From owner-chemistry@ccl.net Wed Oct 31 03:51:00 2007 From: "Szabolcs Csepregi scsepregi/./chemaxon.com" To: CCL Subject: CCL: Yet another software recommendation request Message-Id: <-35534-071031034634-4612-5ntI5CsinuADGjqzj/Ti/Q**server.ccl.net> X-Original-From: Szabolcs Csepregi Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=UTF-8; format=flowed Date: Wed, 31 Oct 2007 07:55:15 +0100 MIME-Version: 1.0 Sent to CCL by: Szabolcs Csepregi [scsepregi{:}chemaxon.com] Andrew D. Fant fant{=}pobox.com wrote: > What I want is something that will let me sketch a small molecule with a > reasonably simple interface, convert it to a 3D form, and do enough of a > structural minimization to get the geometry cleaned up enough that a > real MM/MD or quantum program won't choke on it. MarvinSketch can do all these, and runs nicely on Mac. See: http://www.chemaxon.com/product/marvin_land.html > Oh, and it needs to be > within the budget of a self-funded graduate student, which is to say > that I am not eager to drop several hundred dollars for it. > Marvin is free if used as a desktop application. Furthermore, it seems that you qualify for the academic package, so you can get free license for calculator plugins also. Conditions are here: http://www.chemaxon.com/forum/ftopic193.html From the calculators, at least the "Conformers" and MD plugins seem relevant for you. See the full list at: http://www.chemaxon.com/product/calc_pl_land.html I hope this helps. Szabolcs Szabolcs Csepregi, PhD Director of Search Technologies, ChemAxon Ltd. http://www.chemaxon.com Skype: szabolcs.csepregi Tel: +36 1 4532661 Cell: +36 20 4219863 Fax: +36 1 4532659 From owner-chemistry@ccl.net Wed Oct 31 11:55:01 2007 From: "Daniil Bratashov dn2010]*[gmail.com" To: CCL Subject: CCL: Q optimal configuration for G03 running? Message-Id: <-35535-071031115117-13839-jfV6tO3fFToEnzxvQTowRw..server.ccl.net> X-Original-From: Daniil Bratashov Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII Date: Wed, 31 Oct 2007 18:51:02 +0300 Mime-Version: 1.0 Sent to CCL by: Daniil Bratashov [dn2010(!)gmail.com] On Tue, 30 Oct 2007 16:47:00 -0400 "David Hose anthrax_brothers-$-hotmail.com" wrote: > or a Linux equivalent to use all 8 Gb. 32bit linux can use effectively only slightly less than 1G (1G-128M) due to implementation. All other memory will be used like very fast swap. 64bit variant is highly recommended in this case. WBR, Daniil Bratashov. From owner-chemistry@ccl.net Wed Oct 31 12:30:01 2007 From: "Noel M O Boyle baoilleach(_)gmail.com" To: CCL Subject: CCL: Yet another software recommendation request Message-Id: <-35536-071031115919-16416-vS6wvg2lmmd8TdzBoFhOgw(-)server.ccl.net> X-Original-From: "Noel M O Boyle" Date: Wed, 31 Oct 2007 11:59:15 -0400 Sent to CCL by: "Noel M O Boyle" [baoilleach++gmail.com] Avogadro can do this. It's Open Source and available for Linux, Mac and Windows. It's available from http://avogadro.sf.net Noel > "Andrew D. Fant fant{=}pobox.com" wrote: > > Sent to CCL by: "Andrew D. Fant" [fant]|[pobox.com] > Greetings all, > Having reached that point of my life where I want to spend time less > time being my own systems administrator, I've decided to replace my > (deceased) notebook with a shiny new mac (at least after macworld when > the prices drop). Given this, would people please share their favorite > 3d capable molecular editor that runs natively in MacOSX and Carbon. I > know that there are free options if I use parallels to run something > under Windows, or the X server, but I have had nothing but hassles with > that approach on the lab workstation (nasty cut and paste errors and > random crashes). I was apparently the last to discover that ChemDraw > doesn't do any real 3D functionality on the Mac after buying a copy. > What I want is something that will let me sketch a small molecule with a > reasonably simple interface, convert it to a 3D form, and do enough of a > structural minimization to get the geometry cleaned up enough that a > real MM/MD or quantum program won't choke on it. Oh, and it needs to be > within the budget of a self-funded graduate student, which is to say > that I am not eager to drop several hundred dollars for it. > > Any nominees? > > Thanks, > Andy > > From owner-chemistry@ccl.net Wed Oct 31 13:05:02 2007 From: "Lothar Terfloth lothar.terfloth+/-mol-net.com" To: CCL Subject: CCL: Announcement: Release of SYLVIA 1.0 Message-Id: <-35537-071031120938-20717-L0OZXJMfQcRyJW4rvWCKQg===server.ccl.net> X-Original-From: Lothar Terfloth Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Wed, 31 Oct 2007 16:13:23 +0100 MIME-Version: 1.0 Sent to CCL by: Lothar Terfloth [lothar.terfloth[-]mol-net.com] Dear CCL'ers, -------------------------------------------------------------- Molecular Networks Announces the Release of SYLVIA Version 1.0 for the Fast Estimation of Synthetic Accessibility of Organic Compounds -------------------------------------------------------------- SYLVIA provides a powerful method to rapidly evaluate the synthetic accessibility of organic compounds and to prioritize thousands of structures according to their synthetic complexity, e.g., structures generated by de novo design experiments or stored in large virtual compound libraries. Thus, SYLVIA links the areas of computer-aided molecular design, chemoinformatics and synthetic chemistry. SYLVIA ranks chemical compounds on a scale that reflects whether a structure can be synthesized by a straightforward synthesis route or whether it is a complex, challenging synthesis target. The scoring function is based on structure-based components, similarity to commercially available chemicals and a unique reaction-based component extracted from reaction databases. Further Information about SYLVIA -------------------------------------------------------------- Product Details: Please refer to the product page of SYLVIA at www.molecular-networks.com/software/sylvia or directly contact us at info-#-molecular-networks.com. Versions: SYLVIA is available as GUI version and as XT version (GUI and batch mode version). In addition, a component for SciTegic Pipeline Pilot is available. Evaluation: Interested parties are invited to download a 30 days evaluation copy of SYLVIA at www.molecular-networks.com/php/profile.php (Download Area), or to test it online at www.molecular-networks.com/online_demos/sylvia (SYLVIA web service). Special Introductory Offer: Save 10% of the regular 1st year license price of SYLVIA (GUI version). This offer lasts for orders until December 31, 2007. -------------------------------------------------------------- Molecular Networks GmbH www.molecular-networks.com info-#-molecular-networks.com -- -------------------------------------------------------- Dr. Lothar Terfloth Tel. +49-9131-9790623 Molecular Networks GmbH Fax: +49-9131-815 669 Henkestr. 91 D-91052 Erlangen Germany email: Lothar.Terfloth-#-mol-net.com www: http://www.mol-net.com -------------------------------------------------------- From owner-chemistry@ccl.net Wed Oct 31 14:22:01 2007 From: "Oliver T. Hofmann o.hofmann__tugraz.at" To: CCL Subject: CCL: Optimizing Geometry Optimizations in VASP Message-Id: <-35538-071031112739-10924-gynGQ2pnIlbCNIfo+Slqlw/a\server.ccl.net> X-Original-From: "Oliver T. Hofmann" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=UTF-8; format=flowed Date: Wed, 31 Oct 2007 15:47:26 +0100 MIME-Version: 1.0 Sent to CCL by: "Oliver T. Hofmann" [o.hofmann(!)tugraz.at] Dear CCL'lers We`re doing Geometry optimizations using the band structure code VASP on a regular basis, and are trying to find ways to speed up the convergence behaivour at low gradients . The most sensible way seems to be to stop the calculation at some point, change SMASS and POTIM, and restart. Does anyone in the community have experience doing this? Does this result in significant reduction of optimization cycles? Or is it not worth the extra work? Does anybody know a more efficent approach? Thank you in advance With kind regards Oliver Hofmann From owner-chemistry@ccl.net Wed Oct 31 14:57:00 2007 From: "errol lewars elewars#,#trentu.ca" To: CCL Subject: CCL:G: Calculation of Proton Affinity Message-Id: <-35539-071031133306-12629-bvUGZXQ54wNX6W7bCBH09w%server.ccl.net> X-Original-From: errol lewars Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Wed, 31 Oct 2007 13:33:47 -0400 MIME-Version: 1.0 Sent to CCL by: errol lewars [elewars%x%trentu.ca] 2007 Oct 31 Hello, Here are some relevant refs [I assume you'll learn basic keywords for methods, basis sets, optimization and frequencies and the significance of these]: 1) B J Smith, L Radom, JACS, 1993, 115, 4885 "Assigning absolute values to PSa: differentiating between competing sites" 2) A K Chandra, A Gourset, JPC, 1996, 100, 11595 "...A critical study" 3) Z B Maksic et al, JPCA, 1997, 101, 7445 "Simple ...method..." 4) Z B Maksic, R Vianello, JPCA, 2002, 106, 419 "...neutral nitrogen basers" 5) J F Sanz et al, J Comp Chem, 1998, 9, 784 "...preferential protonation sites." 6) Definition of proton affinity, and calculations: E. Chamorro et al, JPCA, 2005, 109(44), 10068 7) How to handle the proton room temp enthalpy: E I von Felsobuki, K K Kimura, JPC, 1990, 94, 8041. E. Lewars === Raji Raji raji- -anal.chem.tohoku.ac.jp wrote: >Sent to CCL by: "Raji Raji" [raji(0)anal.chem.tohoku.ac.jp] >Dear CCL Members, >I would like to find out the protonation position in a molecule which contains two heterocyclic nitrogen. I think it can be done by calculating the Proton Affinity. Can you please tell me how to calculate this using G03? Please suggest me the keywords and procedure for the same. Thanks in advance. >Raji.> > > > > From owner-chemistry@ccl.net Wed Oct 31 17:18:00 2007 From: "David C Sullivan davidc.sullivan-,-novartis.com" To: CCL Subject: CCL: AutoShim - Tools to train docking re-scoring on activity Message-Id: <-35540-071031154717-31800-nIjecOVyu92Qw9gYdighwg() server.ccl.net> X-Original-From: "David C Sullivan" Date: Wed, 31 Oct 2007 15:47:13 -0400 Sent to CCL by: "David C Sullivan" [davidc.sullivan],[novartis.com] The AutoShim suite of tools for training docking re-scoring functions on activity data is now downloadable from Daylight Contrib and SourceForge: http://www.daylight.com/download/contrib/autoshim.html http://sourceforge.net/projects/autoshim/ AutoShim regresses docking score (or score components, or multiple scoring functions) supplemented with interaction descriptors (e.g. specific hydrogen bonds made/not-made, occupancy of particular pockets, etc.) against experimental activity data. Non-linearities among interactions are learned using decision trees. Pose selection is performed iteratively using the lowest energy pose by the trained model in an iterative model-building/pose-selection process. AutoShim can boost predictivity for any docking program, although included scripts call DockIt and QXPs Flo+. Magnet (from Daylight) calculates protein-ligand interaction descriptors. Predictive modeling uses the freely available R statistical package. AutoShim was developed by David Sullivan and Eric Martin at Novartis (Emeryville) and validated on ~60 kinase activity datasets. An AutoShim calculation on a public dataset, using various flavors of the learning algorithm, is included in the distribution. Happy docking (and training), David Sullivan davidxyz1972 over-at-that yahoo.com