From chemistry-request@server.ccl.net Fri May 14 06:37:55 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA00243 for ; Fri, 14 May 1999 06:37:55 -0400 Received: from mail-c.bcc.ac.uk (mail-c.bcc.ac.uk [144.82.100.23]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id GAA17270 Fri, 14 May 1999 06:34:12 -0400 (EDT) Received: from socrates-a.ucl.ac.uk by mail-c.bcc.ac.uk with SMTP (PP); Fri, 14 May 1999 11:34:07 +0100 From: uccatvm Message-Id: <25362.199905141034@socrates-a.ucl.ac.uk> Subject: Re: high quality basis sets for heavy atoms To: chemistry@www.ccl.net (CCL) Date: Fri, 14 May 1999 11:34:05 +0100 (BST) Cc: T.vanMourik@ucl.ac.uk (Tanja van Mourik) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Hi Ping, For the 3rd-row atoms (Br, Kr) there are correlation consistent basis sets available, up to cc-pV5Z. You can download them from http://www.emsl.pnl.gov:2080/forms/basisform.html. For really large atoms, ECPs may be more appropriate, because they include some relativistic effects. Unless of course you do relativistic calculations. Hope this helps, Tanja -- ==================================================================== Tanja van Mourik phone University College London work: +44 (0)171-504-4665 Christopher Ingold Laboratories home: +44 (0)1895-259-312 20 Gordon Street e-mail London WC1H 0AJ work: T.vanMourik@ucl.ac.uk United Kingdom home: tanja@netcomuk.co.uk ==================================================================== From chemistry-request@server.ccl.net Thu May 13 15:23:06 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id PAA25644 for ; Thu, 13 May 1999 15:23:06 -0400 Received: from mailer.psc.edu (mailer.psc.edu [128.182.58.100]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id PAA09713 Thu, 13 May 1999 15:19:36 -0400 (EDT) Received: from psc.edu (hdsx9.psc.edu [128.182.61.62]) by mailer.psc.edu (8.8.7/8.8.7/mailer) with ESMTP id PAA19972; Thu, 13 May 1999 15:19:36 -0400 (EDT) Message-ID: <373B267E.8E8C5F1B@psc.edu> Date: Thu, 13 May 1999 15:22:38 -0400 From: Craig M Sager Organization: Pittsburgh Supercomputing Center X-Mailer: Mozilla 4.5 [en] (WinNT; I) X-Accept-Language: en-US MIME-Version: 1.0 Newsgroups: bionet.biology.computational,bionet.molec-model To: chemistry@www.ccl.net, beowulf-announce@beowulf.gsfc.nasa.gov CC: ropelews@psc.edu Subject: WORKSHOP: DEVELOPING AND USING PC CLUSTERS AS A BIOMEDICAL RESEARCH TOOL Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The Pittsburgh Supercomputing Center's Biomedical Initiative will be offering the following NIH-sponsored training workshop: DEVELOPING AND USING PC CLUSTERS AS A BIOMEDICAL RESEARCH TOOL Workshop Dates: June 17-19, 1999. Application deadline: May 20, 1999 This workshop is designed to assist biomedical researchers who are interested in establishing a PC cluster for use by their laboratory. The workshop will address software and hardware topics encountered in the process of building, operating and managing a PC cluster. In addition, software development issues will also be described. Topics of discussion will include: "What do you need to consider when creating an operating environment?" "How to select System Software?" "How to Administer the Cluster?" "What applications are appropriate?" and "How to get the best performance for your money?" Full details, including on-line applications and financial support, may be found on the PSC Web site: http://www.psc.edu/biomed/workshops/workshops.html The Pittsburgh Supercomputing Center's Biomedical Initiative is an NIH Supported Resource Center. From chemistry-request@server.ccl.net Thu May 13 18:46:08 1999 Received: from mserv1.dl.ac.uk (root@mserv1.dl.ac.uk [148.79.160.65]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id SAA27654 for ; Thu, 13 May 1999 18:46:08 -0400 Received: from tcsg7 (tcsg7.dl.ac.uk [193.62.112.46]) by mserv1.dl.ac.uk with SMTP id XAA03951 (8.8.8/5.4[ref postmaster@dl.ac.uk] for dl.ac.uk from P.Sherwood@dl.ac.uk); Thu, 13 May 1999 23:42:36 +0100 (BST) Precedence: first-class Received: by tcsg7 (950413.SGI.8.6.12/930817.MJE) id XAA01973; Thu, 13 May 1999 23:28:02 +0100 Date: Thu, 13 May 1999 23:28:02 +0100 From: P.Sherwood@dl.ac.uk (Paul Sherwood) Message-Id: <199905132228.XAA01973@tcsg7> To: CCL , Ping Lin Subject: Re: CCL:high quality basis sets for heavy atoms Ping, did you look on the basis set library at PNNL? http://www.emsl.pnl.gov:2080/forms/basisform.html At least for some elements you mention (certainly Br, I) the all-electron Sadlej pvtz basis sets are there. Paul From chemistry-request@server.ccl.net Fri May 14 04:42:30 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id EAA31778 for ; Fri, 14 May 1999 04:42:30 -0400 Received: from fsuj20.rz.uni-jena.de (fsuj20.rz.uni-jena.de [141.35.1.18]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id EAA17109 Fri, 14 May 1999 04:38:45 -0400 (EDT) Received: from al.bundy.uni-jena.de (al.bundy.uni-jena.de [141.35.147.40]) by fsuj20.rz.uni-jena.de (8.9.1a/8.9.1) with SMTP id KAA09833 for ; Fri, 14 May 1999 10:38:44 +0200 (MET DST) Message-Id: <199905140838.KAA09833@fsuj20.rz.uni-jena.de> Received: by al.bundy.uni-jena.de (1.37.109.4/16.2) id AA16516; Fri, 14 May 99 10:43:30 +0200 From: Michael Braeuer Subject: summary: large ab initio calc's To: CHEMISTRY@www.ccl.net (chemistry list) Date: Fri, 14 May 99 10:43:30 METDST Mailer: Elm [revision: 70.85] Hi, Here comes the summary to my question about large ab initio calcs. Thank you very much to all who replied. Micha Original question: I need to know the approximate size of the largest ab initio calculations done with standard programs so far. And here are the responses. ***************************************************************** Hi Micha, dies haengt natuerlich entscheidend von dem ab, was Du rechnen moechtest. Auf welchem Level und mit welchem Basissatz moechtest Du denn rechnen? Auf MP2/6-31g** wuerde ich sagen ca. 50-60 Atome, also single point! Und dies gilt fuer org. Molekuele ohne Metalle! Viel Erfolg C. -- ----------------------------------------------------------------- Christoph Schneider csc@imb-jena.de Institute for Molecular Biotechnology Beutenbergstrasse 11, -Biocomputing- Tel.: +49-3641-65-6205 P.O. Box 100 813, D-07708 Jena, Germany Fax: +49-3641-65-6210 http://www.imb-jena.de/~csc/ ----------------------------------------------------------------- ******************************************************************* Ich finde, das ist eine sehr unpraezise Frage! "large" kann -meiner Meinung nach- naemlich folgendes bedeuten: 1. viele Atome (am wahrscheinlichsten) 2. viele Basisfunktionen (d.h. man kann kleine Molekuele mit grossen Basissaetzen rechnen und grosse Molekuele werden eher mit kleinen Basissaetzen gerechnet, wobei die >Zahl< der Funktionen gleich sein kann. 3. "large" im Sinne von viel Speicherplatz (Ram oder Festplatte). Dies kann man sehr einfach durch correlierte Methoden (MP4, CCSD, usw.) "erreichen" >Any hint where I could start searching ? z.B. A. Hishikawa, A. Iwamae, K. Hoshina, M. Kono, K. Yamanouchi Mass-resolved two-dimensional momentum imaging of the Coulomb explosion of N2 and SO2 in an intense laser field Chem. Phys. Lett. 282 (1998) 283 Gaussian Inc wirbt mit der Berechnung der Molekuels Taxol auf einem Multiprozessorsystem fuer die Qualitaet des Parallel-codes (siehe Dokumentation) Gruesse Wolfgang Roth ==W=R===========================W=o=l=f=g=a=n=g==R=o=t=h== University of Leeds School of Chemistry Leeds LS2 9JT UK ===== http://www-public.rz.uni-duesseldorf.de/~rothw ===== *********************************************************************** Message 10/34 from Frank Jensen May 11 '99 at 12:14 (noon) Return-Path: Subject: Re: CCL:Large ab initio calculations check some of Gustavo Scuceria's lattest work, primarily in JCP. Frank *********************************************************************** In general I don't know how you'd go about finding such things, apart >from careful and continuous reading of the literature. I have recently done RI-MP2 calculations with 2460 AO basis functions without taking advantage of symmetry using the NWChem parallel computational chemistry code. RI-MP2 is an approximate version of the familiar MP2 correlated method. It is surely among the largest calculations done to date at this level, though I will not claim it is _the_ largest. If you want more information, I can send you pointers on the RI-MP2 method and on the NWChem code; these particular calculations are not quite written up yet. I would be interested to know what other large calculations you learn about. Good luck, David -- David E. Bernholdt | Email: bernhold@npac.syr.edu Northeast Parallel Architectures Center | Phone: +1 315 443 3857 111 College Place, Syracuse University | Fax: +1 315 443 1973 Syracuse, NY 13244-4100 | URL: http://www.npac.syr.edu ************************************************************************* Hi Micha, I'm not sure what you mean by standard, but we have carried out RHF/STO-3G calculations with 2000 atoms and 6000 basis sets using the serial version of MondoSCF. I'm pretty sure this is the largest to date. See my page for info/papers. With best regards, Matt +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + Matt Challacombe, Ph.D. http://www.t12.lanl.gov/~mchalla/ + + Los Alamos National Laboratory email: mchalla@t12.lanl.gov + + Theoretical Division vmail: (505) 698-4112 + + Group T-12, Mail Stop B268 phone: (505) 665-5905 + + Los Alamos, New Mexico 87545 fax: (505) 665-3909 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ************************************************************************ you may want to look at a calculation on crambin, a protein containing 46 aminoacids (642 atoms). JPC A, vol. 102, 2246-2251 (1998) by Yu, Peeters, Martin, Sch\"afer and myself. sincerely, kris --------------------------------------------------------------------------- Kris Van Alsenoy, Quantum Chemistry, Structural Chemistry Group, Department of Chemistry, University of Antwerp (UIA), Universiteitsplein 1, B-2610 Antwerp, Belgium. E-mail : alsenoy@uia.ua.ac.be / Phone : +32(3)8202366 / Fax : +32(3)8202310 --------------------------------------------------------------------------- ************************************************************************ -- Michael Braeuer http://www.uni-jena.de/chemie/anders/micha.htm PhD Student E-mail: brauer@al.bundy.uni-jena.de IOMC FSU Jena Tel: ++49/3641/948217 Humboldtstrasse 10 Fax: ++49/3641/948212 07743 Jena Germany From chemistry-request@server.ccl.net Fri May 14 06:29:25 1999 Received: from mailhost.rdg.ac.uk (exim@sumh1.rdg.ac.uk [134.225.16.4]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA00149 for ; Fri, 14 May 1999 06:29:23 -0400 Received: from rubens ([192.133.244.118]) by mailhost.rdg.ac.uk with smtp (University of Reading Email Service) id {10iF9x-00078N-00} ; Fri, 14 May 1999 11:25:41 +0100 Sender: ssr97lrc@ccl.net Message-ID: <373BFCA4.41C6@rdg.ac.uk> Date: Fri, 14 May 1999 11:36:20 +0100 From: "l.r.cooper" Organization: pedal X-Mailer: Mozilla 3.01SGoldC-SGI (X11; I; IRIX 6.3 IP32) MIME-Version: 1.0 To: CHEMISTRY@ccl.net CC: ssr97lrc@reading.ac.uk Subject: Summary PDB superposition and RMS Content-Type: multipart/mixed; boundary="------------167E2781446B" This is a multi-part message in MIME format. --------------167E2781446B Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable Dear Lee, This question had also been posted several years ago, and I enclose the response received in that time, together with a useful piece of software. Regards. Jose I. Garcia From: = "Rino Ragno (uniroma1.it)" To: = "l.r.cooper" Hi, PROFIT 17 does what you are looking for, have a look at: www.biochem.ucl.ac.uk/~martin/programs/index.html Ciao From: = Paul@pdchem.demon.co.uk (Paul Davis) To: = ssr97lrc@reading.ac.uk Dear Lee, Unless you are looking at different conformations of the same molecule, this is not a trivial problem because you have to decide which atoms in each molecule have to be equivalenced. Here are two suggestions. 1) ProFit by Andrew Martin is an excellent, flexible program. http://www.biochem.ucl.ac.uk/~martin/programs/index.html#profit 2) The 'Motif' routine in WhatIf (by Gert Vriend) is especially worth looking at; it offers automatic alignment of molecules in three dimensions. The program suggests which atoms should be equivalenced. WhatIf is a large and complex modelling and structural analysis package, and so may not be exactly what you were asking for; however you can write scripts to routinely perform a given set of tasks. http://www.sander.embl-heidelberg.de/whatif/ ProFit probably fits the bill. http://www.biochem.ucl.ac.uk/~martin/programs/index.html cheers, Malcolm -- Malcolm Gillies PhD student, computational medicinal chemistry Dept Medicinal Chemistry, Faculty of Pharmacy, Utrecht University, The Netherlands There are many programs try http://www.biochem.ucl.ac.uk/~martin/swreg.html =3D FroFit v1.8 Miri From: = "Mauro A. Cremonini" To: = "l.r.cooper" Hi! I did it once in fortran. I used the orthogonal procrustes algorithm by Golub and Pereira. Unfortunately I haven't the code anymore, but it was some 30 lines of fortran code and one call to the lapack library. If you want I can fax you 4 pages with the original reference. Hope it helps, Mauro Try SwissPDB viewer Claude Aflalo ######################## Phones: Office 972-7-6472118 #### Dept. of Life Sciences Lab 972-7-6472119 Ben Gurion University of the Negev Fax 972-7-6472890 P.O.Box 653 email: aflaloc@bgumail.bgu.ac.il Beer Sheva 84105 Israel URL: http://www.bgu.ac.il/life/aflalo.html Hi Lee, I suggest you use my ProFit program: http://www.biochem.ucl.ac.uk/~martin/programs/ Hi, I recommand you use: (1) the superpose module in Sybly (2) TINKER (3) Modeller both tinker and modeller are public available without any charge for academic coummunity, and Tinker also offers the F77 (?) source codes, and the c codes translated by f2c. Good luck. Dayong I don't know of any c code that will do that = but I know that you can do it with MSI's Cerius. Nathalie Godbout | | University of Illinois at Urbana-Champaign | | 600 S. Goodwin Av. Box 23.6 | | Urbana, IL 61801 = -- = =B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8= =B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4=B0= ` //=3D\Mr. Lee R. Cooper Bsc(Hons) MSc, Parallel, Emergent//=3D\evolution,= \=3D//& Distributed Architechture Laboratory, Dept. of \=3D//creationis= ts = //=3D\Comp. Science, University of Reading, Whiteknights //=3D\have argued, = \=3D//PoBox 225, Reading, Berkshire. RG6 6AY (UK) \=3D/was as lik= ely //=3D\Tel +44 (0)118 987 5123 x7645 (0915-1700)|Fax //=3Dto have made = \=3D//Tel +44 (0)118 931 8188 x7817 (1900-2300)|0118 975 \=3D//man as a= = //=3D\http://www.rdg.ac.uk/~ssr97lrc/ 1994 //=3D\monkey..., =B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,= =B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4=B0`=B4=B0=A4=F8,=B8=B8,=F8=A4= =B0` randomly typing the bible without error. with natural selection acting as an eraser of mistakes; assuming thirteen mistakes for every correct character & a rate of 1 keystroke per second, the 6 million characters of the holy script could be typed in 13x6e6 s, or <3 yrs. --------------167E2781446B Content-Type: text/plain; charset=us-ascii; name="overlay.prog" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="overlay.prog" Dear net people, With the recent interest in molecular super-positioning I thought I'd post a copy of my program that I sent to Mike Smith (who originally wrote to the list wanting a program to overlay two molecules). The program takes two PDB files and two lists of atom numbers and produces a translated/rotated version of the second PDB file so the specified atoms are overlayed as closely as possible. I'm sending the source with this message, hopefully it should get through uncorrupted. If there are any problems due to more than 80 characters per line (my text editor is set up with about 150 columns) I could ftp it to you or something. There are 3 source files, "overlay.c", "utils.c" and "utils.h". After you've chopped them out of this file, make the program by doing cc -o overlay overlay.c utils.c You'll also need a "-lm" flag on SGI's as for some reason the maths part of the standard C library isn't automatically linked in. Then to overlay the two PDB files you do overlay This will produce a file called ".super". And tell you the RMS distance between the atoms you specified to be overlayed. The atom list files should have one atom id per line. The atom id can be just its index number, or you can specify the chain too. The format of each line is... [] where things in brackets [] are optional. eg. 118 4560 etc or 118 A 118 B 4560 B etc Obviously there must be the same number of atoms in each list. I hope this is useful. If there are any problems or queries send me a message. yours, Simon Kilvington (srk@uk.ac.soton) The source... ================================================================= ================================================================= /* overlay.c Simon Kilvington, 1994 */ #include #include #include #include "utils.h" #define MAXFILENAMELEN 256 #define PDBLINELEN 82 /* length of a line in a pdb file, inc space for "\n\0" terminator */ #define INDEXCOLUMN 6 /* position of atom number in ATOM records */ #define CHAINCOLUMN 21 /* position of chain id in ATOM records */ #define XPOSCOLUMN 30 /* position of x coord in ATOM records */ #define YPOSCOLUMN 38 #define ZPOSCOLUMN 46 #define COORDLEN 8 /* width of coord fields */ #define COORDDPS 3 /* number of decimal places for coords */ #define SUPERPOSTOLERANCE 0.000000175 /* when superpositioning give up when tan(rotation needed) is less than this, ~tan(0.00001degs) */ #define MAXSUPERPOSITS 100000 /* or we need more than this number of iterations */ typedef struct { char pdb[PDBLINELEN]; int index; /* pdb atom number */ char chain; /* chain id, resolves atom number clashes */ vector pos; } atomdetails; typedef struct { int natoms; atomdetails atom[1]; /* array extends to atom[natoms-1] */ } fragdetails; #define FRAGHDRSIZE (sizeof(fragdetails) - sizeof(atomdetails)) fragdetails *readpdbdata(char *); void writepdbdata(char *, fragdetails *); int *readatomlist(char *, fragdetails *, int *); int findatomindex(fragdetails *, int, char); void superpositionfrags(int, fragdetails *, int *, fragdetails *, int *); void calcsuperposmatrix(int, fragdetails *, int *, fragdetails *, int *, matrix *, vector *, vector *); /* main params should be the should be lists of atoms (and, optionally, chain id's) to be overlapped, there must be the same number in each file produces a pdb file called ".super" */ int main(int argc, char **argv) { fragdetails *pdb1, *pdb2; int *atom1, *atom2, natoms1, natoms2, a; float rms; vector dr; char filename[MAXFILENAMELEN]; if(argc != 5) { printf("Useage: overlay \n"); return 1; } /* read in the files */ if((pdb1 = readpdbdata(argv[1])) == NULL) return 1; if((atom1 = readatomlist(argv[2], pdb1, &natoms1)) == NULL) return 1; if((pdb2 = readpdbdata(argv[3])) == NULL) return 1; if((atom2 = readatomlist(argv[4], pdb2, &natoms2)) == NULL) return 1; if(natoms1 != natoms2) { printf("Both atom lists must have the same number of atoms\n"); return 1; } printf("Moving \"%s\"\n", argv[3]); superpositionfrags(natoms1, pdb1, atom1, pdb2, atom2); strcpy(filename, argv[3]); strcat(filename, ".super"); writepdbdata(filename, pdb2); /* report the rms distance between atom1[] and atom2[] */ rms = 0.0; for(a=0; aatom[atom1[a]].pos, &pdb2->atom[atom2[a]].pos, &dr)); rms = sqrt(rms / natoms1); printf("RMS distance between atoms to overlap: %.3f\n", rms); printf("Number of atom pairs: %d\n", natoms1); free(pdb1); free(atom1); free(pdb2); free(atom2); return 0; } fragdetails * readpdbdata(char *filename) { fragdetails *frag, *newptr; atomdetails *aptr; FILE *file; char buffer[PDBLINELEN]; BOOL err; if((file = fopen(filename, "r")) == NULL) { printf("Unable to read \"%s\"\n", filename); return NULL; } printf("Reading \"%s\"\n", filename); frag = malloc(FRAGHDRSIZE); frag->natoms = 0; err = FALSE; while(!feof(file) && !err) { /* read in the line */ fgets(buffer, PDBLINELEN, file); /* if this is an ATOM record add it to the growing fragment */ if(strncmp(buffer, "ATOM ", 6) != 0) continue; frag->natoms++; newptr = realloc(frag, FRAGHDRSIZE + (frag->natoms * sizeof(atomdetails))); if(err = (newptr == NULL)) continue; frag = newptr; aptr = &frag->atom[frag->natoms-1]; strncpy(aptr->pdb, buffer, PDBLINELEN); aptr->index = atoi(&buffer[INDEXCOLUMN]); aptr->chain = buffer[CHAINCOLUMN]; aptr->pos.x = atof(&buffer[XPOSCOLUMN]); aptr->pos.y = atof(&buffer[YPOSCOLUMN]); aptr->pos.z = atof(&buffer[ZPOSCOLUMN]); } fclose(file); if(err) { printf("No memory to read \"%s\"\n", filename); free(frag); } return (err) ? NULL : frag; } void writepdbdata(char *filename, fragdetails *frag) { FILE *file; int a; atomdetails *aptr; if((file = fopen(filename, "w")) == NULL) { printf("Unable to write to \"%s\"\n", filename); return; } printf("Writing \"%s\"\n", filename); for(a=0; anatoms; a++) { aptr = &frag->atom[a]; sprintf(&aptr->pdb[XPOSCOLUMN], "%*.*f", COORDLEN, COORDDPS, aptr->pos.x); sprintf(&aptr->pdb[YPOSCOLUMN], "%*.*f", COORDLEN, COORDDPS, aptr->pos.y); sprintf(&aptr->pdb[ZPOSCOLUMN], "%*.*f", COORDLEN, COORDDPS, aptr->pos.z); fprintf(file, "%s\n", aptr->pdb); } fclose(file); return; } int * readatomlist(char *filename, fragdetails *frag, int *natoms) { int *list, *newptr, index, bpos; char chain, buffer[PDBLINELEN]; FILE *file; BOOL err; *natoms = 0; if((file = fopen(filename, "r")) == NULL) { printf("Unable to read \"%s\"\n", filename); return NULL; } printf("Reading \"%s\"\n", filename); list = malloc(sizeof(int)); /* so we can use realloc from the start */ err = FALSE; while(!feof(file) && !err) { (*natoms)++; newptr = realloc(list, (*natoms) * sizeof(int)); if(err = (newptr == NULL)) { printf("No memory read \"%s\"\n", filename); continue; } list = newptr; fgets(buffer, PDBLINELEN, file); sscanf(buffer, "%d%n", &index, &bpos); while(buffer[bpos] != '\0' && isspace(buffer[bpos])) bpos++; chain = (buffer[bpos] == '\0') ? ' ' : buffer[bpos]; if(err = ((list[(*natoms)-1] = findatomindex(frag, index, chain)) == -1)) printf("Error while reading \"%s\"; PDB atom with index %d%c does not exist\n", filename, index, chain); } fclose(file); if(err) free(list); return (err) ? NULL : list; } int findatomindex(fragdetails *frag, int index, char chain) { int atom; BOOL match; atom = 0; match = FALSE; while(!match && atom < frag->natoms) { match = (frag->atom[atom].index == index && frag->atom[atom].chain == chain); if(!match) atom++; } return (atom == frag->natoms) ? -1 : atom; } /* superpositionfrags this moves yfrag so the specified atoms in it and xfrag overlap as closely as possible the xatom and yatom arrays should be natoms long and hold the numbers of the atoms that are required to be overlappped uses the Ferro-Hermans algorithm, Acta Cryst. 33 (1977) p345-347 */ void superpositionfrags(int natoms, fragdetails *xfrag, int *xatom, fragdetails *yfrag, int *yatom) { matrix spos, ypos, result; vector cgx, cgy; int a; atomdetails *aptr; matrix_alloc(&spos, 3, 3, TRUE); matrix_alloc(&ypos, 3, 1, TRUE); calcsuperposmatrix(natoms, xfrag, xatom, yfrag, yatom, &spos, &cgx, &cgy); for(a=0; anatoms; a++) { aptr = &yfrag->atom[a]; matrix_setelement(&ypos, 0, 0, aptr->pos.x - cgy.x); matrix_setelement(&ypos, 1, 0, aptr->pos.y - cgy.y); matrix_setelement(&ypos, 2, 0, aptr->pos.z - cgy.z); matrix_mult(&spos, &ypos, &result); aptr->pos.x = cgx.x + matrix_getelement(&result, 0, 0); aptr->pos.y = cgx.y + matrix_getelement(&result, 1, 0); aptr->pos.z = cgx.z + matrix_getelement(&result, 2, 0); matrix_forget(&result); } matrix_forget(&ypos); matrix_forget(&spos); return; } /* calcsuperposmatrix as superpositionfrags, but yfrag is unmoved by the experience on entry spos should be a matrix_alloc'd 3x3 matrix on exit spos is the superposition matrix, cgx is the centre of gravity of xatom, and cgy is the centre of gravity of yatom to superposition the two frags yfrag should be moved to "cgx + spos(y - cgy)" where y are the original coords of yfrag !!! this routine depends on the fact that a "vector" is composed of 3 consecutive floats that represent x, y and z coords !!! */ #define veccoord(VEC, AXIS) (*(((float *) (VEC)) + AXIS)) #define xcoord(ATOM, AXIS) veccoord(&(xfrag->atom[xatom[ATOM]].pos), AXIS) #define ycoord(ATOM, AXIS) veccoord(&(yfrag->atom[yatom[ATOM]].pos), AXIS) enum {XAXIS, YAXIS, ZAXIS}; void calcsuperposmatrix(int natoms, fragdetails *xfrag, int *xatom, fragdetails *yfrag, int *yatom, matrix *spos, vector *cgx, vector *cgy) { matrix corr; int i, j, k, p, q, r, itno; float oneovern, sum, sigma, gamma, dist, qk, rk; BOOL rotated; if(natoms < 3) { printf("Not enough atoms (%d) to superposition molecules\n", natoms); return; } matrix_alloc(&corr, 3, 3, TRUE); /* calc cgx and cgy */ cgx->x = cgx->y = cgx->z = 0.0; cgy->x = cgy->y = cgy->z = 0.0; for(i=0; iatom[xatom[i]].pos, cgx); vector_add(cgy, &yfrag->atom[yatom[i]].pos, cgy); } oneovern = 1.0 / (float) natoms; vector_scale(cgx, oneovern); vector_scale(cgy, oneovern); /* initialise the correlation matrix, c'' in the paper */ for(j=XAXIS; j<=ZAXIS; j++) { for(k=XAXIS; k<=ZAXIS; k++) { sum = 0.0; for(i=0; i SUPERPOSTOLERANCE) { dist = (float) sqrt(gamma*gamma + sigma*sigma); if(dist != 0.0) { for(k=XAXIS; k<=ZAXIS; k++) { /* the superposition matrix */ qk = matrix_getelement(spos, q, k); rk = matrix_getelement(spos, r, k); matrix_setelement(spos, q, k, (gamma*qk + sigma*rk) / dist); matrix_setelement(spos, r, k, (-sigma*qk + gamma*rk) / dist); /* and the correlation matrix */ qk = matrix_getelement(&corr, q, k); rk = matrix_getelement(&corr, r, k); matrix_setelement(&corr, q, k, (gamma*qk + sigma*rk) / dist); matrix_setelement(&corr, r, k, (-sigma*qk + gamma*rk) / dist); } } rotated = TRUE; /* make a note of the fact that a rotation was still needed */ } } itno++; } while(rotated && itno < MAXSUPERPOSITS); /* until we didnt rotate about any axis, or we give up */ matrix_forget(&corr); return; } #undef veccoord #undef xcoord #undef ycoord ================================================================= ================================================================= /* utils.c Simon Kilvington, 1994 */ #include "utils.h" float vector_len2(vector *vp) { return (vp->x * vp->x) + (vp->y * vp->y) + (vp->z * vp->z); } vector * vector_add(vector *vp1, vector *vp2, vector *vp3) { vp3->x = vp1->x + vp2->x; vp3->y = vp1->y + vp2->y; vp3->z = vp1->z + vp2->z; return vp3; } vector * vector_sub(vector *vp1, vector *vp2, vector *vp3) { vp3->x =vp1->x - vp2->x; vp3->y =vp1->y - vp2->y; vp3->z =vp1->z - vp2->z; return vp3; } vector * vector_scale(vector *v, float scalar) { v->x *= scalar; v->y *= scalar; v->z *= scalar; return v; } BOOL matrix_alloc(matrix *m, int rows, int cols, BOOL quit) { if(!(m->element = (matrix_element_t *) malloc(rows * cols * sizeof(matrix_element_t)))) { printf("Out of memory in matrix_alloc\n"); if(quit) exit(1); return FALSE; } m->rows = rows; m->cols = cols; return TRUE; } matrix * matrix_makeunit(matrix *m) { int i, j; for(i=0; irows; i++) for(j=0; jcols; j++) matrix_setelement(m, i, j, (i==j) ? 1.0 : 0.0); return m; } matrix * matrix_mult(matrix *m1, matrix *m2, matrix *r) { int i, j, k; matrix_element_t e; if(m1->cols != m2->rows) { printf("Matrices are of incompatible sizes in matrix_mult\n"); exit(1); } matrix_alloc(r, m1->rows, m2->cols, TRUE); for(i=0; irows; i++) { for(j=0; jcols; j++) { e=0; for(k=0; kcols; k++) e += matrix_getelement(m1, i, k) * matrix_getelement(m2, k, j); matrix_setelement(r, i, j, e); } } return r; } ================================================================= ================================================================= /* utils.h Simon Kilvington, 1994 */ #include #include typedef enum {FALSE, TRUE} BOOL; typedef struct { float x, y, z; } vector; float vector_len2(vector *); #define vector_len(VP) sqrt(vector_len2(VP)) vector *vector_add(vector *, vector *, vector *); vector *vector_sub(vector *, vector *, vector *); vector *vector_scale(vector *, float); typedef float matrix_element_t; /* a matrix is just a dope vector and a ptr to an array of elements */ typedef struct { int rows, cols; matrix_element_t *element; } matrix; #define matrix_elementptr(mat, i, j) (((mat)->element) + ((j) * (mat)->rows) + (i)) #define matrix_setelement(mat, i, j, val) (*(matrix_elementptr(mat, i, j)) = (val)) #define matrix_getelement(mat, i, j) (*(matrix_elementptr(mat, i, j))) BOOL matrix_alloc(matrix *, int, int, BOOL); #define matrix_forget(mat) (free((mat)->element), (mat)->rows = (mat)->cols = 0) matrix *matrix_makeunit(matrix *); matrix *matrix_mult(matrix *, matrix *, matrix *); ================================================================= ================================================================= Here endeth todays source. --------------167E2781446B-- From chemistry-request@server.ccl.net Fri May 14 06:35:00 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id GAA00225 for ; Fri, 14 May 1999 06:35:00 -0400 Received: from fsuj20.rz.uni-jena.de (fsuj20.rz.uni-jena.de [141.35.1.18]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id GAA17266 Fri, 14 May 1999 06:31:16 -0400 (EDT) Received: from pc04.chemie.uni-jena.de (pc04.chemie.uni-jena.de [141.35.29.60]) by fsuj20.rz.uni-jena.de (8.9.1a/8.9.1) with SMTP id MAA16072 for ; Fri, 14 May 1999 12:31:13 +0200 (MET DST) Received: from pc03.chemie.uni-jena.de by pc04.chemie.uni-jena.de (5.65v3.2/1.1.10.5/28Jan98-0152PM) id AA04385; Fri, 14 May 1999 12:31:12 +0200 Sender: goeller@pc04.chemie.uni-jena.de Message-Id: <373BFB70.794B@pc04.chemie.uni-jena.de> Date: Fri, 14 May 1999 12:31:12 +0200 From: Andreas Goeller X-Mailer: Mozilla 3.04Gold (X11; I; OSF1 V4.0 alpha) Mime-Version: 1.0 To: chemistry@www.ccl.net Subject: Summary free MOPAC Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Summary MOPAC for NT: I asked for a free MOPAC version running on Win NT with the following features: > 1) run on Intel/WinNT > 2) does optimization AND force (mopac 6/NBO does not) > 3) writes graph-files (I get fileio-errors) > > and perhaps also > > 4) CIS and CISD (not only CI) > 5) NBO analysis > 6) ESP > 7) FF hyperpolarizabilities The following web-pages / collegues offer solutions. Thanks to all. Some answers, on the other hand, were not so helpful and are listed at the end. Point 4 seems not to be implemented at all in the available MOPAC versions, but I was used to it from the commercial VAMP and AMPAC packages. >From the CCL archive, there is source code and a mopac NBO version, which refuses to do force calcs and topics 3, 4, 6 and I think 7. _____________________________________________________________________________ Unless for some reason you insist on Mopac6, you are welcome to try extended DOS executable (it should run Ok in a DOS box of WNT) of Mopac7 either at: ftp://ftp.ccl.net/pub/chemistry/software/MS-DOS/mopac_for_dos/mopac7/ or at: http://www.cobalt.chem.ucalgary.ca/ps/hobby/mopac7/ Due to the way it was ported to DOS/Windows, this version will be limited to only a handful of atoms under WNT, but it should be enough for an undergrad course. If you wish, you can even have a (DOS, but will run in a DOS box, too) GUI to go with it, see: http://www.cobalt.chem.ucalgary.ca/ps/moldemo/ Regards, /Serge.P This version always gave me "not enough memory errors" so I couldn't test it. _____________________ Try this one. It should do at least FORCE, ESP and NBO. However, ESP is not applicable when FORCE is used... This version is limited to 50 heavy and 100 light atoms. Use at your own risk :) Victor S. Lobanov, PhD fullfilled 1, 2, 5, 6, 7 but not 3 and 4. ________________________ There is a version available from Ralph Puchta at CCC Erlangen. It fulfills 1, 2, 3, 7. Please contact puchta@organik.uni-erlangen.de This version includes a sparcle for the element Eu. AUTHOR = "A. V. M. de Andrade, N. B. da Costa Jr., A. M. Simas, G. F. de Sa", TITLE = "Sparkle model for the quantum chemical AM1 calculation of europium complexes", JOURNAL = Chemical Physics Letter, YEAR = 1994, VOLUME = "227", PAGES = "349"} _________________________ _________________ MOPAC_2000 will do all that you ask, and on large systems. I have easily run a Lipase molecule (4100 atoms) in parallel under MPI on SGI O2K and an PC/NT cluster, also with MPI. MOPAC_2000 code is distributed by (yes!) Schroedinger, Inc. (help@Schroedinger.com) John McKelvey ___________________ Dear Andreas, You might be interested in ArgusLab. It's free and runs on Win95/98/NT. http://www.seanet.com/~mthompson/ArgusLab/index.htm Mark Thompson _______________ Thanks to all who helped me. Now the students have a version which does all we need in the course and which they also can use at home. _____________________________________________________________________________ -- Andreas Goeller --------------------------------------------------------------- Dr. Andreas Goeller Institut fuer Physikalische Chemie Friedrich-Schiller-Universitaet Lessingstr. 10 phone: +49(0)-3641-948352 D-07743 Jena fax: +49(0)-3641-948302 (secretary) Germany email: goeller@pc04.chemie.uni-jena.de http://www.uni-jena.de/chemie/photo/goeller/goeller.html --------------------------------------------------------------- Dr. Andreas Goeller ehemals Computer Chemie Centrum email: goeller@organik.uni-erlangen.de http://www.organik.uni-erlangen.de/clark/goeller/goeller.html --------------------------------------------------------------- From chemistry-request@server.ccl.net Fri May 14 08:43:54 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id IAA02144 for ; Fri, 14 May 1999 08:43:54 -0400 Received: from CHEMNA.DICHI.UNINA.IT (chemna.dichi.unina.it [192.135.165.1]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id IAA17810 Fri, 14 May 1999 08:40:12 -0400 (EDT) From: ENZO@CHEMNA.DICHI.UNINA.IT Received: from CHEMNA.DICHI.UNINA.IT by CHEMNA.DICHI.UNINA.IT (PMDF V5.2-29 #30460) id <01JB6W16CNS000069H@CHEMNA.DICHI.UNINA.IT> for chemistry@www.ccl.net; Fri, 14 May 1999 14:40:17 CET Date: Fri, 14 May 1999 14:40:16 +0100 (CET) Subject: ChiTEL99 To: chemistry@www.ccl.net Message-id: <01JB6W16FC8I00069H@CHEMNA.DICHI.UNINA.IT> X-VMS-To: IN%"chemistry@www.ccl.net" MIME-version: 1.0 Content-type: text/plain; CHARSET=US-ASCII Dear Collegues, the Organizing Committee is proud to announce the XXV International Congress of Theoretical Chemists of Neolatin Languages (XXV ChiTEL, Naples 13-18 September 1999). Communications in Italian, French, Spanish, Portuguese and any other Neolatin languages are welcome. Instructions for participants can be found in the Congress web site: http://fidel.dichi.unina.it/chitel99/chitel99.html Best regards , XXV ChiTEL Organizing Committee Dr. Andrea di Matteo From chemistry-request@server.ccl.net Fri May 14 12:05:59 1999 Received: from mail1.uct.ac.za (exim@mail1.uct.ac.za [137.158.128.73]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id MAA03858 for ; Fri, 14 May 1999 12:05:57 -0400 Received: from mort by mail1.uct.ac.za with local (Exim 2.01 #3) for chemistry@ccl.net id 10iKPc-0004ln-00; Fri, 14 May 1999 18:02:12 +0200 Received: by mail1.uct.ac.za (Mort 2.23alpha) id 580 from ACHIM@PSIPSY; Fri May 14 18:02:12 1999 From: "Achim Lienke" Organization: University of Cape Town To: chemistry@ccl.net Date: Fri, 14 May 1999 18:06:00 SAST-2 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: Hydrogen bonds with DFT, summary Priority: normal X-mailer: Pegasus Mail for Windows (v2.42a) Message-Id: A few weeks ago I posted the following question > can anybody point me to a review which compares the accuracy of different DFT > methods with respect to predicting the geometry and strength of > hydrogen bonds, especially for interactions between water and simple anions. Thanks to all of the people who replied. It seems that there is not much literature about anion-water interaction with DFT, the best article I was pointed at is Andrew Pudzianowski, J. Phys. Chem. 100, 4781-89 (1996) He published a somewhat more detailed review in Recent Res. Devel. in Physical Chem., 1 (1997), which is hard to obtain in ordinary libraries. However, Andrew was very kind to send my a reprint and I am sure he will do that to another people interested (pudziana@bms.com) I received numerous replies to my question, about 90 % of them asked me to summarize the answers. For all those, who are more generally interested in a comparison of different methods I thought it might be a good idea to make a list on the web summarizing very briefly the content of useful articles in that area. This list is under http://nitrogen.cem.uct.ac.za/achim/hbond.htm and I am very happy to add other refs. Achim ---------------------------------------------------------------------- Achim Lienke Department of Chemistry University of Cape Town Private Bag Rondebosch 7701 South Africa Phone: ++27-21-650-2325 or -2530 Fax: ++ 27-21-689-7499 email: achim@psipsy.uct.ac.za ---------------------------------------------------------------------- From chemistry-request@server.ccl.net Fri May 14 12:17:28 1999 Received: from gate-sl1.mdli.com (ns2.mdli.com [208.200.221.3]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id MAA04027 for ; Fri, 14 May 1999 12:17:27 -0400 Received: (from smap@localhost) by gate-sl1.mdli.com (8.8.8/8.8.8) id JAA16003 for ; Fri, 14 May 1999 09:03:06 -0700 (PDT) Received: from puffin.mdli.com(191.254.19.10) by gate-sl1.mdli.com via smap (V2.1) id xma016001; Fri, 14 May 99 09:03:04 -0700 Received: from shepherd.mdli.com by puffin.mdli.com (8.8.5/BCH1.0) id JAA15047; Fri, 14 May 1999 09:13:44 -0700 (PDT) Received: from mdli.com (ostrich.mdli.com [191.254.14.135]) by shepherd.mdli.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2232.9) id DG28DWJM; Fri, 14 May 1999 09:16:10 -0700 Sender: lchen@mdli.com Message-ID: <373C411E.38D5E1B8@mdli.com> Date: Fri, 14 May 1999 08:28:30 -0700 From: Lingran Chen Organization: MDL Information Systems, Inc. X-Mailer: Mozilla 4.07C-SGI [en] (X11; I; IRIX 6.5 IP22) MIME-Version: 1.0 To: chemistry@ccl.net Subject: Re: CCL:Unexpected predictions of chemical reactivity Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit My reply to jas@mail.fct.unl.pt could not be sent, so I resend it to chemistry@ccl.net. J. Aires de Sousa wrote: > > Dear CCLers, > > Could anyone send me examples of unexpected predictions of chemical > reactivity by artificial intelligence / statistical methods that were > experimentally confirmed ? The paper in J.Am.Chem.Soc. 1997, 119, 4033-4042 showed how a Kohonen neural network can be used to predict reactions. -Lingran > I'll summarise the replies that I receive. > ______ > > Joao Aires-de-Sousa, Ph.D. > Departamento de Quimica, Fac. Ciencias e Tecnologia, Univ. Nova de Lisboa, > 2825 - 114 CAPARICA, Portugal; Fax: +351 1 2948550; email: jas@mail.fct.unl.pt > From chemistry-request@server.ccl.net Fri May 14 12:41:23 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 MAA04816 for ; Fri, 14 May 1999 12:41:22 -0400 Received: [from rzusgi.unizh.ch (chburger@rzusgif.unizh.ch [130.60.68.19]) by rzusuntk.unizh.ch (8.8.5/SMI-5.25) with SMTP id SAA20174 for ; Fri, 14 May 1999 18:37:41 +0200 (MET DST)] Date: Fri, 14 May 1999 18:37:40 +0200 (MDT) From: "Dr. Peter Burger" To: CHEMISTRY@ccl.net Subject: COSMO & UHF & DFT & ECP Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear CCLers, after having tried thee+ programs without success in order to do a COSMO DFT calculation using ECP's on an open shell system (UHF) I was wondering whether Gaussian 98 is capable of doing such a thing. According to the manual it should be possible in principle, but unless you do it ...(that's what happened to me before). Best regards Peter -------------------------------------------------------------------- Peter Burger University of Zuerich chburger@aci.unizh.ch From chemistry-request@server.ccl.net Fri May 14 14:05:16 1999 Received: from charleston.softhome.net (qmailr@charleston.SoftHome.net [204.144.231.41]) by server.ccl.net (8.8.7/8.8.7) with SMTP id OAA07061 for ; Fri, 14 May 1999 14:05:15 -0400 Received: (qmail 9303 invoked by uid 417); 14 May 1999 18:22:55 -0000 Received: from 1cust165.tnt1.det1.da.uu.net (HELO kressj.www.microsoft.com) (153.34.15.165) by smtp.softhome.net with SMTP; 14 May 1999 18:22:55 -0000 Message-ID: <001901be9e33$cd39f1c0$a50f2299@kressj.www.microsoft.com> From: "Jim Kress" To: Subject: Re: CCL:COSMO & UHF & DFT & ECP Date: Fri, 14 May 1999 14:01:32 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.72.3155.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3155.0 Be careful with Gaussian 98. The documentation may say that it will perform the calculation but, as I have found, the functionality that actually works within Gaussian 98, especially the windows version, may be different from what is stated in the documentation. 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: Dr. Peter Burger To: CHEMISTRY@ccl.net Date: Friday, May 14, 1999 12:51 PM Subject: CCL:COSMO & UHF & DFT & ECP >Dear CCLers, > >after having tried thee+ programs without success in order >to do a COSMO DFT calculation using ECP's on an open shell system >(UHF) I was wondering whether Gaussian 98 is capable of doing >such a thing. According to the manual it should be possible in >principle, but unless you do it ...(that's what happened to me before). > >Best regards > >Peter > >-------------------------------------------------------------------- >Peter Burger >University of Zuerich >chburger@aci.unizh.ch > > > >-= 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 14 15:16:59 1999 Received: from ccshst09.cs.uoguelph.ca (ccshst09.cs.uoguelph.ca [131.104.96.18]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id PAA09314 for ; Fri, 14 May 1999 15:16:59 -0400 Received: from ccshst01 (ccshst01.cs.uoguelph.ca [131.104.96.14]) by ccshst09.cs.uoguelph.ca (8.8.6 (PHNE_17135)/8.8.6) with SMTP id PAA28927 for ; Fri, 14 May 1999 15:13:17 -0400 (EDT) Date: Fri, 14 May 1999 15:13:16 -0400 (EDT) From: WeiQuan Tian X-Sender: wtian@ccshst01 To: CHEMISTRY@ccl.net Subject: Re: CCL:COSMO & UHF & DFT & ECP In-Reply-To: <001901be9e33$cd39f1c0$a50f2299@kressj.www.microsoft.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII It's true to G98, especially I am interested in the new DFT functionals. The people in Gaussian told that some of the new functionals specified in the manual are not in the program actually. On Fri, 14 May 1999, Jim Kress wrote: > Be careful with Gaussian 98. The documentation may say that it will perform > the calculation but, as I have found, the functionality that actually works > within Gaussian 98, especially the windows version, may be different from > what is stated in the documentation. > > 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: Dr. Peter Burger > To: CHEMISTRY@ccl.net > Date: Friday, May 14, 1999 12:51 PM > Subject: CCL:COSMO & UHF & DFT & ECP > > > >Dear CCLers, > > > >after having tried thee+ programs without success in order > >to do a COSMO DFT calculation using ECP's on an open shell system > >(UHF) I was wondering whether Gaussian 98 is capable of doing > >such a thing. According to the manual it should be possible in > >principle, but unless you do it ...(that's what happened to me before). > > > >Best regards > > > >Peter > > > >-------------------------------------------------------------------- > >Peter Burger > >University of Zuerich > >chburger@aci.unizh.ch > > > > > > > >CHEMISTRY@ccl.net -- To Everybody | CHEMISTRY-REQUEST@ccl.net -- To > Admins > >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 > > > > > > > > > > -= 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 14 15:39:38 1999 Received: from mail.chem.tamu.edu (mail.chem.tamu.edu [165.91.176.8]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id PAA09962 for ; Fri, 14 May 1999 15:39:38 -0400 Received: from mail.chem.tamu.edu ([165.91.176.116]) by mail.chem.tamu.edu (Netscape Messaging Server 3.62) with ESMTP id 282 for ; Fri, 14 May 1999 14:37:57 -0500 Sender: "Ping Lin" Message-ID: <373C7B1A.67D0947@mail.chem.tamu.edu> Date: Fri, 14 May 1999 14:35:55 -0500 From: Ping Lin X-Mailer: Mozilla 4.51 [en] (X11; I; IRIX64 6.2 IP28) X-Accept-Language: en MIME-Version: 1.0 To: CCL Subject: CCL:summary:high quality basis sets for heavy atoms Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, all Many thanks to all those replies. Almost all the replies refer to the basis set library at EMSL: http://www.emsl.pnl.gov:2080/forms/basisform.html There are some basis set including some of the elements I am asking for, i.e. the Sadlej pvtz basis sets, cc-pVNZ( up to 5), Ahlrichs basis sets, etc. I did visit the library before posting my question, because what I most expect is a basis set in cluding all the elements I am going to deal with. I am not sure about the performance of mixed basis sets. For some elements, I will probably use basis sets with ECPs. Can someone comment on the addition of the diffused and polarization functions to these basis sets? I would like to give my thanks to P.Sherwood@dl.ac.uk (Paul Sherwood) Herbert Fruchtl T.vanMourik@ucl.ac.uk (Tanja van Mourik) larry@melanie.ucd.ie (Larry Cuffe) Ping Lin From chemistry-request@server.ccl.net Fri May 14 17:34:27 1999 Received: from www.ccl.net (www.ccl.net [192.148.249.5]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id RAA12113 for ; Fri, 14 May 1999 17:34:25 -0400 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id RAA04691 Fri, 14 May 1999 17:30:38 -0400 (EDT) Received: from rrzs2.rz.uni-regensburg.de (rrzs2.rz.uni-regensburg.de [132.199.1.2]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id RAA05018 for ; Fri, 14 May 1999 17:30:31 -0400 (EDT) Received: from rchs4.chemie.uni-regensburg.de (rchs4.chemie.uni-regensburg.de [132.199.48.4]) by rrzs2.rz.uni-regensburg.de (8.8.8/8.8.8) with SMTP id XAA17388; Fri, 14 May 1999 23:28:56 +0200 (MET DST) Date: Fri, 14 May 1999 23:32:49 +0200 From: Herbert Homeier t4720 Message-Id: <9905142132.AA28990@rchs4.chemie.uni-regensburg.de> To: CHEMISTRY@ccl.net Subject: Methods for transition states of radical cations (Summary) Cc: echamot@chamotlabs.com, frj@dou.dk, churca@q1.fcen.uba.ar, sudhirk@MahindraBT.com, nch425@merle.acns.nwu.edu Content-Transfer-Encoding: quoted-printable Mime-Version: 1.0 Reply-To: herbert.homeier@na-net.ornl.gov X-Sun-Charset: US-ASCII Dear CCL members, some time ago I posted the a question to CCL and received some answers included below and some requests for a summary. I would like to thank all those who answered. I appreciate your efforts highly. Best regards The original question: ------------------------------------------------ Some colleagues are interested in calculations of radical cations containing about 10 C, some N and O and about 15 H with extended pi systems (distonic ions). They have the following questions that I=20 would like to forward to the experts of such calculations in practice: A) What do you consider as state-of-the-art method(s) for calculating the energy of transitions states of such systems with an accuracy of about 5 kcal/mol? (Is this accuracy really realistic to expect?) B) Do you regard semiempirical methods as reliable for the calculation of the energy barriers of such systems? If yes, which ones? Also, pointers to literature on such issues would be appreciated. ------------------------------------------------- Answer no. 1: ----- Begin Included Message ----- From: Frank Jensen 5 kcal/mol should be borderline possible with various DFT methods, try e.g. B3LYP or B3PW91 (available in G98). System of this size should be no problem with DFT, unless you have very limited computing facilities. AM1/PM3 would probably have a hard time getting 5 kcal/mol accuracy, but they may be useful as a starting point for getting geometries roughly OK before turning on the DFT. Frank ----- End Included Message ----- Answer no. 2: ----- Begin Included Message ----- From: Ernest Chamot I don't have specific information about distonic ions, but (depending on what reactions you are considering) based on the literature I've seen (radical additions, abstraction reactions, electrocyclic reactions) I don't believe you will get the 5 kcal/mol accuracy you want from semiempirical methods. PM3 can come close for some activation energies (as opposed to absolute energies of transition states), but AM1 and MNDO will only be accurate to within about 10-20 kcal. See: B. H. Bessler, W. L. Hase, & K. C. Haas, J. Phys. Chem., 96, 9639-76 (1992). J. Baker, M. Muir, J. Andzelm, J. Chem. Phys., 102(5), 2063-79 (1995). M. A. Ries & J. Rodriquez, J. Comp. Chem., 13(7), 860-6 (1992). This is starting to be a big system for ab initio HF calculations, but accuracy will be a problem for you here, too. Even with MP2 correlation you may only get accuracies within about 10 kcal/mol. See: Baker et. al. A. Pross et. al., J. Phys. Org. Chem., 4, 135-40 (1991). L. Fan & T. Ziegler, J. Amer. Chem. Soc., 114, 10890-7 (1992). I think your best bet would be to use DFT, but you will need to use GGA rather than Local Density Functionals to get within 2-6 kcal/mol. See: Baker et. al. Fan & Ziegler. Recent work by B. Jursac and by J. Durant (I think these are in press, but they may have come out in the last year or so.) I hope this helps. EC --- Ernest Chamot Chamot Laboratories, Inc. 530 E. Hillside Rd. Naperville, IL 60540 (630) 637-1559 echamot@chamotlabs.com http://www.chamotlabs.com/cl ----- End Included Message ----- From chemistry-request@server.ccl.net Fri May 14 17:53:13 1999 Received: from cerberus.ulaval.ca (cerberus.ulaval.ca [132.203.250.10]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id RAA12324 for ; Fri, 14 May 1999 17:53:13 -0400 Received: from promethium.chm.ulaval.ca (promethium.chm.ulaval.ca [132.203.70.61]) by cerberus.ulaval.ca (8.8.7/8.7.3) with ESMTP id RAA13533 for ; Fri, 14 May 1999 17:57:31 -0400 (EDT) Received: (from anh@localhost) by promethium.chm.ulaval.ca (8.6.12/8.6.12) id LAA17507 for chemistry@ccl.net; Fri, 14 May 1999 11:01:12 -0400 From: "Nguyen Nam Anh" Message-Id: <9905141101.ZM17505@promethium.chm.ulaval.ca> Date: Fri, 14 May 1999 11:01:09 -0400 X-Mailer: Z-Mail (3.2.1 6apr95 MediaMail) To: chemistry@ccl.net Subject: electric field + H2+ Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Dear cclers, I'm doing very "simple" abinitio calculations on the H2+ molecular system using g94. A strong electric field has been applied (using Field keyword) which distorts the original Coulomb potential of the molecule, i.e. a linear potential is added to the Coulomb one. Thus the molecule's LUMO and HOMO can interact with continuum states of the linear potential of the electric field to form resonant states, and electron can tunnel out. The difficulty is: to generate correct resonant states, we have to provide basis sets that can support continumm states of the electric field. Standard Gaussian basis sets centered at the H atoms alone are not sufficient. My question: Does anyone know of any similar abinitio calculation? Or any basis set that supports continuum states? Thank you very much -- Nguyen Nam Anh Quebec, Canada E-mail: anh@chm.ulaval.ca WWW: http://promethium.chm.ulaval.ca/~anh/