From chemistry-request@server.ccl.net Fri May 3 04:10:50 2002 Received: from linux5.theochem.uni-stuttgart.de ([129.69.55.55]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g438AiI01255 for ; Fri, 3 May 2002 04:10:48 -0400 Received: from localhost (teomgs@localhost) by linux5.theochem.uni-stuttgart.de (8.11.6/8.11.6/SuSE Linux 0.5) with ESMTP id g438AJ430580 for ; Fri, 3 May 2002 10:10:19 +0200 X-Authentication-Warning: linux5.theochem.uni-stuttgart.de: teomgs owned process doing -bs Date: Fri, 3 May 2002 10:10:18 +0200 (METDST) From: "Dr. Martin Schuetz" To: Subject: energy level diagrams, plotprogram ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, Is anybody out there on that list aware of a reasonable procedure (or ideally a computer program) to generate plots of energy level diagrams, (e.g. for two-atomic system) from a given set of computet orbital energies ? I will summarize the response I get. Thanks Martin Schuetz ---------------------------------------------------------------------------- PD Dr. Martin Schuetz Phone: 0049 (0)711-685-4409 University of Stuttgart Fax: 0049 (0)711-685-4442 Institute for Theoretical Chemistry Pfaffenwaldring 55 Room Nr: 8.161 D-70569 Stuttgart Email: schuetz@theochem.uni-stuttgart.de WWW : http://www.theochem.uni-stuttgart.de/~schuetz/ ---------------------------------------------------------------------------- From chemistry-request@server.ccl.net Fri May 3 04:30:22 2002 Received: from vytautas.vdu.lt ([193.219.38.33]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g438UMI02017 for ; Fri, 3 May 2002 04:30:22 -0400 Received: from vaidila.vdu.lt ([193.219.38.52] ident=mail) by vytautas.vdu.lt with esmtp id 173YSL-0003h3-00; Fri, 03 May 2002 10:30:21 +0200 Received: from vejas.vdu.lt ([193.219.38.82] helo=TYREX) by vaidila.vdu.lt with esmtp id 173YU2-0006Um-00; Fri, 03 May 2002 10:32:06 +0200 Date: Fri, 3 May 2002 11:30:22 +0300 From: a3arzi X-Mailer: The Bat! (v1.44) UNREG / CD5BF9353B3B7091 Reply-To: "Art'" X-Priority: 3 (Normal) Message-ID: <184908712.20020503113022@vaidila.vdu.lt> To: Joe Leonard CC: chemistry@ccl.net Subject: Re[2]: CCL:scoring functions In-reply-To: <5.1.0.14.1.20020502181500.00ac2510@pop.theworld.com> References: <5.1.0.14.1.20020502181500.00ac2510@pop.theworld.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, Yes - I know CScore. But the main disadvantage is it runs ONLY on HP and IBM computers. I prefer Linux/MS Windows OS'es. If anybody could suggest anything similar - please, let me know. Something CScore-like would be the real I need. ----------------------------> >>Question about where get some free/commercial good scoring function to >>evaluate docking results. Any suggestions are welcome. JL> Tripos has a product called CScore, which applies several scoring functions JL> to a docked protein/ligand pose... JL> Joe <--------------------------- Good weekend Arturas http://biologija.vdu.lt/person/ziemar/index.html From chemistry-request@server.ccl.net Fri May 3 03:08:09 2002 Received: from gauss.uniba.it (IDENT:root@[193.204.176.40]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43788I30333 for ; Fri, 3 May 2002 03:08:09 -0400 Received: from mozart.uniba.it (IDENT:root@mozart.uniba.it [193.204.184.2]) by gauss.uniba.it (8.9.3/8.9.3) with ESMTP id JAA01704 for ; Fri, 3 May 2002 09:01:53 +0200 Received: from farmo2 (farmo2.farmacia.uniba.it [193.204.179.198]) by mozart.uniba.it (8.11.0/8.11.0) with SMTP id g437GbY30554 for ; Fri, 3 May 2002 09:16:37 +0200 Message-ID: <001801c1f271$42f0ad20$c6b3ccc1@farmo2> From: "Angelo Favia" To: Subject: analysis of qsar data set. Date: Fri, 3 May 2002 09:08:00 +0200 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0015_01C1F282.06666A50" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 This is a multi-part message in MIME format. ------=_NextPart_000_0015_01C1F282.06666A50 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear netters, I am a student in the last year of the pharamaceutical chemistry at=20 the Univ. of Bari (Italy). As a project for my degree, I need to=20 analise a qsar data set (~130 compounds) in which pKi data are=20 unevenly distributed. Specifically, pKi data, which span from 7 to=20 12, are binned as follows 7-8: ~5% 8-9: ~16% 9-10: ~57% 10-11: ~26% 11-12: ~2% Given this distribution, my point is that there can be some risk to=20 find out unreliable q2 values during the leave one out calculation=20 and also that some validation analyses as the scrambling of data=20 can't help in a situation like that. I wonder if I can still be happy performing qsar on the whole data=20 set ( which is what I whish, possibly ) or if it would be better select=20 a more uniform subset of points. Have you any clue about that? thanks, Angelo Favia ------=_NextPart_000_0015_01C1F282.06666A50 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Dear netters,
I am = a student in=20 the last year of the pharamaceutical chemistry at
the Univ. of Bari = (Italy).=20 As a project for my degree, I need to
analise a qsar data set (~130=20 compounds) in which pKi data are
unevenly distributed. Specifically, = pKi=20 data, which span from 7 to
12,  are binned as follows
7-8:=20 ~5%
8-9: ~16%
9-10: ~57%
10-11: ~26%
11-12: ~2%
Given = this=20 distribution, my point is that there can be some risk to
find out = unreliable=20 q2 values during the leave one out calculation
and also that some = validation=20 analyses as the scrambling of data
can't help in a situation like = that.
I=20 wonder if I can still be happy performing qsar on the whole data
set = ( which=20 is what I whish, possibly ) or if it would be better select
a more = uniform=20 subset of points.

Have you any clue about = that?

thanks,
Angelo=20 Favia
------=_NextPart_000_0015_01C1F282.06666A50-- From chemistry-request@server.ccl.net Fri May 3 05:43:24 2002 Received: from umhsp02.umh.ac.be ([193.190.192.27]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g439hNI04672 for ; Fri, 3 May 2002 05:43:23 -0400 Received: from europe.umh.ac.be (europe.umh.ac.be [193.190.132.12]) by umhsp02.umh.ac.be (8.9.3/8.9.3) with ESMTP id LAA04660 for ; Fri, 3 May 2002 11:43:08 +0200 Received: by europe.umh.ac.be with Internet Mail Service (5.5.2653.19) id ; Fri, 3 May 2002 11:38:45 +0200 Message-ID: From: Jean-Philippe Calbert To: "'chemistry@ccl.net'" Subject: Zoa v2.0 annonce Date: Fri, 3 May 2002 11:38:44 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id g439hOI04673 *************** * Zoa v 2.0 * *************** Zoa v2.0 is a Windows-based software designed to analyze the results of quantum chemical calculations. It mainly targets the field of organic materials and provides a series of tools to: - Visualize and analyze properties such as molecular orbitals, charges, vibration mode... - Perform in-depth analysis of absorption spectra and other excited state properties. - Build molecules and polymer chains, measure distances and many other features... To know more about Zoa v2.0, visit our web site at http://zoa.freeservers.com You can also freely download the viewer/demo version of Zoa v2.0 at http://zoa.freeservers.com/viewer.htm If you have other questions, feel free to contact us at zoa@averell.umh.ac.be The Zoa development team ------------------------ Jean-Philippe Calbert E-mail: Jean-phi@averell.umh.ac.be Service de Chimie des Matériaux Nouveaux Profs J.L. Brédas/R. Lazzaroni University of Mons-Hainaut Belgium From chemistry-request@server.ccl.net Fri May 3 10:05:07 2002 Received: from pcserver.iqm.unicamp.br ([143.106.51.10]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43E57I16971 for ; Fri, 3 May 2002 10:05:07 -0400 Received: from localhost (localhost.iqm.unicamp.br [127.0.0.1]) by pcserver.iqm.unicamp.br (Postfix) with ESMTP id A74093E17; Fri, 3 May 2002 11:05:05 -0300 (BRT) Received: from iqm.unicamp.br (luxor.iqm.unicamp.br [143.106.51.146]) by pcserver.iqm.unicamp.br (Postfix) with ESMTP id BF9F43E0C; Fri, 3 May 2002 11:05:03 -0300 (BRT) Message-ID: <3CD2990A.7060106@iqm.unicamp.br> Date: Fri, 03 May 2002 11:04:58 -0300 From: Milton Taidi Sonoda User-Agent: Mozilla/5.0 (X11; U; FreeBSD i386; en-US; rv:0.9.7) Gecko/20020123 X-Accept-Language: en-us MIME-Version: 1.0 To: weizhuang Cc: CHEMISTRY@ccl.net Subject: Re: CCL:help References: <200205022202.g42M29Rw6381812@mail.rochester.edu> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: by AMaViS perl-11 Hi As you have done your first box with xyzedit I guess your molecules are too close to each other. Try to run the minimize program with this very firt box, and after that the newton program. In the dynamic, try to do long runs with use very short time steps (~0.1fs or even lesser) to avoid to get superposing sites. And, as the system is getting relaxed, try use longer time steps. Yet, I sugest you to use the canonical ensemble, with the key TAU-TEMPERATURE with the modifier equal to your time step, during this relaxing runs. If you are simulating "infinite" system, you must use the A-AXIS keyword. Good luck Milton Taidi Sonoda weizhuang wrote: >Dear Sir: > >I met some problems here. I am using tinker.I want to do MD to a system with >32 CH2(OH)CH(OH)CH2(OH) molecules( this is a preconditioning of the solvent >which I want to use in a ab initio MD simulation). first I create a molecule, >then I use "xyzedit" to create a periodic box (15.87A for each side) and fill >in the 32 molecules. then I wrote the key file. it is: > >-------------------------------- >ARCHIVE >EWALD >A-AXIS 15.87 >ALPHA 90.0 >THERMOSTAT ANDERSEN >parameters mm3.prm >----------------------------- >then I use "newton" to optimize the system ( by the way, it didn't reach a >0.01 GMS but a "SMALLfct" what is it?) > >then I run "dynamic". but it seems a terriblly huge calculation because after >1 hour the monitor is still idle. and when I took off the "A-AXIS 15.87 >ALPHA 90.0" keywords and then run the dynamic. the system just "fly off"( it >is natual because it is a isolated system) > >so could you please tell me what is wrong with my operations above? > >-= 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 Fri May 3 14:56:52 2002 Received: from baton.phys.lsu.edu ([130.39.182.71]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43IuqI29227 for ; Fri, 3 May 2002 14:56:52 -0400 Received: from [130.39.183.42] (cclms9600b.phys.lsu.edu [130.39.183.42]) by baton.phys.lsu.edu (8.9.2/8.9.2) with ESMTP id NAA09051 for ; Fri, 3 May 2002 13:54:25 -0500 (CDT) User-Agent: Microsoft-Entourage/10.0.0.1309 Date: Fri, 03 May 2002 13:57:03 -0500 Subject: Making Initial configurations of polymers From: Satyavani Vemparala To: Message-ID: Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Hi: Is there any software (free or otherwise) out there which can build the initial chain of PMMA, given bond lengths and bond angles? Iam looking for a united atom model initial chain. And also once one chain is made , what is the best way to proceed to make the initial configuration of the system?( I usually make duplicates, put them all in the box together and apply some repulsive potential to introduce excluded volume but that takes quite a while)? Thanks Vani From chemistry-request@server.ccl.net Fri May 3 15:02:51 2002 Received: from dedalus.lcc.ufmg.br ([150.164.65.10]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43J2oI29519 for ; Fri, 3 May 2002 15:02:51 -0400 Received: from localhost (noronha@localhost) by dedalus.lcc.ufmg.br (8.9.3/8.9.3) with ESMTP id QAA62802 for ; Fri, 3 May 2002 16:02:31 -0300 Date: Fri, 3 May 2002 16:02:26 -0300 (BSC) From: antonio luiz oliveira de noronha X-X-Sender: noronha@dedalus To: chemistry@ccl.net Subject: QM/MM ( Gamess/TINKER ) help Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCL friends I had some error messages when running the tests for the GAMESS/TINKER QM/MM. Any of you had ever get this error messages and was able to figure out how to deal with then?? These calculations are the ones that came in the tinker.tar.gz file as tests to chech the execution of the program. I'm using a SP2 machine and the program was compiled to use in parallel. When unsing the IMOMM I get this one --- ENCODED Z MATRIX --- COORD TYPE I J K L M N . . . EIGENVECTOR ROUTINE EINVIT DID NOT CONVERGE FOR VECTOR 7. NORM = 1.71E-14 PERFORMANCE INDEX = 2.93E +12 (AN ERROR HALT WILL OCCUR IF THE PI IS GREATER THAN 100) EIGENVECTOR ROUTINE EINVIT DID NOT CONVERGE FOR VECTOR 102. NORM = 2.41E-14 PERFORMANCE INDEX = 2.07E +12 (AN ERROR HALT WILL OCCUR IF THE PI IS GREATER THAN 100) EINVIT HAS FAILED TO CONVERGE FOR VECTOR 102 NORMALIZATION ERROR IN -ZMTSYM- FOR VECTOR J = 13 EXECUTION OF GAMESS TERMINATED ABNORMALLY AT 12:53:15 CST 3-MAY-2002 ON NODE 0, STEP CPU TIME = 2.67 TOTAL CPU TIME = 2.8 ( .0 MIN) TOTAL WALL CLOCK TIME= 6.0 SECONDS, CPU UTILIZATION IS 46.00% And for the SIMOMM I get this ---- QMMM procedure is ON ---- ***** CURRENT QMMM IMPLEMENTATION REQUIRES DLC COO RDINATES ***** EXECUTION OF GAMESS TERMINATED ABNORMALLY AT 11:42:28 CST 3-MAY-2002 ON NODE 0, STEP CPU TIME = .02 TOTAL CPU TIME = .3 ( .0 MIN) TOTAL WALL CLOCK TIME= .0 SECONDS, CPU UTILIZATION IS 100.00% *** ERROR TERMINATION *** IN COMPUTE PROCESS 0 DATA SERVER STATS: TOTAL DISTRIBUTED MEMORY USED (MEMDDI)= 1 MWORDS. FIRST DATA SERVER'S MAXIMUM MEMORY= -1 WORDS, CPU= .2 SECONDS. Thank you for your help | | |Antonio Luiz Oliveira de Noronha - noronha@dedalus.lcc.ufmg.br | |EEPEMM - Eletronic Spectra and Electromagnetic Properties of Molecules| |Theoretical Physical Chemistry | |Chemistry Departament - DQ | |UFMG - Federal University of Minas Gerais - Brazil | |Av. Antonio Carlos, 6627 - Campus Pampulha | |CEP 31270-901 - Belo Horizonte - MG - Brazil | |VOICE +55 31 499 5757 - FAX +55 31 499 5700 | |______________________________________________________________________| From chemistry-request@server.ccl.net Fri May 3 13:50:37 2002 Received: from umhsp02.umh.ac.be ([193.190.192.27]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43HobI27243 for ; Fri, 3 May 2002 13:50:37 -0400 Received: from europe.umh.ac.be (europe.umh.ac.be [193.190.132.12]) by umhsp02.umh.ac.be (8.9.3/8.9.3) with ESMTP id TAA22470 for ; Fri, 3 May 2002 19:50:27 +0200 Received: by europe.umh.ac.be with Internet Mail Service (5.5.2653.19) id ; Fri, 3 May 2002 19:46:02 +0200 Message-ID: From: Jean-Philippe Calbert To: "'chemistry@ccl.net'" Subject: Appology (Zoa v2.0 announcement) Date: Fri, 3 May 2002 19:46:01 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id g43HobI27248 Dear CCL members, After the announcement of Zoa v2.0 today, we had a little problem with our account on the server that give rise to commercial pop up appearing on your screen. This problem has now been fixed! We do apology again The Zoa development team Zoa web site: http://zoa.freeservers.com ------------------------ Jean-Philippe Calbert E-mail: Jean-phi@averell.umh.ac.be Service de Chimie des Matériaux Nouveaux Profs J.L. Brédas/R. Lazzaroni University of Mons-Hainaut Belgium From chemistry-request@server.ccl.net Fri May 3 15:18:30 2002 Received: from postal.sdsc.edu (IDENT:pMYW6VZCt+MATk7K5Pt78l8I3JopSDk0@[132.249.20.114]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43JITI30223 for ; Fri, 3 May 2002 15:18:29 -0400 Received: from koa.sdsc.edu (IDENT:1dpFsNxKA3JiXEm32G0pgX+yJpdZ0JNk@koa.sdsc.edu [132.249.23.116]) by postal.sdsc.edu (8.11.6/8.11.6/server/38) with ESMTP id g43JIPn20129; Fri, 3 May 2002 12:18:25 -0700 (PDT) Received: from localhost by koa.sdsc.edu (SGI-8.9.3/1.11-IRIXclient) with ESMTP id MAA31260; Fri, 3 May 2002 12:18:24 -0700 (PDT) Date: Fri, 3 May 2002 12:18:24 -0700 From: Wibke Sudholt To: Tapas Kar cc: Subject: Re: CCL:Non convergence in SCF calculation (GAMESS program) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=X-UNKNOWN Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by server.ccl.net id g43JIUI30224 Hi, this worked in my case: * If you use direct SCF, set FDIFF=.FALSE. in $SCF group. * If you have many diffuse functions, enlarge QMTTOL in $CONTRL group. You may also change the optimizer and play around with other options in the $SCF group. I hope this helps! Best regards, Wibke Sudholt University of California, San Diego On Thu, 2 May 2002, Tapas Kar wrote: > Hi: > I got SCF convergence problem in Gamess. It looks energy is oscillating. > Is there any option in gamess to manage the convergence problem? > Tapas > *********************************************** > "We owe a lot to the Indians, who taught us how to count, without > which no worthwhile scientific discovery could have been made." > - Albert Einstein - > "The value of pi was first calculated by Budhayana, and he explained > the concept of what is now known as the Pythagorean Theorem. British > scholars have last year (1999) officially published that Budhayan’s > works dates to the 6th Century, which is long before the European > mathematicians." > ---------------------------------------------------------------------------- > ---- > Tapas Kar, Ph. D > Department of Chemistry & Biochemistry > Utah State University > Logan, UT 84322-0300 > > Tel: 435-797-7230 > Fax: 435-797-3390 > Email: tapaskar@cc.usu.edu > tapas@risky3.chem.usu.edu From chemistry-request@server.ccl.net Fri May 3 14:42:12 2002 Received: from dedalus.lcc.ufmg.br ([150.164.65.10]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43IgBI28771 for ; Fri, 3 May 2002 14:42:12 -0400 Received: from p333 (lcc-ip11.lcc.ufmg.br [150.164.65.210]) by dedalus.lcc.ufmg.br (8.9.3/8.9.3) with SMTP id PAA53420; Fri, 3 May 2002 15:41:48 -0300 Message-ID: <001301bd165b$803da300$d241a496@p333> From: "=?iso-8859-1?Q?Andr=E9_M._de_Oliveira?=" To: "Angelo Favia" , References: <001801c1f271$42f0ad20$c6b3ccc1@farmo2> Subject: Re: CCL:analysis of qsar data set. Date: Thu, 1 Jan 1998 00:17:58 -0200 Organization: UFOP MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0010_01BD164A.B6A68800" 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 This is a multi-part message in MIME format. ------=_NextPart_000_0010_01BD164A.B6A68800 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Mr. Favia, Try splitting your dataset into two or more subsets, performing the = analysis in separate. Greets. ----- Original Message -----=20 From: Angelo Favia=20 To: chemistry@ccl.net=20 Sent: Friday, May 03, 2002 5:08 AM Subject: CCL:analysis of qsar data set. Dear netters, I am a student in the last year of the pharamaceutical chemistry at=20 the Univ. of Bari (Italy). As a project for my degree, I need to=20 analise a qsar data set (~130 compounds) in which pKi data are=20 unevenly distributed. Specifically, pKi data, which span from 7 to=20 12, are binned as follows 7-8: ~5% 8-9: ~16% 9-10: ~57% 10-11: ~26% 11-12: ~2% Given this distribution, my point is that there can be some risk to=20 find out unreliable q2 values during the leave one out calculation=20 and also that some validation analyses as the scrambling of data=20 can't help in a situation like that. I wonder if I can still be happy performing qsar on the whole data=20 set ( which is what I whish, possibly ) or if it would be better = select=20 a more uniform subset of points. Have you any clue about that? thanks, Angelo Favia ------=_NextPart_000_0010_01BD164A.B6A68800 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Dear Mr. Favia,
 
Try splitting your dataset into two or = more=20 subsets, performing the analysis in separate.
 
Greets.
----- Original Message -----
From:=20 Angelo=20 Favia
Sent: Friday, May 03, 2002 5:08 = AM
Subject: CCL:analysis of qsar = data=20 set.

Dear netters,
I am a student in = the last year=20 of the pharamaceutical chemistry at
the Univ. of Bari (Italy). As = a=20 project for my degree, I need to
analise a qsar data set (~130 = compounds)=20 in which pKi data are
unevenly distributed. Specifically, pKi = data, which=20 span from 7 to
12,  are binned as follows
7-8: ~5%
8-9: = ~16%
9-10: ~57%
10-11: ~26%
11-12: ~2%
Given this = distribution, my=20 point is that there can be some risk to
find out unreliable q2 = values=20 during the leave one out calculation
and also that some validation = analyses as the scrambling of data
can't help in a situation like=20 that.
I wonder if I can still be happy performing qsar on the whole = data=20
set ( which is what I whish, possibly ) or if it would be better = select=20
a more uniform subset of points.

Have you any clue about=20 that?

thanks,
Angelo=20 Favia
------=_NextPart_000_0010_01BD164A.B6A68800-- From chemistry-request@server.ccl.net Fri May 3 16:15:10 2002 Received: from rrzs2.rz.uni-regensburg.de ([132.199.1.2]) by server.ccl.net (8.11.6/8.11.0) with ESMTP id g43KFAI32070 for ; Fri, 3 May 2002 16:15:10 -0400 Received: from listserv.ngate.uni-regensburg.de (kroesus2.rz.uni-regensburg.de [132.199.2.208]) by rrzs2.rz.uni-regensburg.de (8.9.3/8.9.3-URRZ-Sol-2.7-01) with ESMTP id WAA15060 for ; Fri, 3 May 2002 22:15:09 +0200 (MET DST) Received: from KROESUS2/SpoolDir by listserv.ngate.uni-regensburg.de (Mercury 1.48); 3 May 02 22:15:09 +0200 Received: from SpoolDir by KROESUS2 (Mercury 1.48); 3 May 02 22:15:06 +0200 From: "Georg Schmeer" Organization: Universitaet Regensburg To: chemistry@ccl.net Date: Fri, 3 May 2002 22:14:58 +0200 MIME-Version: 1.0 Content-type: text/plain; charset=ISO-8859-1 Subject: Compiling g98 A11 with ATLAS:Summary Message-ID: <3CD30BE2.17859.38477F@localhost> Priority: normal X-mailer: Pegasus Mail for Win32 (v3.12c) Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from Quoted-printable to 8bit by server.ccl.net id g43KFEI32071 Dear all, thanks to Malcolm Gillies, Dr. Peter Burger, and Philipp Matz for their answers concerning my problems with gaussian98 compiling. My original mail from Friday, May 2 2002 was: *********** Dear G98-users, I need some help for the compiling of gaussian98 Rev.A11. I got Gaussian98 Rev.A11 from Gaussian,Inc on one CD containing two directories. One directory contains the gaussian98 Rev.A11 tar file and the other the tar file of the ATLAS-linear algebra system and also blas- and f77blas-libraries for IBM AIX-Systems. After dearchiving the two tar-files I compiled the ATLAS-library according to the appropriate instructions on a Pentium III processor under Suse- LINUX 7.2. The compilers are gcc and PGF90. After a rather long time the compilation finished without error messages and produced 7 libraries in the directory ATLAS/lib/Linux_PIIISSE1_2: libatlas.a libcblas.a libf77blas.a liblapack.a libptcblas.a libptf77blas.a libtstatlas.a In order to compile g98 I copied these libraries in the directory $g98root/g98/lib and made the following modifications in the file g98.make: BLAS = $g98root/g98/lib/libatlas.a \ $g98root/g98/lib/libcblas.a \ $g98root/g98/lib/libf77blas.a \ $g98root/g98/lib/liblapack.a \ $g98root/g98/lib/libptcblas.a \ $g98root/g98/lib/libptf77blas.a \ $g98root/g98/lib/libtstatlas.a The compilation of g98 via "bsd/bldg98 >& bldg98.log &" stops shortly before the end with the following lines in bldg98.log: >. >. if ( 1 ) then >. >. make[1]: Leaving directory `&g98root/g98´ pgf90 -O2 -tp p6 -Munroll -Mvect=cachesize:524288 -o g98 ml0.o util.so util.so: undefined reference to `atl_f77wrap_dspmv_´ util.so: undefined reference to `atl_f77wrap_dgemm_´ make: *** [g98] Error 1 endif The error message is the same when compiling with pgf77. In the directory ATLAS/interfaces/blas/F77/src/f77wrap I found the generic routines atl_F77wrap_gemm.c and atl_F77wrap_spmv.c but no hint for the double-precison routines respectively. The Gaussian 98 Release Notes Revisions A.11.2 and A.11.3 give instructions for Fortran Compiler and Libraries for Building the Intel Linux Version on page 14, these instructions are useless, however, since the appropriate libraries are lacking. Therefore any help to locate the error which I made, would be appreciated. Is there any people familiar with compiling of gaussian98 together with ATLAS? ******************** But all answers are not really valuable to solve my problem and by accident I found the solution, outlined in the following: 1. Instead of atlas3.2.1.tar.gz I downloaded atlas3.3.15.tar.gz. After gunzip atlas3.3.15.tar.gz and tar -xvf atlas3.3.15.tar making a directory ATLAS with several files and subdirectories the compiling procedure is the following in the directory ATLAS/ according to INSTALL.txt: 2. make answer to all questions with the default values until > use express setup? 3. answer with n and then answer to all questions with the default values until > Enter f77 compiler [/usr/bin/g77]: Here the appropriate Fortran-compiler with complete path must be given ( In my case it is PGF77 or PGF90 with the appropriate path) The resulting questions can be replied by the default values. 4. the command make install arch= may take some hours. It produces 7 libraries in the directory ATLAS//lib/ 5. The libraries are copied into g98root/lib/ and 6. in g98root/g98/bsd/g98.make the lines 51 and 52 are commented and then the following line is inserted BLAS = $g98root/g98/lib/libstsatlas.a \ $g98root/g98/lib/libf77blas.a \ $g98root/g98/lib/libcblas.a \ $g98root/g98/lib/liblatlas.a \ $g98root/g98/lib/libptcblas.a \ $g98root/g98/lib/libptf77blas.a 7. After setting g98root and making sure that the appropriate pgi-compiler is available start the g98-compilation in the directory g98root/g98: 8. bsd/bldg98 >& bldg98.log & The compilation worked well until the end without error messages. A test-run with test429.com (to find in the subdirectory g98root(g98/tests/com/) yields the following results: Gaussian98 Rev.A.11 with ATLAS-routines Job CPU time: 2 minutes and 42.8 seconds Gaussian98 Rev.A.11 with standard BLAS-routine Job CPU time: 11 minutes and 59.7 seconds The difference in time is remarkable but naturally one test is not enough to calculate a mean gain in computer time. Obviously there is a problem in the release 3.2.1 of atlas, which is provided as stable version, whereas the release 3.3.15 is marked as unstable developper version. However this version seems to give good and reliable results in conjunction with gaussian98. I hope this information can also help other users of gaussian98 even with lower releases than A.11. Sincerely yours Georg Schmeer Prof. Dr. Georg Schmeer Institut fuer Physikalische und Theoretische Chemie Universitaet Regensburg D-93040 Regensburg Georg.Schmeer@chemie.uni-regensburg.de Tel: 0941-943-4745 Fax: 0941-943-1690