one of those is MSEED. It is supposed to be available via QCPE, but I actually couldn't find any working webpage with access to QCPE programs. Has QCPE disappeared?