From shen@athena.cs.uga.edu Thu Jun 3 22:52:48 1993 Date: Fri, 4 Jun 93 02:52:48 EDT From: Mingzuo Shen Message-Id: <9306040652.AA16384@athena.cs.uga.edu> To: chemistry@ccl.net Subject: computational chemists employed in fields other than drug design? Hi all, Dr. Philippe Youkharibache posted a very informative mail on "how to become a computational chemist" in the pharmaceutical industry. How about industries working in materials (nonlinear optics, for example; and strange materials, etc.) and energy "production" (of course, energy is conserved, but there are ways to "use" and "transform" energy which are better than others?)? What are they looking for in a potential employee (me included)? Just wondering ... Mingzuo Shen From martin@link.sunet.se Fri Jun 4 11:45:18 1993 Date: Fri, 4 Jun 93 09:45:18 +0200 Message-Id: <9306040745.AA26917@decnet.sunet.se> From: martin@link.sunet.se (Martin Norin, Dept. Biochem., Royal Inst. Technol., Stockholm, tel. +46-8-7907512, e-mail; martin@physchem.kth.se) To: "chemistry@ccl.net"@kth.sunet.se Subject: octanol/water partition coefficients...reply I think that the following reference might be of interest. The method is maybe not purely theoretical. Hydrophobic fragment constants are values of the hydrophobicity of fragmental chemical groups. If added together thay can be directly converted to a log P value (partition coefficient octanol/water). Data was derived from 1000 compunds. The reference is: Rekker, R.F., de Kort, H.M. (1979). "The hydrophobic fragmental constants; an extension to 1000 data point set". Eur. J. Med. Chem.-CHIMICA THERAPUTICA. 14, pp 479-488. The reference is rather old, maybe someone other on the net has some updates ? Greetings, Martin Norin e-mail:martin@physchem.kth.se Dept. Biochem. KTH Stockholm From moret@far.ruu.nl Fri Jun 4 08:48:49 1993 Message-Id: <199306041203.AA19790@accucx.cc.ruu.nl> From: Ed Moret Subject: response to summary modelling in industry To: chemistry@ruucmc.far.ruu.nl Date: Fri, 4 Jun 93 14:01:32 METDST Dear netters, In addition to the summary of Philippe Youkharibache, I would like to suggest that would-be modellers in the pharmaceutical industry learn from previous experience in Computer Aided Molecular Design. To complement his list of books, the following should not be forgotten: P.M. Dean, Molecular Foundations of Drug-Receptor Interaction, Cambridge University Press 1987 CAMD tools, more specific than mechanics and dynamics, have been put into algorithms like: GRID, DOCK, LUDI, GROW, HOOK, CoMFA, APEX, ALADDIN, CAVEAT, SPERM, AAA, MTD and many others. Regards Ed Moret -- ------------------------------------------------------------------------- E.E. Moret (@more@) moret@far.ruu.nl Computational Medicinal Chemistry/Department of Pharmaceutical Chemistry Faculty of Pharmacy/Utrecht University/the Netherlands Telephone (31-30)536979/536958 Facsimile (31-30)516674 ------------------------------------------------------------------------- From DSMITH@uoft02.utoledo.edu Fri Jun 4 04:13:30 1993 Date: Fri, 04 Jun 1993 09:13:30 -0500 (EST) From: "DR. DOUGLAS A. SMITH, UNIVERSITY OF TOLEDO" Subject: Re: computational chemists employed in fields other than drug design? To: shen@athena.cs.uga.edu Message-Id: <01GYZ3P1RIS6001N6R@UOFT02.UTOLEDO.EDU> In my limited knowledge, most of the computational chemistry done outside of academia is done at pharmaceutical firms. Most does not necessarily mean 1) the most money spent or 2) the most cpu cycles burned. Part of the reason for this is, I believe, that drugs and proteins have a long history of being modeled as isolated molecules or scantily solvated systems, and only recently with the advent of new hardware and software have these approximations to reality started to disappear. In the materials sciences, bulk systems and bulk properties are more important and require significantly more horsepower to calculate. So polymers, crystalline materials, zeolites, etc. are just now possible to handle in a meaningful way. What that really means is that industries which deal with these systems are 10+ years behind the pharmaceutical industries in terms of methods and uses of molecular modeling. It won't take them 10 years to catch up, however - this field moves at an exponential rate. Most of the work in these "non-pharmaceutical" areas seems to have historically taken place at federal laboratories. For example, Wright- Patterson AFB has a large program in computational chemistry in the materials sciences, from polymers to nonlinear optical materials and conducting polymer systems. Battelle Pacific Northwest Laboratory (a Dept. of Energy lab) has all sorts of interesting things going on, many classified, but certainly some in the area of crystallography and catalysis. There is an active group at the Army's Edgewood Research, Development and Engineering Center (ERDEC, formerly CRDEC) doing work in several areas of "biological" chemistry such as simulants, nerve gasses, etc. - mostly small molecules (ask Chris Cramer or George Famini about these). Certainly there are many other federal laboratories I am not mentioning which have significant programs. I am only talking about the few I know something about - so my apologies to any one who feels slighted. As for industry, Ohio is full of polymer, aerospace, and related industries. The modeling efforts appear to me to be small relative to the pharmaceutical industries, but they are there and active. For example, PPG in Pittsburgh, Goodyear in Akron, BP and Lubrizol in Cleveland, and more have programs in place. In fact, there is now the NorthEast Ohio Modeling Interest Group (NEOMIG), organized by Lynn Brower, formerly of BP America, and Wayne Mattice, of the University of Akron. In addition to meetings, there is (sigh) another mail exploder. If there is enough interest I will post information about the exploder at a later date. Doug Douglas A. Smith Assistant Professor of Chemistry The University of Toledo Toledo, OH 43606-3390 voice 419-537-2116 fax 419-537-4033 email dsmith@uoft02.utoledo.edu ============================================================================ >Hi all, > > Dr. Philippe Youkharibache posted a very informative mail >on "how to become a computational chemist" in the pharmaceutical >industry. How about industries working in materials (nonlinear >optics, for example; and strange materials, etc.) and energy >"production" (of course, energy is conserved, but there are >ways to "use" and "transform" energy which are better than >others?)? What are they looking for in a potential employee >(me included)? > > Just wondering ... > >Mingzuo Shen From C1790@slvaxa.umsl.edu Tue Jun 4 03:24:30 1993 Date: 04 Jun 1993 09:24:30 -0600 (CST) From: C1790@slvaxa.umsl.edu Subject: CALCULATING MOLECULAR POLARIZABILITIES To: CHEMISTRY@ccl.net Message-Id: <01GYZ4EHRQSOGKUKVK@slvaxa.umsl.edu> DEAR NETTERS, I HAVE TWO QUESTIONS REGARDING CALCULATING POLARIZABILITIES. FIRST, IS THERE A READILY AVAILABLE METHOD FOR CALCULATING THE POLARIZABILITY OF A MOLECULE AS A FUNCTION OF FREQUENCY (WAVELENGTH)? I UNDERSTAND THAT THE FINITE-FIELD APPROACH IN MOPAC WILL NOT DO THIS. EVEN A QUICK-AND-DIRTY WOULD BE USEFUL TO ME. SECOND, I WOULD LIKE TO EXTRACT BOTH THE REAL AND IMAGINARY COMPONENTS OF THE CALCULATED POLARIZABILITIES. AGAIN, IS THERE A METHOD AVAILABLE FOR CALCULATING THE TWO COMPOENTS? THANKS. BILL WELSH DEPT. OF CHEM. UNIV. OF MISSOURI-ST. LOUIS ST. LOUIS, MO 63121 314-553-5318 EMAIL: C1790@UMSLVAXA From mes@atlas.chemistry.uakron.edu Fri Jun 4 08:35:57 1993 Date: Fri, 4 Jun 93 08:39:24 EDT From: mes@atlas.chemistry.uakron.edu (Mary Ellen Scott) Message-Id: <9306041239.AA09695@atlas.chemistry.uakron.edu> To: chemistry-request@ccl.net Subject: YSN post Jan, If this Ok, would you post it in the net. Thanks Mary Ellen There has been some inquiry on the net about job shortage and job opportunities. I would like to suggest anyone interested become a member of YSN. It is a netlink of over 3000 scientists and it free. Issues such as those above are discussed and some positive results have been obtained. To subscribe: ysn-adm@zoyd.ee.Washington.edu - just say I would like to become a member...that's all Mary Ellen From CFCIJZER@rulgl.LeidenUniv.nl Fri Jun 4 11:49:20 1993 Date: Fri, 4 Jun 1993 16:58 MET From: CFCIJZER@rulgl.LeidenUniv.nl Subject: modeling of ATP To: chemistry@ccl.net Message-Id: <01GYZKGSTYNK8WX20C@rulgl.LeidenUniv.nl> Dear netters: Nucleotides, such as adenosine-5'-triphosphate (ATP), are cofactors and/or substrates to probably 30 or 40% of all enzymes that we know of. A few of these enzymes have been crystallized, and their 3D-structure is in the Brookhaven protein databank. As such these structures would be ideal starting points for molecular modelling, e.g. for the design of other substrates of enzyme-inhibitors. Unfortunately, the tri- phosphate moiety poses considerable problems to theoretical chemists, in terms of parametrization etc. My general question is: are you aware of any modeling or theoretical studies aimed at the structure and/or charge distribution of the phosphate chain of nucleotides, such as ATP (in the presence or absence of divalent cations, such as Ca2+ and Mg2+), either as such or complexed with a protein structure. Which type of methodology seems most appropriate? A more specific question is: what does the replacement by CH2 or NH of the oxygen links do to the ATP molecule. I will summarize the responses to the list in due course. Ad IJzerman email: cfcijzer@rulgl.LeidenUniv.nl From djh@ccl.net Fri Jun 4 07:31:03 1993 From: David Heisterberg Date: Fri, 4 Jun 1993 11:31:03 -0400 Message-Id: <199306041531.AA02616@xipe.ccl.net> To: chemistry@ccl.net Subject: Diels-Alder TS follow-up Many thanks to the large number of people who responded on this topic! I got several good references, sample input files, and various helpful hints and words of encouragement. Part of the problem was due to a less than intelligent choice of Z-matrix. But it also seems, from some attempts with Gaussian and DMOl, that the BFGS method runs into some problems in this case. We're having some success now using the NLLSQ minimizer in MOPAC rather than the TS eigenvector following, with a subsequent FORCE calculation to establish the transition state. Gaussian sometimes gave problems unless a analytic hessian was calculated. DMol using BFGS tended to terminate the optimization, and using conjugate gradient it proceeded through the TS right to product. Thanks again, I really appreciate all the help. Some references and authors are listed below. Dave Heisterberg -------- Jorgensen, W.L.; Lim, D.; Blake, J.F.; "JACS" _115_ 2936 (1993) Blake, J.F; Jorgensen, W.L.; "JACS" _113_ 7430 (1991) Gajewski; Gilbert Poirier, R.A.; Burnell J. Clark, T.; "A Handbook of Computational Chemistry" and references therein Houk K.N.; "Angew. Chem. Int. Ed. Engl." _31_ 682 (1992) Houk K.N.; "JACS" _111_ 9172 (1989) Dewar; Jie; "Acc. Chem. Res." _25_ 537 (1992) Hehre; "JACS" _108_ 7381 (1986) Hehre; "JACS" _109_ 663 (1987) Loncharich, R; "JOC" _54_ 1129 (1989) From sridar%nil@isis.MNI.McGill.CA Fri Jun 4 08:46:06 1993 Date: Fri, 4 Jun 93 12:46:06 EDT From: sridar@nil.mni.mcgill.ca (Sridar Narayanan) Message-Id: <9306041646.AA16990@NIL.MNI.McGill.CA> To: jle@world.std.com Subject: 8-bit color X server for Macintosh >> Is anybody aware of a product that allows a Mac with an ethernet card to act as an 8-bit color X terminal? Also, what communication software's required to make it work? >> MacX and MacTCP (included) from Apple. Sridar From tsilva@harpo.aaec.com Fri Jun 4 08:23:17 1993 Date: Fri, 4 Jun 93 12:23:17 EDT From: Tony Silva Message-Id: <9306041623.AA01083@harpo> To: chemistry@ccl.net Subject: Mac X servers Here are many notes I've collected (and written myself) on Mac X servers. Tony ----------------------------------------------------------------------------- From scherock@allegro.mit.edu Fri Oct 23 12:07:50 1992 Date: Thu, 22 Oct 92 23:27:51 EDT From: scherock@allegro.mit.edu (Stephen F. Scherock) To: tsilva@aaec.com Subject: A bit of info on MacX Subject: 69) Where can I get an X server on a Macintosh running MacOS? eXodus from White Pine Software (603-886-9050) runs on any Mac with at least 1MB of memory and runs the X server within a standard Macintosh window. Version 3.0 [6/91] supports intermixing of X and Mac windows and the ADSP protocol. The version supports the SHAPE extension and includes DECwindows support. Apple's MacX runs on MacPlus or newer machines with >= 2MB of memory and system software 6.0.4 or later. Version 1.1 is fully X11R4-based. It supports full ICCCM-compatible cut and paste of text AND graphics between the Macintosh and X11 worlds, the SHAPE extension (including SHAPEd windows on the Macintosh desktop), an optional built-in ICCCM-compliant window manager, X11R4 fonts and colors, a built-in BDF font compiler, and built-in standard colormaps. Upgrades to MacX are available by ftp from aux.support.apple.com. Info: 408-996-1010. [Note: MacX is also the name of a vax-mac xmodem transfer utility.] Also: Liken (1-800-245-UNIX or info@qualix.com) software enables monochrome 68000 Mac applications to run on a SPARC system running X. Xport (1-800-245-UNIX (415-572-0200) or xport@qualix.com) enables Mac applications to display on an X-based workstation by turning the Mac into an X client. Intercon has a product called Planet-X which enables Mac applications to display on an X server. ----------------------------------------------------------------------------- call from Tony Silva to Jay Martin @ White Pine Software (603-886-9050) 10/23/92: - eXodus is $295/copy, 20% discount for 5-30 copies (25% for 31+), GSA discount ~$30 ----------------------------------------------------------------------------- call from Tony Silva to MacConnection 10/28/92: - don't carry MacX ----------------------------------------------------------------------------- call from Tony Silva to MacWarehouse 10/28/92: - don't carry MacX ----------------------------------------------------------------------------- call from Tony Silva to Apple (408)996-1010 10/28/92: - call Customer Assistance Center (800)776-2333 (on west coast) for product info ----------------------------------------------------------------------------- Return-Path: <@lenny.kaman.com:alan@kaman.com> Date: Thu, 29 Oct 92 06:36:21 EST From: Alan Piszcz To: tsilva@sparky.aaec.com Subject: MacX MacX is an exellent product. Though you may not enjoy using it on a MacPLUS. ----------------------------------------------------------------------------- Return-Path: <@dartvax.dartmouth.edu:Richard.A.Damon.III@dartmouth.edu> Date: 29 Oct 92 09:10:39 EST From: Richard.A.Damon.III@dartmouth.edu Subject: X windows server for MacPlus (Q) To: Tony Silva Just a quick note. I've recently been working with White Pines' eXodus. I'm about to return to MacX to veryify my subjective findings, but right now all I can give you is impressions. eXodus beats MacX in every category. Much smaller footprint in both disk space and memory requirements; much faster in execution; better, more extensive protocol support. The price is better, too. MacX is available from DEC at (I understand) $300/user as part of Pathworks. eXodus is about $199 through MacConnection. good luck, Rick Damon ----------------------------------------------------------------------------- Return-Path: Date: Thu, 29 OCT 92 05:50:26 PST From: Micro Mauler To: CRVAX:;@crvax.sri.com Mmdf-Warning: Parse error in original version of preceding line at nic.near.net Subject: Re: X windows server for MacPlus (Q) Reply-To: MICRO2.SCHWER@crvax.sri.com Mmdf-Warning: Parse error in original version of preceding line at nic.near.net Does any one know of any X windows servers (free, shareware, or commercial) that will run on a MacPlus? Does anyone have any experience in comparing the commercial products MacX and eXodus? I know that eXodus costs about $300, and that MacX is from Apple, but that's about it. Where can I buy MacX? MacX is available thru authorized Apple dealers and costs about $300+. The latest version is for System 7 only. I don't have any experience with this software, but would be interested in a summary of response you receive. --Len micro2.schwer@sri.com ----------------------------------------------------------------------------- call from Tony Silva to MacConnection (800)334-4444 10/28/92: - eXodus is $199 - there may be a quantity discount for 10+ copies - currently on back order, they should receive s/w end of next week ----------------------------------------------------------------------------- Return-Path: <@relay2.uu.net:hallofjustice!bursik@uunet.uu.net> Date: Fri, 30 Oct 92 01:47:45 EST From: Dave Bursik To: tsilva@sparky.aaec.com Subject: Re: X windows server for MacPlus (Q) Tony, Unless you've got a hefty accelerator stashed in that Mac Plus, I wouldn't plan on trying to run X on it. Given that you would also need to run TCP/IP, I suspect the processing load would eat your machine and make X effectively unusable. Also, if you only have the built-in monitor, you don't really have enough screen real estate to run X. (X is a pig even on SPARCstations.) If all you need is terminal windows, then something like VersaTerm (with CTB) might do the trick. I think it can do something like 8 simultaneous sessions (term windows). Also, there's the free MacLayers program (assumes you have a UNIX host on the other end). I use it and it works pretty well. Sorry I can't be more upbeat. -Dave Bursik LCC, Inc. / Arlington, VA / (703) 516-7398 / uunet!lccinc!bursik ----------------------------------------------------------------------------- Return-Path: To: tsilva@aaec.com Subject: MacX Organization: The Dimensional Gate, Inc. From: brg@dgate.org (Brian Gaeke) Date: Wed, 28 Oct 92 22:17:01 -0500 X-Mailer: Fernmail 1.1 In info-mac 10:259, >Where can I buy MacX? MacX can be had through APDA, try apda@applelink.apple.com Brian Gaeke, The Dimensional Gate, Inc. Internet: brg@dgate.org --- ----------------------------------------------------------------------------- Return-Path: Date: Tue, 3 Nov 1992 15:05:04 -0500 (EST) From: HOLT@dstl86.gsfc.nasa.gov Subject: X windows server for MacPlus (A) To: tsilva@sparky.aaec.com X-Vmsmail-To: SMTP%"tsilva@sparky.aaec.com" You asked, > Does any one know of any X windows servers (free, shareware, or > commercial) that will run on a MacPlus? Does anyone have any > experience in comparing the commercial products MacX and eXodus? I > know that eXodus costs about $300, and that MacX is from Apple, but > that's about it. Where can I buy MacX? I'm a little late on catching up with back issues of Info-Mac, but since I don't see any replies to your questions, I'll give it a try. I am using MacX on a Mac II with a 19" B&W display, and it works great. I've never compared it with eXodus. The reviews I read (MacWorld, I think) suggested that you need a good deal of power to be happy with X on a Mac (or any platform for that matter), but I haven't tried it on something on the level of a MacPlus. You can buy MacX from an Apple dealer, or even straight from Apple itself. In my case, the company I work for bought it from Falcon Microsystems, a GSA computer supplier. MacX should be cheaper than $300; I think list is around $200. >From the MacX user guide: "MacX version 1.1.7 runs on any Macintosh with at least 2 MB of RAM, no less than 128K of ROM, and either a hard disk or a floppy disk drive and access to a file server. MacX operates at its best if your Macintosh has 4 MB or more of RAM and a hard disk drive. MacX requires system software version 6.0.5 or later (including system software version 7.0)..." Hope this helps, Brett Holt ** holt@dstl86.gsfc.nasa.gov ----------------------------------------------------------------------------- call from Tony Silva to Christie @ MacConnection (800)334-4444, 10/28/92: - eXodus is $199 + $3 shipping = $202 - I gave her AAEC account #16888 - I should call back w/ PO # - s/w (in-stock) will then be shipped airborne overnight ----------------------------------------------------------------------------- call from Tony Silva to @ MacConnection (800)334-4444, 10/28/92: - I gave person PO #B2617 - should receive s/w tomorrow ----------------------------------------------------------------------------- Date: Thu, 12 Nov 92 14:14:46 PST From: ke_reich@rom.tcpl.ucalgary.ca Subject: X Windows on the Mac If anyone has information on X Windows packages on the mac, I would appreciate hearing about them. I have looked at both MacX and Exodus and neither package fits my needs. My needs are fairly straightforward. I have an SCO Unix box running Open DeskTop. On this machine, I have a phone switch management package that is X unaware. However, In a SCO Term window, I am able to focus the mouse and use the Xmouse to control the cursor for the management application. With MacX, I am unable to get color windows, because MacX foolishly (and I stress foolishly) creates new windows that have nothing to do with the Open DeskTop screen, hence I lose color in the windows, and this is important to the users. As well, focusing the mouse doesn't work, and mouse support is critical. Exodus at least gives me the color and mouse support, but it's slower than pushing rope up a hill. What is interesting to me is that my Mac IIcx with A/UX running X is significantly faster than running Exodus on my Mac IIci. Again if anybody has information directing me to a better package (or platform - a DOS machine running X will be significantly faster, but politically incorrect :-). Thanks in advance, and I will summarize to the net, provided of course that there are answers. As an aside, please reply to the below listed address as I do not read info-mac. Kevin Reichert phone: (403) 267-6290 ke_reich@rom.tcpl.ucalgary.ca fax: (403) 267-8991 ----------------------------------------------------------------------------- Date: 21 Apr 1993 14:32:58 -0500 (CDT) From: "Michael W. Wheeler" Subject: Summary: MacX or eXodus (2nd try) This is my second try at sending this to info-mac. Bill sent the file back saying it seemed to be messed up. The MAC-L list didn't have any problem with it. The returned message looked fine all except it had somebody elses message appended to it. But I changed the line that said "Thanks to:" to be "Thanks to" just in case some mailer was choking on the "to:" thinking it was supposed to mail it somewhere else. Hope it works this time and sorry for the trouble. - - - original post - - - I'd like to thank all those who took the time to reply to my questions about MacX and eXodus. In summary o Both MacX and eXodus support color. o You need MacX 1.2 for system 7.1. Which is availble on the March 93 CD for those on DEC's educations CSLG program. o eXodus from White Pine may be a little slower than MacX but eXodus is more configurable. However, White Pine is announcing eXodus 5.0 which is supposed to have some performance improvments according to MacWeek. o Performance can be improved by using native Mac windows for your clients rather than a remote window manager. This may depend on the speed of the host and the connection between Mac and host. o One report of eXodus crashing when a low level debugger was installed. No mention of the version of eXodus or the debugger. o MacX doesn't come with FTP support however eXodus does. o Some said it was slower than an X-terminal while others stated the opposite. Is probably dependant on differences in X-terminal speed as well as Mac speed, network speed and load, etc. o There was some comments saying that the speed bottle-neck was on the drawing speed of the Mac. Possibly the translation of X commands to QuickDraw commands and the speed of QuickDraw. o Those with experience on X workstations said performance would be slower regardless of the Mac setup. If I get a Mac for doing X-windows on then I can report my 2 cents worth on that subject later. o Over whelming response that MacX was totally brain dead when it came to mapping the 3-button mouse to the keyboard. eXodus is supposedly a little better in this respect but everyone said BUY A 3-BUTTON MOUSE! o Some recommended MacX other recommended eXodus. But the basic tone of all the messages was that there wasn't anyone that hated either product. So since MacX is included with Pathworks for the Macintosh I'll use it. If it's extremely frustrating then I'll take a look at eXodus. Thanks to Stan Kerr, kerr@ux1.cso.uiuc.edu Spencer Marks, spencer@MathWorks.COM Peter Fenyes, PFENYES@CMSA.gmr.com Glockzin Donald, Glockzin_Donald@macmail1.fwrdc.rtsg.mot.com Jeff Schlimmer, schlimme@eecs.wsu.edu Dave Leiner, leiner@leiner.mtv.gtegsc.com Mel Martinez, mem@jhufos.pha.jhu.edu Trevor Hyde, trevor.hyde@NSCSupport.COM Tony Silva, tsilva@sparky.aaec.com F. Arthur Cochrane, cochranea@srs.gov JP Weaver, weaver@sherlock.chinalake.navy.mil Brian Hutchison, brian@harpo.ampr.ab.ca Malcolm Slaney, malcolm@apple.com for replying to me. Michael W. Wheeler (Bitnet: mww@tntech) (Internet: mww@tntech.edu) VAX/VMS Systems Support, Tennessee Technological University ----------------------------------------------------------------------------- Date: Tue, 20 Apr 93 20:37:30 EDT From: Tony Silva Subject: X server h/w and s/w, including 3-button mice (S) Here are my OPINIONS on various X-Windows-related Mac products that I've compared. I've tried both the Mouse Systems A3 Mouse and the Logitech MouseMan on my Mac LC III. I like the action of the A3 Mouse (it's optical; the MouseMan has a roller ball). However, the current CDEV for the A3 Mouse does not yet work on the LC III. Rumor has it that it's due to an Apple ADB (not Mouse Systems) bug. The folks at Mouse Systems told me a new CDEV would be released in a month or two. They have an electronic bulletin board, so getting a new version should be fast once it becomes available. The CDEV shipped with my MouseMan also did not work on the LC III, but a new version was available on Logitech's bulletin board, and it works like a champ. The MouseMan CDEV also is more flexible to configure. And now, moving on to X server software... On various intensive client-update operations that I tried (xdvi document previewing, xterm text typeout, plotting and homebrew GUI's in matlab, etc.), MacX was almost twice as fast as eXodus. I set each server application up for maximal performance at the expense of all else. I should also mention that I have an FPU in my LC III, and that the version of eXodus I was using (and presumably the MacX application, as well) uses it. There doesn't seem to be an easy way to use the Command key as a Control key in MacX as can be done using eXodus (with the appropriate "xmodmap" command on the client side). However, you can use the Caps Lock key instead. This requires removing the latching mechanism inside the keyboard, which is not that hard to do on my Apple Keyboard II. eXodus is a little more flexible w.r.t. assignments of mouse/keystroke equivalents. I'm very interested in hearing other people's opinions on X-related Mac h/w and s/w, particularly if they don't agree with my own. Many thanks to all who responded to my requests for info: tuc@cs.wisc.edu (Brian A. Cole) Bob Stewart James Stricherz Fearghas McKay mldickens@bbn.com (Michael L. Dickens) "Alan Hewat, ILL Grenoble" kerr@ux1.cso.uiuc.edu (Stan Kerr) Tony Silva, Atlantic Aerospace Elec. Corp., (617)890-4200, tsilva@aaec.com _______________________________________________________________________________ Tony Silva tel: (617)890-4200 Atlantic Aerospace Electronics Corp. FAX: (617)890-0224 470 Totten Pond Road Internet: tsilva@aaec.com Waltham, MA 02154 From dreeves@ese.ogi.edu Fri Jun 4 03:20:37 1993 Date: Fri, 4 Jun 93 10:20:37 -0700 Message-Id: <9306041720.AA00824@ese.ogi.edu> To: chemistry@ccl.net From: dreeves@ese.ogi.edu (B. Douglas Reeves) Subject: xStones benchmarks In light of the recent interest in Macintosh xWindows servers, I thought I'd post some results for some of our xWindows terminals, PCs and Quadras. Platform Colors Connection xStones HP700 8 10BaseT 69411 RS/6000 -340 8 Unix-socket 50760 Quadra 800 8 10BaseT 29437 NCD-15b 1 10BaseT 14431 486/33 clone 8 10BaseT 12734 The Q800 was using MacX and the PC was using PC-X from NCD (I believe) and 8M RAM. The HP700 is a 19" (?) xWindows terminal and the NCD-15b a 15" monochrome xWindows term. The RS/6000-340 data is for the console with the CPU under background loads only. The Q800 uses a 33HMz 68040. The PC clone was running windows and has a moderately speedy graphics card and 13" monitor. D. Reeves |B. Douglas Reeves Dept. Env. Sci.& Eng. Oregon Graduate Institute| |N6XHW on 146.900 in PDX dreeves@ese.ogi.edu everywhere else| |"..life itself represents a nonequilibrium condition." - J. F. Pankow | From d3f012@pellucidar.pnl.gov Fri Jun 4 03:04:48 1993 Date: Fri, 4 Jun 93 10:04:48 -0700 From: d3f012@pellucidar.pnl.gov Subject: Re: UV-VIS simulation responses... To: "Jon A. Erickson" Message-Id: <9306041704.AA23844@pellucidar.pnl.gov> Another implementation of the INDO1/S method can be found in my own code, Argus. I bring this up merely as noncommercial alternative to codes such as CAChe's ZINDO, Biosym's ZINDO, or HyperChem's INDO/s implementations. Argus is freely distributed via anonymous ftp (instructions below), runs on Sun, HP, IBM/6000, Cray YMP, Cray 2, DECstation. Though is does not have the extensive breadth of features of ZINDO, is is it is free, and will run much larger systems than ZINDO is capable of running. The INDO1/s method will typically reproduce low-energy UV/vis excited states within about 1000 cm-1 of experiment. It is parameterized to vertical excitations (not adiabatic), and does much better for transitions with lots of pi-pi* character. INDO/s does a less impressive job on n-pi* states. Typically, oscillator strengths are overestimated, which is in part, a byproduct of the use of singles-only CI. Including doubles (for some systems like the paracyclophanes), or using random phase approximation (RPA) for the excitation energies can dramatically improve computed oscillator strengths. However, I stress that the present INDO1/s method is parameterized for singles-only CI, vertical excitations, for low-energy excited states (< ~ 50,000 cm-1). The theory is currently incapable of reproducing Rydberg states since Rydberg orbitals are not included in the basis. The CNDO/s methods also work very well. Del Bene and Jaffe originally devised a CNDO/s method, which provided the starting point for the INDO1/s method of Zerner and coworkers. Also, I believe that Charlie Duke (Xerox?) has his own CNDO/s method. The advantage one would get from using an INDO-based method over a CNDO-based method, is that the latter does not distinguish triplet from singlet states (CNDO ignores one-center exchange integrals). Here are some publications that give representative results of the INDO1/s method in Argus: 1. "A Theoretical Examination of the Electronic Structure and Spectroscopy of the Photosynthetic Reaction Center from Rhodopseudomonas viridis" JACS 113, (1991) p. 8210 2. "Effect of a Polarizable Medium on the Charge-Transfer States of the Photosynthetic Reaction Center from Rhodopseudomonas viridis" JACS 112 (1990) p. 7828 3. "A Theoretical Examination of the Electronic Structure and Excited States of the Bacteriochlorophyll-b Dimer from Rhodopseudomonas viridis" J. Phys. Chem. 95, (1991) p. 5693 Mark Thompson Argus is available at no cost via anonymous ftp. The ftp site is pnlg.pnl.gov (130.20.64.11). Look in the argus directory. Make sure you download and read the README file first. Argus is Copyrighted and Licensed software. All you have to do is read the files License.doc and Copyright, and agree with the conditions contained in those documents. Feel free to contact me with problems/suggestions. Enjoy, Mark. ************************************************************************** Mark A. Thompson Sr. Research Scientist email: d3f012@pnlg.pnl.gov Molecular Science Research Center FAX : 509-375-6631 Pacific Northwest Laboratory voice: 509-375-6734 PO Box 999, Mail Stop K1-90 Richland, WA. 99352 Argus available via anonymous ftp from pnlg.pnl.gov (130.20.64.11) (in the argus directory). Download the README file first. Disclaimer: The views expressed in this message are solely my own and do not represent Battelle Memorial Institute, Pacific Northwest Laboratory, or any of its clients. ************************************************************************** ************************************************************************** Mark A. Thompson Sr. Research Scientist email: d3f012@pnlg.pnl.gov Molecular Science Research Center FAX : 509-375-6631 Pacific Northwest Laboratory voice: 509-375-6734 PO Box 999, Mail Stop K1-90 Richland, WA. 99352 Argus available via anonymous ftp from pnlg.pnl.gov (130.20.64.11) (in the argus directory). Download the README file first. Disclaimer: The views expressed in this message are solely my own and do not represent Battelle Memorial Institute, Pacific Northwest Laboratory, or any of its clients. ************************************************************************** From whitbeck@maxey.unr.edu Fri Jun 4 03:42:03 1993 Date: Fri, 4 Jun 93 10:42:03 PDT From: whitbeck@maxey.unr.edu Message-Id: <9306041742.AA21979@snowlake.maxey> To: chemistry@ccl.net Subject: Re: M. Whitbeck's kinetic simulation program > Dear Netters, > > I would like to hear from people who have succeded to port Mike Whitbecks > kinetic simulation program (available from www.ccl.net) to the > MS-DOS platform. > > Thx in advance > > Peter > > I too would like to hear of any PC ports (I get requests all the time but my real job interferes with doing the things I want to do--- isn't that always the way? :-) Not having a PC the following may or may not help- can't try it to see. The compiler that I used automatically initializes allocated storage to nil. Apparently many (most?) compilers don't. I would suggest changing the malloc() calls to calloc()'s and setting newly allocated memory to nil. [warning! some compilers in the past had ill behaved calloc() as a built-in... make sure yours works!] Also I believe there is a bug in the code for lsoda.c Either pick another comparable stiff ode solver or try this: The last routine in lsoda.c is freevectors(). I believe it is being called prematurely. It need not be called at all on most platforms. On my development platform free() does not really do anything terribly useful so premature calls to freevectors() has no ill effects. On a PC this may not be the case at all. Try making freevectors() a "dummy" or do-nothing call. here is what I did static void freevectors() { /* everything commented out! free( yh ); free( wm ); free( ewt ); free( savf ); free( acor ); free( ipvt ); */ } /* end freevectors */ Finally I have had heard, informally, of mixed results using Borland and Microsoft C compilers. I hope to find the time this fall/winter to rewrite the program with more documentation and make a version that is PC and Mac friendly (assuming I can find the development platforms)--- I never really considered the PC a viable platform for serious work because of the memory restrictions of DOS. (serious wrt to large # of reactions). I won't program around the DOS limitations so any final code would still be limited by the 640K barrier. Again this is assuming I can find a DOS box with a C compiler. I will also use a more robust and more portable memory allocation scheme (like dbmalloc) in the future. One last tip. It is my understanding that compilers for DOS like machines typically default to a relatively small stack size. Bumping the stack to a larger size might be necessary for PC or Mac. Try 8k - 16k as a start. [some compilers produce executable code that warns of stack overflow and some don't] With the changes made above the program runs on sun/sparc and sgi platforms. cheers, Mike _-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-_ Mike Whitbeck whitbeck@maxey.unr.edu 702-673-7348 _-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-__-_-_-_-_-_-_ From mail Fri Jun 4 12:58:32 1993 Date: Fri, 4 Jun 1993 13:05:55 -0400 From: hyper!hurst (Graham Hurst) Message-Id: <9306041705.AA27269@hyper.hyper.com> To: jaeric@mtcamm.monsanto.com, chemistry@ccl.net (OSU Comp. Chem. List) Subject: Re: UV-VIS simulation responses... One of the responses in Jon Erickson's summary about codes for simulation/prediction of UV-Vis spectra (it's not clear to me in the post whether John P Lafemina wrote it or not) included: > HyperChem from AutoDesk also has ZINDO, I believe. It has the same methods as the ZINDO program, but does not include any code from the ZINDO program. It has published ZINDO *science* (and results from HyperChem can thus be published) but not proprietary ZINDO code. HyperChem Release 3 for Windows can calculate and display UV/vis spectra using CNDO, INDO, MINDO/3, MNDO, AM1, PM3, INDO/1 or INDO/S semi-empirical quantum mechanics methods. (The latter two are the methods in Zerner's ZINDO program, and INDO/S is the one specifically developed for reproducing UV/vis spectra.) It is a standard feature of HyperChem, so running the calculation is straightforward yet quite powerful, with modest hardware requirements. The implementation is entirely written in C, so there is no software limitation on the number of atoms. The code can handle elements up to Xe (though only INDO/1 and INDO/S can do transition metals), though parameters may need to be added for some elements (we included all the published parameters we could find and users can add more). To obtain a UV/vis spectrum, one first performs a CI calculation limited to singly-excited configurations. Either an orbital criterion or energy criterion can be used to select configurations. (HyperChem's CI can also use multiply excited states.) HyperChem Release 3 for Windows requires at least a 386 PC with fpu, 4 Mb RAM and Windows 3.1 (running in Enhanced mode). The output can be displayed as a spectrum, written to a file, or queried by another Windows application via DDE (for example into an Excel spreadsheet). UV/vis data can also be input to HyperChem for display via a text script file or DDE. Cheers, Graham ------------ Graham Hurst Hypercube Inc, 7-419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 internet: hurst@hyper.com From maloneyhuske%phvax.dnet@smithkline.com Fri Jun 4 12:15:32 1993 Date: Fri, 4 Jun 93 16:15:32 -0400 Message-Id: <9306042015.AA16741@smithkline.com> From: maloneyhuske%phvax.dnet@smithkline.com To: "chemistry@ccl.net"%INET.dnet@smithkline.com Subject: Synthetic chemistry software info needed I'm trying to find out how computational methods other than modelling are being used to help synthetic and medicinal chemists. I've seen a few of the synthetic chemistry software packages (e.g. CAMEO, CHIRON, LHASA), and am wondering if and how this type of software is being used. Wendy Warr's Directory of Chemistry Software is a good listing, but I would really like to get comments from users out there. If people using any software as an aid in synthesis would email me about their experiences, I'd be happy to summarize. Thanks --keith Keith MaloneyHuss maloneyhuske@smithkline.com From chiremv!andromeda!jeffb@uunet.UU.NET Fri Jun 4 08:00:39 1993 Date: Fri, 4 Jun 93 01:26:57 -0700 From: chiremv!andromeda!jeffb@uunet.UU.NET (Jeff Blaney) Message-Id: <9306040826.AA16522@andromeda> To: uunet!ccl.net!chemistry-request@uunet.UU.NET Subject: Re: summary: shortage of compuational chemists Interesting discussion. Perhaps it might be helpful to summarize by pointing out that for most of the chemical, pharmaceutical, agricultural, and biotech industry that computational chemistry calculations and software are not our product. Therefore, your chances of being hired are usually directly proportional to your ability to influence experiments. This requires the ability to interpret experimental data, model them, and design new experiments. These skills are difficult to pick up without good experimental experience (not just Chem 1 and undergrad Organic Chem). A solid background in physical organic chemistry and structure-activity and/or structure-activity relationships is a good asset. Many computational chemists are skilled at how to run various calculations, but fewer are trained in which calculations to run and what kinds of problems are amenable to computational approaches. Note that computational approaches that require more time to perform than the experiments they're trying to simulate are of questionable value. Jeff Blaney Chiron From whitbeck@maxey.unr.edu Fri Jun 4 09:22:07 1993 Date: Fri, 4 Jun 93 16:22:07 PDT From: whitbeck@maxey.unr.edu Message-Id: <9306042322.AA22466@snowlake.maxey> To: chemistry@ccl.net Subject: Re: summary: shortage of compuational chemists I appreciate the points made by Jeff Blaney. Quite often we get too involved in what we are doing and lose sight of why we are doing it! I am an experimentalist by nature and even in doing experimental research I fall into that same trap. > ... Note that computational approaches > that require more time to perform than the experiments they're trying to > simulate are of questionable value. > Of course if time is not the critical or most costly factor this may not hold. I can imagine experiments that are so costly that waiting for modeling results is quite desirable. Ultimately of course "theory guides, experiment decides" :-) I recall awhile back there was some traffic about 'has comp. chem. methods ever developed a new drug ?' I no longer recall the consensus of that conversation (or was it on the drug discovery list out of oz?). If anyone recalls a concrete example where the calculations made the difference tell us about it. mike whitbeck whitbeck@maxey.unr.edu From raman@bioc01.uthscsa.edu Fri Jun 4 14:18:50 1993 From: raman@bioc01.uthscsa.edu (C.S.RAMAN) Message-Id: <9306050018.AA03420@bioc01.uthscsa.edu> Subject: Buckminsterfullerene - Listserv. To: chemistry@ccl.net Date: Fri, 4 Jun 1993 19:18:50 -0500 (CDT) Dear Netters The new form of Carbon know as bucminsterfullerene has created an exorbitant amount of interest and excitement in various fields including Chemistry. The number of articles published in the field of "buckyball" has grown exponentially and new journals have been devoted to this area of research. The geometry and chemistry of this molecule is very interesting!! Here is a NEWS-SERVICE devoted to providing the current activity in this field and anyone interested in subscribing to it should do the following: send email to "bucky@sol1.lrsm.upenn.edu" with Subject: as a blank field (don't put anything here!) The body of the letter must say, SUBSCRIBE and please do not include any signature file with it. You can also get an encyclopedic bibliography on buckball, with > 2000 references (compiled originally by Prof. Richard Smalley!) that cites information dating back to B.C. You can get this list by including the word BIBLIO in the body of your mail. The other words accepted by this electronic inforamation router/distributer are: INTRO, UNSCUBSCRIBE, SUBMISSION, UPDATE. Enjoy! -raman -- C.S.Raman raman@bioc01.uthscsa.edu - Internet UNIX Programming & Administration 70412.2354@compuserve.com - CIS SPARC & SGI Systems raman@hermes.chpc.utexas.edu - CHPC Department of Biochemistry craman@launchpad.unc.edu 7703 Floyd Curl Dr. (210) 567-6623 [Tel] San Antonio, TX 78284-7760 (210) 567-6595 [Fax] ****************************************************************************** All Science is either Physics or stamp collecting -E.Rutherford ****************************************************************************** From raman@bioc01.uthscsa.edu Fri Jun 4 14:30:45 1993 From: raman@bioc01.uthscsa.edu (C.S.RAMAN) Message-Id: <9306050030.AA03443@bioc01.uthscsa.edu> Subject: RISC/6000 software & hardware.... To: chemistry@ccl.net Date: Fri, 4 Jun 1993 19:30:45 -0500 (CDT) Dear netters Here is a new database that catalogs huge amounts of data on the various software and hardware options available for the RISC/6000 platforms: InfoExplorer:RISC System/6000 software and hardware information TELNET JUMBO.HRZ.UNI-GIESSEN.DE or 134.176.2.14 Login: info With InfoExplorer, you can access RISC System/6000 software and hardware information in an online information base consisting of articles connected by hypertext links. A hypertext link is a one-way path from one piece of information to another. Cheers -raman -- C.S.Raman raman@bioc01.uthscsa.edu - Internet UNIX Programming & Administration 70412.2354@compuserve.com - CIS SPARC & SGI Systems raman@hermes.chpc.utexas.edu - CHPC Department of Biochemistry craman@launchpad.unc.edu 7703 Floyd Curl Dr. (210) 567-6623 [Tel] San Antonio, TX 78284-7760 (210) 567-6595 [Fax] ****************************************************************************** All Science is either Physics or stamp collecting -E.Rutherford ****************************************************************************** From mail Fri Jun 4 20:37:08 1993 Message-Id: <04JUN93.21135517.0029.MUSIC@NEMOMUS> Date: Fri, 04 Jun 93 19:34:11 EDT From: SC18000 To: Subject: RE:MOPAC93 & small UNIX Hi Netters, I'm at a small university and have been operating MOPAC6.0 via coprocessor boards (AEON Inc., no defunct) I'm not likely to get anything like an RS6000 anytime soon, but I have limited access to one in physics. I'd like to get MOPAC93 in a DOS friendly disk, because my system recognizes only DOS flavor in 3.5" format. It will do a standard UNIX (GENIX5.3) from my 5.25" port. I can get the MOPAC 93 on tape and read it into the RS6000 and transfer it to 5.25" media and thence take my chances with GENIX5.3. Now, my real question. Has anybody out there transferred anything successfully from an RS6000 directory to a 3.5" media in DOS WRITE format? If so would you share the keystrokes it takes to do this? Likewise, if you have successfully detarred the MOPAC93 to an RS6000 please share the exact keystrokes you used to do this. Thank you very much for all the help you can give. Sincerely, Ken Fountain sc18@NEMOMUS