From owner-chemistry@ccl.net Thu Aug 31 03:54:01 2006 From: "Andreas Klamt klamt * cosmologic.de" To: CCL Subject: CCL: Bad calculated geometries in CCCBDB Message-Id: <-32459-060831034639-14389-VbnlrxofR0lpINGtJSnhRg|a|server.ccl.net> X-Original-From: Andreas Klamt Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-15 Date: Thu, 31 Aug 2006 09:46:24 +0200 MIME-Version: 1.0 Sent to CCL by: Andreas Klamt [klamt#%#cosmologic.de] Hi all, the CCL note on databases this morning let me look into the CCCBDB again. I humbled into the section "Bad calculated geometries", curious to see whether a few molecules which tend to make us problems in our COSMO-RS method may suffer from bad geometries. Looking into the table I was shocked to see compounds like "Ethane, 1,1,1-trichloro-" . The page suggests that all quantum methods mis-predict the product of the moments of inertia by roughly a factor of 70.000 !!! Looking into details, it turns out that the experimental listing of the moments of inertia (just one value) wants to make us believe that this molecule is linear. I have my doubts. Indeed, the geometry reported as experimental geometry fits quite well with the calc. data, no real surprise. I got the impression that the same kind of error is present for many other molecules in that list. In these cases apparently the exp. moments of inertia do not fit together with the reported exp. geometries. Nevertheless, the database lists them as "bad calculated geometries". The database suggests as reasons for the bad geometries just calculation errors. I believe, that meanwhile computational chemistry has reached a status where we should dare to have a second, critical look at the exp. data, when we find strong discrepancies between exp. and good quality calculations on several levels of theory, and especially before we claim that all computational results are wrong. In this sense it would be a good idea to reduce the list of "Bad calculated geometries" to those where really the exp. product of inertia is in agreement with the exp. geometry, and deviates from the calculated data. Otherwise this list is rather confusing and worthless, and sheds a bad light on the work of the comp. chem. community. I just had a look to the "Bad Calculated Bond Lengths" list and clicked on AM1. This lists first the CC-single bond in "2-Butene, (Z)-" as mis-calculated by 1 Angstrom. Then it lists "2H-Thiopyran, tetrahydro-" with 2 "mis-calculated" bond lengths in the order of 1 Angstrom. Apparently, in both cases atom numbering was mixed up when analyzing the AM1 results. How much shall we trust the rest of the list, if even the leading entries are rubbish? Regards Andreas (P.S.: Let me just point out that I consider the CCCBSD altogether as a really valuable effort!) -- ----------------------------------------------------------------------------- Dr. habil. Andreas Klamt COSMOlogic GmbH&CoKG Burscheider Str. 515 51381 Leverkusen, Germany Tel.: +49-2171-73168-1 Fax: +49-2171-73168-9 e-mail: klamt+*+cosmologic.de web: www.cosmologic.de ----------------------------------------------------------------------------- COSMOlogic Your Competent Partner for Computational Chemistry and Fluid Thermodynamics ----------------------------------------------------------------------------- From owner-chemistry@ccl.net Thu Aug 31 08:31:01 2006 From: "Marcel Swart marcel.swart###udg.es" To: CCL Subject: CCL: Mulliken Charges and Augmented Functions Message-Id: <-32460-060831065153-11908-cGGj1xLM/NMeE68txuzLNg]_[server.ccl.net> X-Original-From: Marcel Swart Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=WINDOWS-1252; delsp=yes; format=flowed Date: Thu, 31 Aug 2006 09:09:16 +0200 Mime-Version: 1.0 (Apple Message framework v752.2) Sent to CCL by: Marcel Swart [marcel.swart() udg.es] See for instance: C. Fonseca Guerra, J.-W. Handgraaf, E. J. Baerends, F. M. Bickelhaupt, "Voronoi Deformation Density (VDD) charges. Assessment of the Mulliken, Bader, Hirshfeld, Weinhold and VDD methods for Charge Analysis", J. Comput. Chem. 2004, 25, 189-210 Specifically look at Figure 2, which is very illustrative. On Aug 31, 2006, at 4:43 AM, Abrash, Samuel sabrash * richmond.edu wrote: > Sent to CCL by: "Abrash, Samuel" [sabrash _ richmond.edu] > Hi to All CCLers, > > Recently there was a discussion on how adding augmented functions to a > basis set made Mulliken charge distributions (even) less accurate. > Can > anyone point me to a reference, either a book or an article, that > discusses this problem? > > Thank you. > > Samuel A. Abrash > Department of Chemistry > University of Richmond > Richmond, VA 23173 > Phone: 804-289-8248 > Fax: 804-287-1897 > E-mail: sabrash:_:richmond.edu > Web-page: http://oncampus.richmond.edu/~sabrash > > "The time it takes to do a scientific project is given by the > equation t > = 2a + b, where a is the original estimate of the length of the > project, > 2 is a correction factor, and b is a number large compared to 2a." > > > > -= This is automatically added to each message by the mailing > script =- > To recover the email address of the author of the message, please > change> Conferences: http://server.ccl.net/chemistry/announcements/ > conferences/ > > Search Messages: http://www.ccl.net/htdig (login: ccl, Password: > search)> > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- > +-+-+ > > > –––––––––––––––––––––––––––––––––––––––––––– dr. Marcel Swart ICREA researcher at Institut de Química Computacional Universitat de Girona Campus Montilivi 17071 Girona Catalunya (Spain) F +34-972-418356 E marcel.swart-,-icrea.es E marcel.swart-,-udg.es W http://www.icrea.es/pag.asp?id=Marcel.Swart W http://iqc.udg.es/~marcel –––––––––––––––––––––––––––––––––––––––––––– From owner-chemistry@ccl.net Thu Aug 31 12:02:00 2006 From: "Tapan Gupta tapanjames||yahoo.com" To: CCL Subject: CCL:G: Selecting the Cheap and Best Computer Configuration for Gaussian03 Message-Id: <-32461-060831120007-13035-XZRDMublr/wV0SdRz5ONvw*|*server.ccl.net> X-Original-From: "Tapan Gupta" Date: Thu, 31 Aug 2006 12:00:06 -0400 Sent to CCL by: "Tapan Gupta" [tapanjames.[-].yahoo.com] Dear Users, I have started using Gaussian 03 for windows. Presently I am having a single processor Computer with following configuration Intel Pentium 2.66 GHz, 1 GB of Ram, 160 Gb of Hard Disk. Now I am planning to change my computer. My Question is that "IS AMD ATHALON a better option than pentium for using Gaussian 03" ? Can anybody please tell me a better configuration in detail ? Also I would like to know that whether Gaussian for windows is better or gaussian for Linux is better IN ALL POSSIBLE WAYS AND HOW IT WORKS BETTER WITH LINUX ? Thanks a lot in advance Sincerely, Tapan From owner-chemistry@ccl.net Thu Aug 31 12:36:00 2006 From: "Dimitrios A Pantazis pantazis]![hotmail.com" To: CCL Subject: CCL: Bad calculated geometries in CCCBDB Message-Id: <-32462-060831122624-14296-WuV5P6Gmx+8RMLTflM8nKQ~!~server.ccl.net> X-Original-From: "Dimitrios A Pantazis" Date: Thu, 31 Aug 2006 12:26:23 -0400 Sent to CCL by: "Dimitrios A Pantazis" [pantazis[#]hotmail.com] The "Bad Calculations" section of CCCBDB is, unfortunately, still full of entries that are simply nonsense. For example, it may come as a surprise to many that a B3LYP calculation with the 6-311+G(3df,2p) basis set on pentadiene predicts the CC double bond length to be 3.551 Angstroms and the C-H bonds more than 5.5 Angstroms long! Or that a B3PW91/cc-pVTZ calculation on cyclopentene results in a CC bond length of over 2.35 A! Of course, despite their appearance in the database, none of that is true. The majority of entries like the above (in fact, the majority of entries in this part of the database) are the result of messy atom numbering: for example, the CC "bond length" of 2.35 A for cyclopentene in the calculation mentioned above is in fact the distance between two non-adjacent carbons. Clearly, the entries must have been created by some automated method and never checked for their validity before being made public. One may double-check each entry by looking up the molecule of interest in the "Calculated Data" section, where the distance matrix is given, albeit with the wrong numbering -- or repeat the calculation. Although it is relatively easy to spot errors such as these, the disagreement pointed out by Andreas between reported moments of inertia and experimental geometries raises another question about the reliability of the database. Browsing through the entries, I noticed that some of the past spurious entries have already been removed from the "Bad Calculations" part of CCCBDB, especially for lower levels of theory. Too many still exist, however, making this part of the database rather unreliable as a reference. Dimitrios ---- Dr. Dimitrios A. Pantazis Computational Chemistry Group WestCHEM, Department of Chemistry Joseph Black Building University of Glasgow Glasgow G12 8QQ United Kingdom Telephone: +44 (0)141 330 8121 E-mail: d.pantazis() chem.gla.ac.uk ---- From owner-chemistry@ccl.net Thu Aug 31 13:45:01 2006 From: "D.BIO- Gerard Pujadas gerard.pujadas a urv.cat" To: CCL Subject: CCL: Announce: BDT, an easy-to-use front-end application for AutoDockautomation Message-Id: <-32463-060831134233-14565-sliByFFiM1SWE1mWnnnlmw-#-server.ccl.net> X-Original-From: "D.BIO- Gerard Pujadas" Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=iso-8859-1; format=flowed Date: Thu, 31 Aug 2006 19:42:03 +0200 Mime-Version: 1.0 Sent to CCL by: "D.BIO- Gerard Pujadas" [gerard.pujadas,urv.cat] Dear CCL list subscribers, we would like to announce BDT. MOTIVATION: AutoGrid/AutoDock is one of the most popular software packages for docking, but its automation is not trivial for tasks such as (1) the virtual screening of a library of ligands against a set of possible receptors; (2) the use of receptor flexibility and (3) making a blind-docking experiment with the whole receptor surface. This is an obstacle for research teams in the fields of Chemistry and the Life Sciences who are interested in conducting this kind of experiment but do not have enough programming skills. To overcome these limitations, we have designed BDT, an easy-to-use graphic interface for AutoGrid/AutoDock. AVAILABILITY: BDT is available for free, upon request, for non-commercial research. Look at http://www.quimica.urv.cat/~pujadas/BDT/Download.html for donwload instructions. SUPPLEMENTARY INFORMATION: Learn more about BDT at Bioinformatics. 2006 Jul 15;22(14):1803-4 and in the BDT website http://www.quimica.urv.cat/~pujadas/BDT ACKNOWLEDGMENTS: We thank the authors of AutoGrid/Autodock for providing us with version 3.0.5 of their software and, especially, Dr Garrett Morris and Dr Ruth Huey for their help. This work is supported by grant number CO3/O8 > from the Fondo de Investigación Sanitaria (FIS) and AGL2005-04889 from the Comisión Interministerial de Ciencia y Tecnología (CICYT) of the Spanish Government. Yours sincerely, Gerard -- ============================= NOTE: NEW E-MAIL ============================ Dr. Gerard Pujadas Grup de recerca en Nutrigenòmica Dept. Bioquímica i Biotecnologia room 106 Campus de Sant Pere Sescelades e-mail: gerard.pujadas~!~urv.cat Univ. Rovira i Virgili phone: 34-977 559565 C/ Marcel·lí Domingo fax: 34-977 558232 43007 Tarragona (CATALONIA) State: Spain (European Union) ===========================================================================