From chemistry-request@server.ccl.net Mon Jan 22 03:41:01 2001 Received: from hawk.dcu.ie (mail.dcu.ie [136.206.1.5]) by server.ccl.net (8.11.0/8.11.0) with ESMTP id f0M8f0d05262 for ; Mon, 22 Jan 2001 03:41:01 -0500 Received: from dcu.ie (136.206.86.63) by hawk.dcu.ie (5.1.050) id 3A3F6BC300086E4B for chemistry@ccl.net; Mon, 22 Jan 2001 08:41:54 +0000 Message-ID: <3A6BF262.64723463@dcu.ie> Date: Mon, 22 Jan 2001 08:42:11 +0000 From: Dermot Brougham X-Mailer: Mozilla 4.73 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 To: chemistry@ccl.net Subject: Summary:molecular modelling workshop Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit A couple of weeks ago I asked a question about running a molecular modelling workshop, thanks to all who replied. The original question and summary are below. Sorry about the delay Original question: >I am currently applying for funding to run a workshop in molecular >modelling for our 4th year group (40 students). I plan to give 4-5 >lectures and then have a series of workshops (maybe 4x3 hours) with >progressively less supervision as time goes on. Ending up with the >students picking a problem of their own (probably from their organic >chemistry lectures) and tackling it using whatever methods they see as >fit and then writing a short report. Maybe the people who read the CCL >have done something similar. I would be grateful for any advice, >particularly about: >Software: Best? Value for money? >Hardware: Do you need a PC each or can you have them in pairs, >threes....? >Pedagogy: >Has there been much published on this type of course? >What are the major issues? > Does the course I have outlined make pedagogical >sense? >Or is there a superior alternative approach? Replies: The consensus is that Spartan is the best software Andreas Ehlers replied "It provides a very good GUI and contains all the methods from MM, semi empirical to ab initio, so the students don't have to change programs and write specific inputs for each problem. But it is expensive." Yes cost is the problem here, several others agreed with Andreas. Among the free options were MOE, recommended by David ENOT "Recently, I ve seen a demonstration of MOE which seems to be advantageous : you can perform classical modeling tasks as well as drug design in a highly pedagogical way, all the platforms are supported and it's free for education. Have a look at http://www.chemcomp.com " I haven't had a chance to try it yet. Also free is WebMolecules Over 200,000 Molecular Models Online in Full 3D http://www.webmolecules.com Requires 3D plug-in. Indexed and searchable. Although this isn't quite what I was looking for myeself! For a little more money: Molecules-3D 3D Molecular Model Building http://www.molecules.com/m3d25int.shtml Great for new students. Only $49 Molecules-3D Pro 3D Molecular Modelling http://www.molecules.com/m3d25int.shtml Ideal for students interested in Computational Chemistry. $99 As recommended by John Vermeer Several people from MSI contacted me talking about Materials Studio "http://staging.msi.com/materials/studio/ Or if you wish to evaluate a copy of Materials Visualizer: http://staging.msi.com/materials/studio/quote/" I have yet to try this! Several respondants recommended third party individuals who have run workshops, I am in the process of contacting these people, I won't give their names as they haven't replied yet. But if they throw up any nuggets I will inform the CCL. In terms of pedagogy there appears to be many ways to teach MM successfully, thanks to Hens Borkent, Michael Nolan and others here. Most replies suggested that what I had outlined will work, but that it is very important to keep the groups small (stagger the sessions if necessary). People disagreed as to whether one or two per PC is best, probably two if the students are "general chemists". It was agreed that you should never have more than two per PC and also that the course should be short if it is for general chemistry, i.e. 2 or at most 3 weeks. What Hens said was good: He uses "a web interface to mopac calculations, based on the Chime plugin, as you can find under http://www.cmbi.kun.nl/wetche/organic/ It requires W95/98 Netscape 4.5 - 4.7 to function properly, in particular the 'editor'. On the editor page you will also find suggestions for exercises. Since this is not really a modelling course, a number of input files have been constructed beforehand, and the students simply submit them, but they can also edit the files if they want to. On practical questions: We prefer them to work in pairs, which forces them to discuss the subject among eachother, talk about it. Three people with one mouse is too many. It is nice indeed to have a goal in the end: either a kind of competition, who comes up with the best figures, lowest minimum, etc., or a presentation, in Powerpoint or HTML (chime), as you propose. It is also our experience that courses of this type should not take too long. There should be regular interrupts, classical lectures etc, otherwise they start "playing" with the computer, chatting, e-mail, browsing. The main goal is to keep them interested in the subject, add the right amount of interactivity to keep them busy for a while, then have a break and get the group together in a discussion of the results. Sometimes you have luck if you ask them to suggest their own problems, but it is wise to have some problems at hand. I noticed that many organic chemists work on very large molecules these days, biological or supra- molecular, that are not so easy to model." One follow up question, I was expecting that Tinker would be mentioned, but it wasn't; Tinker is free; is it any good? Easy for students? GUI? Thanks again Dermot -- Dr. Dermot Brougham, School of Chemical Sciences, Dublin City University, Dublin 9, Ireland. Tel: 353 1 7005472 Fax: 353 1 7005503