From chemistry-request@server.ccl.net Wed Oct 27 04:08:14 1999 Received: from edda.vcp.monash.edu.au (edda.vcp.monash.edu.au [130.194.216.165]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA13642 for ; Wed, 27 Oct 1999 04:08:11 -0400 Received: from localhost (david@localhost) by edda.vcp.monash.edu.au (980427.SGI.8.8.8/980728.SGI.AUTOCF) via ESMTP id TAA49658 for ; Wed, 27 Oct 1999 19:01:01 +1100 (EDT) Date: Wed, 27 Oct 1999 19:01:01 +1100 From: David Chalmers X-Sender: david@edda.vcp.monash.edu.au To: CHEMISTRY@ccl.net Subject: Graphing data with CACTVS Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi All, I have just found the CACTVS suite of programs and it seems to be a very nice package for dealing with SAR data. In particular it contains an XY graphing routine (csxy) which can associate a 2D chemical structure with each point in the graph. At the moment I am having trouble importing MDL 'sdf' files which can contain biological data as well as the molecule coordinates. When I import my SDF file the biological data (a single floating point number) is imported in to CACTVS with property names of the type E_*ACTIVITY*. I have used the property editor to create a new property with thisB names and have defined this property as being a FLOAT. When I try to graph the data I get the following error: Property E_*ACTIVITY* has no suitable data type. Can anybody help me here and point out what I might be doing wrong? I can plot using the inbuilt property types with no problem. David _____________________________________________________________________________ David Chalmers Lab: 9903 9110 Victorian College of Pharmacy Fax: 9903 9582 381 Royal Pde, Parkville, Vic 3053 http://synapse.vcp.monash.edu.au Australia David.Chalmers@vcp.monash.edu.au _____________________________________________________________________________ From chemistry-request@server.ccl.net Wed Oct 27 08:06:17 1999 Received: from terra.ifi.unicamp.br (terra.ifi.unicamp.br [143.106.6.20]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id IAA15335 for ; Wed, 27 Oct 1999 08:06:16 -0400 Received: from ifi.unicamp.br (lua [143.106.6.13]) by terra.ifi.unicamp.br (8.9.3/8.9.3) with ESMTP id JAA09674; Wed, 27 Oct 1999 09:54:20 -0200 (BDB) Received: (from daemon@localhost) by ifi.unicamp.br (8.9.3/8.9.3) id JAA24799; Wed, 27 Oct 1999 09:54:13 -0200 (BDB) Received: from axion.ifi.unicamp.br(143.106.39.170) via SMTP by lua.ifi.unicamp.br, id smtpdAAAa0063N; Wed Oct 27 09:54:07 1999 Received: from localhost (dasf@localhost) by axion.ifi.unicamp.br (8.9.3/8.9.3) with SMTP id JAA37671; Wed, 27 Oct 1999 09:54:10 -0200 (EDT) X-Authentication-Warning: axion.ifi.unicamp.br: dasf owned process doing -bs Date: Wed, 27 Oct 1999 09:54:10 -0200 (EDT) From: Demetrio Antonio da Silva Filho - doutorado To: Demetrio Antonio da Silva Filho - doutorado cc: CHEMISTRY@ccl.net Subject: Geometry Optimization using CIS and AM1-CI In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, I'm studing the geometry of the first excited state using AM1-CI (implemented im MOPAC 6) and CIS (implemented in GAUSSIAN 94). I increased the size of the CI in AM1-CI and the final geometry changed, as I expected. But the optimized geometries using CIS method for diferent number of excitations used in the calculation were the same. It's a particularity associated to the sistem I'm working or it's a particularity of the method? Does anybody have references of geometry studies of the excited state changing the number of excitations and using the CIS method? Thanks in advance, Demetrio FIlho _______________________________________________________________ Demetrio A. da Silva Filho UNICAMP - IFGW Predio D - Sala 17 CEP 13083-970 C.Postal 6165 Campinas - SP - Brazil _______________________________________________________________ "Se nao houver frutos, valeu a beleza das flores. Se nao houver flores, valeu a sombra das folhas. Se nao houver folhas, valeu a intencao da semente." Henfil From chemistry-request@server.ccl.net Wed Oct 27 08:43:46 1999 Received: from icp.ac.ru (icp.ac.ru [193.233.43.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id IAA15488 for ; Wed, 27 Oct 1999 08:43:35 -0400 Received: from phot2.pc.icp.ac.ru (phot2.pc.icp.ac.ru [192.168.1.93]) by icp.ac.ru (8.9.1/8.9.1) with SMTP id QAA21378 for ; Wed, 27 Oct 1999 16:34:08 +0400 (MSD) Message-ID: <38178BE9.42B9@icp.ac.ru> Date: Wed, 27 Oct 1999 16:34:01 -0700 From: "Mikhayl F.Budyka" Reply-To: budyka@icp.ac.ru Organization: Institute of Problems of Chemical Physics X-Mailer: Mozilla 3.0Gold (Win95; I; 16bit) MIME-Version: 1.0 To: chemistry@ccl.net Subject: Fe parameters for MOPAC (PM3) Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit Dear CCL'ers, I am looking for Fe parameters for MOPAC (PM3). Thanks in advance. M.Budyka From chemistry-request@server.ccl.net Wed Oct 27 09:45:06 1999 Received: from volvic.chem.nagoya-u.ac.jp (volvic.chem.nagoya-u.ac.jp [133.6.85.130]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id JAA15827 for ; Wed, 27 Oct 1999 09:45:05 -0400 Received: from localhost (napoleon.chem.nagoya-u.ac.jp [133.6.85.143]) by volvic.chem.nagoya-u.ac.jp (8.9.2/3.7W.bar.20Nov98) with ESMTP id WAA20109 for ; Wed, 27 Oct 1999 22:35:44 +0900 (JST) To: chemistry@ccl.net Subject: charm parameter in amber X-Mailer: Mew version 1.93 on Emacs 19.34 / Mule 2.3 (SUETSUMUHANA) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: <19991027223749H.yosh@aqua.chem.nagoya-u.ac.jp> Date: Wed, 27 Oct 1999 22:37:49 +0900 (JST) From: Fujitsuka Yoshimi X-Dispatcher: imput version 980905(IM100) Lines: 15 Dear CCL, I'd like to use charm22 parameters in amber. Is it possible? If it's possible, are there any programs which convert charm22 parameters to amber format files like 'parm94.dat'? Thanks YOSHIMI FUJITSUKA Theoretical Chemistry Laboratory Department of Chemistry,Faculity of Science Nagoya Univercity Japan (e-mail) yosh@aqua.chem.nagoya-u.ac.jp From chemistry-request@server.ccl.net Wed Oct 27 12:32:17 1999 Received: from relay2.scripps.edu (relay2.scripps.edu [137.131.200.30]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id MAA16571 for ; Wed, 27 Oct 1999 12:32:17 -0400 Received: from gamow.scripps.edu (gamow.scripps.edu [137.131.252.67]) by relay2.scripps.edu (8.9.3/TSRI-3.1.0r) with ESMTP id JAA00286; Wed, 27 Oct 1999 09:21:04 -0700 (PDT) Received: (from case@localhost) by gamow.scripps.edu (8.9.1a/TSRI-2.8) id JAA26065; Wed, 27 Oct 1999 09:21:05 -0700 (PDT) Message-ID: <19991027092105.B25710@gamow.scripps.edu> Date: Wed, 27 Oct 1999 09:21:05 -0700 From: David Case To: Fujitsuka Yoshimi Cc: chemistry@ccl.net Subject: Re: CCL:charm parameter in amber References: <19991027223749H.yosh@aqua.chem.nagoya-u.ac.jp> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.93.2 In-Reply-To: <19991027223749H.yosh@aqua.chem.nagoya-u.ac.jp>; from Fujitsuka Yoshimi on Wed, Oct 27, 1999 at 10:37:49PM +0900 On Wed, Oct 27, 1999, Fujitsuka Yoshimi wrote: > > I'd like to use charm22 parameters in amber. Is it possible? > If it's possible, are there any programs which convert charm22 > parameters to amber format files like 'parm94.dat'? > We've got close to this, but never had the resources to really finish it off, basically by having someone familiar enough with both CHARMM and Amber to decide if things are really working. If you want to give things a try, you can look at the material in ftp://ftp.scripps.edu/pub/case/c22_in_amber.tar.gz Equivalent (slightly updated) material will be a part of Amber 6, due out soon. But be warned: you will need to recompile Amber with special flags, run some scripts to convert CHARMM files to Amber format (and then hand-check the outputs). You will also need access to CHARMM to decide for yourself on the level of compatability you require. Many CHARMM options for handling long-range cutoffs are not implemented in Amber, although using PME for long-range electrostatics should be nearly the same in the two codes. We would love to have someone volunteer to complete this project! (And, of course, would give that person proper attribution on the resulting files.) ....dac -- ================================================================== David A. Case | e-mail: case@scripps.edu Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896 The Scripps Research Institute | phone: +1-858-784-9768 10550 N. Torrey Pines Rd. | WWW: La Jolla CA 92037 USA | http://www.scripps.edu/case ================================================================== From chemistry-request@server.ccl.net Wed Oct 27 14:19:59 1999 Received: from wugate.wustl.edu (wugate.wustl.edu [128.252.120.1]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA17196 for ; Wed, 27 Oct 1999 14:19:59 -0400 Received: from mirlink.wustl.edu (mirlink.wustl.edu [128.252.155.1]) by wugate.wustl.edu (8.8.8/8.8.5) with SMTP id NAA08458 for ; Wed, 27 Oct 1999 13:10:47 -0500 (CDT) Message-ID: Date: 27 Oct 1999 13:09:18 -0500 From: "David Reichert" Subject: DFT & freq. summary To: "CCL post" X-Mailer: Mail*Link SMTP-MS 3.0.2 Mime-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; Name="Message Body" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id OAA17197 Hi, I posted the following message awhile ago, and several people asked for a summary of the responses. I apologize for the delay in posting this. The orginal post was: Hi, I'm hoping that someone can provide some enlightenment. I've been running some DFT calculations on some copper - amine complexes which appear to end up having only C1 symmetry. In order to ensure that these are indeed minimum energy structures, I've run frequency calculations on them as well. My question is what should I really be seeing in the frequency calculations? I know that for a transition structure you look for one imaginary frequency, and that ideally the translational and rotational modes should drop out. But in real life with non-perfect calculations what do you want to see that says, "Yes, this is indeed a low energy fully converged struture". Any rules of thumb are more than welcome. Following are the responses that I recieved. I'd like to thank everyone who responded for the additional info and helpful suggestions. -Dave Reichert ------------------------------------------------------------------------------------------------------------ From: Brian Salter-Duke There are some real issues here. What code are you running? I'll comment about G94. I have G98 on order. Ideally as you say you should have 6 frequencies that are close to zero for the rotational and translational degrees of freedom. If you have a local minimum all the others should be positive. If you have a transition structure you should have one imaginary frequency which will appear as a negative frequency. With RHF, MP2 etc. you get this. Particularly with opt=tight, you should get the "zero" frequencies in +1 <-> -1. Many people are happy if they are in +10 <-> -10 but I prefer to get them smaller. With DFT in G94 you are using a numerical grid that means the results are not rotationally invarient - the results differ slightly if you rotate the molecule in the cartesian frame. This means that the "zero" frequencies are never really zero. The situation is worst for heavy atoms so with Cu you will have more problems than with a straight 1st row organic molecule. Using a larger grid improves the situation but increases the cost. My rule of thumb for Ga calculations is accept the results if the zero frequencies are in +50 <-> -50 provided the smallest real frequency or for a TS the negative frequency is well outside that range - e.g. more than 200 cm-1. If it is not well outside that range use a larger grid and look at the changes in the frequencies etc. Always use opt=tight. It is a good idea to run two grids - the default and one slightly larger and compare the results. If the real frequencies and the TS negative frequency differ by less than 1 cm-1 you are probably OK. The rotational invariance problem is solved in the qchem code. I do not know about G98. From: Robert Zellmer Of course your frequencies should all be positive if you have a minimum. Whether it's a global minimum is another story. Depending on the number of degrees of freedom, this could be an almost impossible thing to determine. However, you can try. One way is to use MM or semi-emp. methods to produce many low energy conformations and then choice the most likely ones (lowest energy structures) and run single point DFT calculations to see which one has the lowest energy. This is something I'm doing now on several compounds. One reason I've had to do this is in running some AM1 calculations I got different optimized structures from Gaussian98 and Spartan. It can be a bit tedious and I know people are working on the problem with global vs. local minima but I don't recall that anyone has "solved" it. From: Per-Ola Norrby Are you using Gaussian for this? Because in that case, you always get the output corresponding to a geometry optimization step, with four criteria that have to be fullfilled. If all of them say "YES", you're OK... Gaussian will also list the low frequencies twice. The first list is unprojected, in the second list the translations and rotations have been projected out (that means that you won't see the 6 values closest to zero). If you are at a stationary point, these should match closely (except for the extra 6 in the first list), say within a few wavenumbers. For a minimum, all remaining vibrations should be positive. If some are close to zero, you may have free rotations in your molecule. If any of them are below, say, 30-50 cm-1, you should treat thermodynamic contributions with caution, because the harmonic approximation is no longer valid. From: sergiusz kwasniewski Hi, for a fully optimized structure, you shouldn't have any imaginary (or negative in case of gaussian freqs) frequencies left (or very small ones, i.e. just a couple cm-1). The translational and rotational modes should be dropping out here as well. A transition state (first order saddle point) then has two imaginary freqs, a second order saddle point has three ... I hope this is what you were looking for ... From: Huub van Dam As you well know for a minimum energy geometry you should find all positive frequencies. Furthermore my guess would be that all the frequencies should come the same as with an exact calculation (i.e. rotational and translational modes drop, modes across the minimum of a well should have positive frequencies and modes across a maximum imaginary frequencies). But in the case of DFT you would find deviations due to the errors introduced by the numerical integration. I am not sure about the accuracy one can achieve, but I think the errors should not be larger than a few dozen wavenumbers. If it is more than either the structure you found isn't what you where looking for either the integration was to inaccurate and you should retry the calculation at a higher precision. From: Larry Cuffe I dont think there are any hard and fast rules. For a transition structure I prefer to see one negative frequency preferrable greater than 100cm/-1. If there are very low frequency negatives around say -20 or -30 I dont worry about them. If there is simething in between I try and visualize the movement involved in the frequency and see what moving where. Generaly you can see if this is the negative frequency of interest coresponding to the sought for TS like this. If you are using gaussian then gaussview is usefull for visualizing negative modes. also in the achive entry it gives the number of negative modes it feels are significant this is the entry NIMAG= From: Wilson, Bruce E Ideally, what you want is to not see any imaginary frequencies in the frequency calc. However, we do DFT on organometallics, and no imaginary frequencies doesn't always happen. In general, I don't worry about imaginary frequencies below about 100 cm-1, at least not in terms of deciding that something is a local minimum instead of a saddle point of some order. A good minimizer, however, helps this immensely. We've almost never had problems with Jaguar (www.schrodinger.com). If it minimizes, it seems to minimize to a true minimum, particularly if symmetry is turned off. It has also broken symmetry on several structures when it determined that it needed to do so in order to get to a true minimum. That's another rule of thumb I've followed. If the output structure is C1 and "looks reasonable", I don't worry about the frequency calc to verify the minimum. The only times I run frequencies to verify a minimum is when the final structure has some type of symmetry. From: Michael Braunschweig the nonexistance of imaginary frequencies is only one criterion for a minimum. looking at the first derivatives of the energy you can see the forces that are used as convergence criteria from some programms (for example gaussian). if they are zero the geometry stay at an (global oder local) minumum. also the last changes in energy or in the displacements of the coordinates at the last optimization step are often use as convergence criteria. but all these indicate both global and local minima at the energy hyper surface. a minimum you have found in this way may also be a local one and not a fully converged structure. there are different methods for searching for a global minimum. try monte carlo or simulated annealing to examine different conformation. ____________________________________________ | David Reichert, Ph.D | | Mallinckrodt Institute of Radiology | | Washington University School of Medicine | | St. Louis, MO | | | | e-mail: reichertd@mirlink.wustl.edu | | voice: (314) 362-8461 | | fax: (314) 362-9940 | ____________________________________________ From chemistry-request@server.ccl.net Wed Oct 27 14:27:29 1999 Received: from mailhost.det.ameritech.net (mpdr0.detroit.mi.ameritech.net [206.141.239.206]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id OAA17261 for ; Wed, 27 Oct 1999 14:27:28 -0400 Received: from piilibrary.ameritech.net ([209.18.27.185]) by mailhost.det.ameritech.net (InterMail v4.01.01.07 201-229-111-110) with SMTP id <19991027181617.JJRG5474.mailhost.det.ameritech.net@piilibrary.ameritech.net>; Wed, 27 Oct 1999 14:16:17 -0400 Message-ID: <006e01bf20a7$5caf8fe0$cfd18dce@ameritech.net> Reply-To: "Jim Kress" From: "Jim Kress" To: , "Fujitsuka Yoshimi" References: <19991027223749H.yosh@aqua.chem.nagoya-u.ac.jp> Subject: Re: CCL:charm parameter in amber Date: Wed, 27 Oct 1999 14:16:16 -0400 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 You can do this with Tinker. http://dasher.wustl.edu/tinker/ Jim Check out my web site http://www.kressworks.com/ It'll blow your mind (politically), stimulate your senses (artistically) and provide scientific insights that boggle the mind! ----- Original Message ----- From: Fujitsuka Yoshimi To: Sent: Wednesday, October 27, 1999 9:37 AM Subject: CCL:charm parameter in amber > Dear CCL, > > I'd like to use charm22 parameters in amber. Is it possible? > If it's possible, are there any programs which convert charm22 > parameters to amber format files like 'parm94.dat'? > > Thanks > > > YOSHIMI FUJITSUKA > Theoretical Chemistry Laboratory > Department of Chemistry,Faculity of Science > Nagoya Univercity > Japan > (e-mail) yosh@aqua.chem.nagoya-u.ac.jp > > -= 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 > > > > From chemistry-request@server.ccl.net Wed Oct 27 17:20:52 1999 Received: from rzusuntk.unizh.ch (rzumail2.unizh.ch [130.60.128.10]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id RAA18030 for ; Wed, 27 Oct 1999 17:20:51 -0400 Received: [from rzusgi.unizh.ch (chburger@rzusgif.unizh.ch [130.60.68.19]) by rzusuntk.unizh.ch (8.8.5/SMI-5.31) with SMTP id XAA23008 for ; Wed, 27 Oct 1999 23:11:36 +0200 (MET DST)] Date: Wed, 27 Oct 1999 23:11:36 +0200 (MDT) From: "Dr. Peter Burger" To: CHEMISTRY@ccl.net Subject: DFT-NMR & ECP's Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCL'ers is there a final word on the reliabilty/quality of the calculation of 13-C-NMR spectra by DFT methods if the carbon atoms are bound to heavy elements (particularly transtion metals) for which ECP's including rel. corrections have been applied? Is this approach valid at all? Are there any calibration calculations, i.e. with all electron basis sets vs ECP, references? Best regards Peter -------------------------------- Peter Burger Anorg.-chem. Institut University of Zuerich chburger@aci.unizh.ch