From ZPZCM@jazz.ucc.uno.edu Sat Aug 10 16:53:04 1993 Date: 10 Aug 1993 22:53:04 -0600 (CST) From: "ZORAN P. ZDRAVKOVSKI" Subject: QSAR software and references To: chemistry@ccl.net Message-Id: <01H1LIB3DBOY8WWN3L@jazz.ucc.uno.edu> Hi Folks, We would like to do some QSAR studies on anti-cancer drugs so I would appreciate some advice on: - Software packages especially ones for the PC (or VAX). We already have Chem-X but not the QSAR modules (Chem DBS-1 and ChemStat) but before getting them, since they are at least $1000, we'd like to know how they compare to other packages. Any in the public domain? - Introductory references (books, reviews, software reviews). Sincerely, Zoran ******************************************************************** * Zoran Zdravkovski * * Dept. of Chemistry (504) 286-7090 * * University of New Orleans (504) 286-6062 (fax) * * New Orleans, LA 70148 zpzcm@uno.edu * ******************************************************************** From B_DUKE@DARWIN.NTU.EDU.AU Wed Aug 11 14:37:31 1993 Date: Wed, 11 Aug 1993 14:37:31 GMT From: B_DUKE@DARWIN.NTU.EDU.AU (Brian Duke) Message-Id: <930811143731.e46d@DARWIN.NTU.EDU.AU> Subject: Gaussian92, subg92 and NQS To: CHEMISTRY@ccl.net G'day, comp chem netters. I'm having problems on a IBM RISC 6000 M2A (same as M20) running G92 with the NQS queuing system, version 3.34 from wuarchive. subg92 is supposed to run NQS but there are clearly different versions of NQS, and the subg92 as provided seems to want a different version from the one I have. A simplified script is below. I have removed the option in subg92 for more than 2 parameters and incorporated the 1st - the queue - in the script. Problems of files ending up in my home directory b_duke rather than b_duke/work. lead me to stick various cd /home/b_duke/work lines in, but they may not all be required. #!/bin/csh # # Submit a Gaussian 92 job to an nqs batch queue. # # $1 -- name of input file. # cd /home/b_duke/work # cat <$1.job # QSUB -lc 4000 # QSUB -ld "32 mb" # QSUB -lf "408 mb" # QSUB -lm unlimited # QSUB -ls "32 mb" # QSUB -lt 360000 # QSUB -lw "32 mb" # QSUB -q molecule # QSUB cd /home/b_duke/work g92 <$1.com END # echo Now submit it. # qsub -s /bin/csh -nr -eo -r $1 -x -o $1.batch-log $1.job Actually stripping out all the QSUB lines also gives the same result, which is, for the file $1.batch-log:- Warning: no access to tty; thus no job control in this shell... Killed logout I can find nothing useful in the nqs-logfile. I put this problem to Doug Fox at Gaussian Inc. He suggested either a problem of setting the C shell (hence the -s /bin/csh) or a memory problem. For the latter, he suggested taking -bmaxdata:0x40000000 from the make file and relinking. This did not work either. Does anyone have any experience with this version of NQS and G92? One feature I thought odd was that the qlimit command does not show -lm or -lM for memory. By the way, I have similar scripts to the above for running gamess, mopac6 and argus. They all work. Also G92 works fine with "nohup g92 test &", with a "printer" queue with backend=/bin/csh, and with batch command. It is just with NQS thta it fails. I would welcome any experience anyone has on this problem. Thanks. Regards from "down under", Brian. Associate Professor Brian Salter-Duke (Brian Duke) School of Chemistry and Earth Sciences, Northern Territory University GPO Box 40146, Casuarina, NT 0811, Australia. Phone 089-466702 FAX 089-410460 E-mail B_DUKE@DARWIN.NTU.EDU.AU From schaedler@informatik.uni-halle.d400.de Wed Aug 11 10:25:28 1993 Date: Wed, 11 Aug 1993 08:25:28 +0200 From: schaedler@informatik.uni-halle.d400.de Message-Id: <930811082528*/S=schaedler/OU=informatik/PRMD=UNI-HALLE/ADMD=D400/C=DE/@MHS> To: chemistry@ccl.net Subject: Mol. Design Expert System & tools? Dear netters, we are working on an Expert System for Liquid Crystals Molecular Design. The problem differs significantly from Drug Design. We are searching for a) tools for low energy conformation searching, molecular geometry, charge density / electrostatic potential surfaces etc., which can be included in our system (platform: sun sparc, object-oriented LISP CLOS, but C sources are suitable, too). the system will provide advanced AI-methods, which shall be combined with safe, robust tools from comp. chemistry, maintaining the possibility to change them in the future. b) contacts with people who are working on the same or a similar topic, especially Artificial Intelligence applications in Organic and Phys. Chem. or Drug Design (NOT concentrating on local receptor-site interaction) Help is urgently wanted, because we lost a lot of time by looking for suitable software in commercial offers etc., but as it has been discussed here before - no real TOOLs were available, only packages are offered. Thanks Kristina Schaedler (please remind in your answer, i'm not a chemist and i have only little background in theoretical chemistry!) From xlh@wag.caltech.edu Tue Aug 10 16:52:51 1993 Date: Tue, 10 Aug 93 23:52:51 -0700 From: Xinlei Hua Message-Id: <9308110652.AA08318@sgi1> To: chemistry@ccl.net Subject: chemical eq. software inquiry Dear Netters, Does anyone know if there are any expert system type of softwares that can write chemical equations? For example users enter the reactants and condition, the software will print out the products and the right chemical equations. Maybe I stretched my dream too much, but if you know something that do jobs in the similar direction I would appreciate very much your response. And I will be happy to do a sum up. Thanks. Xinlei Hua 139-74 Beckman Institute Caltech, Pasadena,CA 91125 (818)577-5108 From JKONG@ac.dal.ca Wed Aug 11 09:44:13 1993 Date: Wed, 11 Aug 1993 12:44:13 -0300 From: JKONG@ac.dal.ca Subject: nice? To: chemistry@ccl.net Message-Id: <01H1MBB7DJQM00DJ7A@AC.DAL.CA> Dear netters, This query has been posted twice on newsgroup comp.unix.aix but no response yet. The commands "nice" and "renice" seemingly do not work on our two ibm risc/6000's(580 and 355). The commands we tried are renice 20 pid /usr/bin/nice -n 15 job nice +15 job (csh) The speed of the jobs simply were not affected in the existance of other jobs. And also we don't know how to schdedule the queues. The queues we use now are printer queues. We want a couple of queues with different priorities to meet the demands of different types of jobs. We would very appreciate any suggestions. Jing Kong Dept. Chem. Dalhousie U. From h.rzepa@ic.ac.uk Wed Aug 11 16:41:33 1993 Message-Id: <9308111544.AA03856@cscmgb.cc.ic.ac.uk> Date: Wed, 11 Aug 1993 16:41:33 +0000 To: CHEMISTRY@ccl.net From: h.rzepa@ic.ac.uk (Henry Rzepa) (Henry Rzepa) Subject: Re: chemical eq. software inquiry >Dear Netters, > >Does anyone know if there are any expert system type of softwares that can >write chemical equations? >For example users enter the reactants and condition, the software >will print out the products and the right chemical equations. >Maybe I stretched my dream too much, but if you know something that >do jobs in the similar direction I would appreciate very much your response. >And I will be happy to do a sum up. >Thanks. I have tried a Hypercard stack that seems to do this. It is called MacElements Utilities 1.01 and is on the sumex macintosh archive (ie only for Mac). Dr Henry Rzepa, Dept. Chemistry, Imperial College, LONDON SW7 2AY; rzepa@ic.ac.uk via Eudora 1.3.1, Tel:+44 71 225 8339, Fax:+44 71 589 3869. From albo@RUUCI9.chem.ruu.nl Wed Aug 11 09:49:10 1993 From: albo@RUUCI9.chem.ruu.nl (Alexandre Bonvin) Message-Id: <9308120049.AA00956@RUUCI9> Subject: NMR chemical shift refinement To: CHEMISTRY@ccl.net Date: Wed, 11 Aug 1993 17:49:10 -0800 (PDT) I am actually writting up my PhD thesis on the determination of biomolecular structures by NMR. I would like to include something about structure refinement based on chemical shift information. Does someone know some recent references on this subject? Thanks Alexandre Bonvin NMR spectroscopy Utrecht University The Netherlands Email: albo@nmr.chem.ruu.nl From d3g359@rahman.pnl.gov Wed Aug 11 08:28:15 1993 Date: Wed, 11 Aug 93 15:28:15 -0700 From: d3g359@rahman.pnl.gov Subject: HF To: CHEMISTRY@ccl.net Message-Id: <9308112228.AA07753@rahman.pnl.gov> Hi netters. I would be interested in hearing from people who have done calculations on HF, H2F-, and various HF dimers, trimers, etc. Thanks, John ------------------------------------------------------------------------------ John Nicholas Senior Research Scientist Molecular Science Research Center Battelle Pacific Northwest Laboratory Mailstop K1-90 Richland, WA 99352 (509) 375-6559 ------------------------------------------------------------------------------ From mercie@cumc.cornell.edu Wed Aug 11 18:56:56 1993 Date: Wed, 11 Aug 1993 22:56:56 -0400 (EDT) From: Gustavo Mercier Subject: chem-station in sgi To: chemistry@ccl.net Message-Id: Hi, Netters! Has anybody ported Chem - Station to an INDIGO R4000 Elan? We are in the process of acquiring the package and comments on this process (i.e. problems, solutions, pitfalls, etc.) would be appreciated! I would like to avoid rediscovering the wheel! thanks gus mercier,jr. mercie@cumc.cornell.edu