From owner-chemistry@ccl.net Mon Dec 5 03:47:00 2011 From: "Carlos T Nieto eneas,+,usal.es" To: CCL Subject: CCL: Getting ride of additional imaginary frecuencies Message-Id: <-45981-111205034539-28622-oMGwKsedgn6NrU9gNhhnkA%a%server.ccl.net> X-Original-From: "Carlos T Nieto" Date: Mon, 5 Dec 2011 03:45:37 -0500 Sent to CCL by: "Carlos T Nieto" [eneas ~ usal.es] Hi, everybody One of the typical problems in a optimization is the appearrance of additional imaginary frecuencies in optimization or transition state minimizations. Im an user of Jaguar. In many frequency calculations (DFT/B3LYP) i obtain small negative frequencies, e.g. < -30 cm-1.It's very frequently i obtained it in: - Solvent calculations - Transition states when freezing atoms involved in it. It's to optimize the different substituents. When this frequencies are associated with certain type of easy identifiable displacement, i modify the geometry along the suspected vibration and reoptimize. This works well. Nevertheless, the most difficult cases is when the displacement is difficult (some type of small apparent rotation where almost all atoms participate) i don't know how to disturb it. It often appears when i freeze atoms, let's say to block the atoms involved in a TS to add substituents and reoptimize or in solvent calculations. With this type of frequencies i've tried several ways, but inefficiently: - Disturbing bulky groups and reoptimizing. - Some authors suggest it's a problem of the grid build up of the dft process. I tried to change to fine or ultrafine type grids without success. - Other opinion is that is a mathematical artefact and such frequencies could be ignored. I've cheked a few publications and it's indicated. - In Transition state searchs, the small negative frequency is the eigenvector n 2. Trying to force to search along eigenvector 1 or lowest eigenvector doesn't works again. So, What else can i do? I'm not sure if i'm doing something wrong... Is there a way to getting rid of it? If not, May i accept the result and ignore the small imaginary frequency? Other authors said that, accepting this, it's not accurate taking the thermodinamic data from the frequency calculations... Will be reportable my results ignoring this additional frecuencies? Shall I indicate them? Ill appreciate all the ideas from you. From owner-chemistry@ccl.net Mon Dec 5 04:22:00 2011 From: "Tobias Schwabe tobba]^[uni-muenster.de" To: CCL Subject: CCL:G: wb97XD in g09 Message-Id: <-45982-111205031550-29612-1+HrRAfPm8GP427ipzO9ow^_^server.ccl.net> X-Original-From: Tobias Schwabe Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Mon, 05 Dec 2011 09:15:13 +0100 MIME-Version: 1.0 Sent to CCL by: Tobias Schwabe [tobba!^!uni-muenster.de] Hi Quartarolo, On 12/01/2011 11:40 AM, quartarolo|a|unical.it wrote: > Is there a way to define this parameter manually in the input file and > eventually where to search for vdW radius of lutetium. I don't know if there is any way to specify a certain parameter for DFT-D in G09. One idea would be to override the addition of the dispersion correction to wB97XD at all (which, if I remember right, is NOT just wB97X): you can do so by an IOps overlay (3/124), see http://www.gaussian.com/g_tech/g_iops/ov3.htm Then, you have the "pure" DFT energy. Afterwards you can use Grimme's DFT-D3 program - available from http://toc.uni-muenster.de/DFTD3/getd3.html to add the missing dispersion term. I think, it has the parameters you need. Be aware, that the different flavors of dispersion correction might not be quite the same (I haven't checked), and by this you kind of create your own new model chemistry. You might want to test this on some reference data before applying it to an unknown problem. Certainly, you cannot compare energies obtained like this with "normal" wB97XD energies. Of course, this whole set up is only working for single point energies and is rather inelegant. Maybe someone has another idea? Regards, Tobias From owner-chemistry@ccl.net Mon Dec 5 05:08:00 2011 From: "Pierre Archirel pierre.archirel[-]u-psud.fr" To: CCL Subject: CCL:G: imaginary frequencies Message-Id: <-45983-111205050555-18231-tMecKnlm0BeiKN+6dG/9qA#,#server.ccl.net> X-Original-From: "Pierre Archirel" Date: Mon, 5 Dec 2011 05:05:53 -0500 Sent to CCL by: "Pierre Archirel" [pierre.archirel/a\u-psud.fr] Dear Carlos, The only true way for dropping imaginary frequencies is to disturb your molecule in the eigen direction of the nuclear hessian corresponding to the imaginary frequency. You just have to write a little code for making the displacement. I hope Jaguar provides you with the eigen vectors together with the frequencies, as gaussian does... If you have this vector available, then: 1- try both directions, with positive and negative displacements 2- try several values of the displacement. In gaussian 03 the reaction field was modeled with point charges, which provided the potential surface with such wavelets that .01 A displacements were necessary to drop imaginary frequencies. The gaussian charges of gaussian 09 have been a qualitative improvement in this respect. Is Jaguar using point charges? Pierre From owner-chemistry@ccl.net Mon Dec 5 08:48:01 2011 From: "Rene Thomsen rt ~ molegro.com" To: CCL Subject: CCL: Molegro releases Molegro Virtual Docker 5.0 Message-Id: <-45984-111205081511-17611-fcqI64xEjq7PoMeoW9/D9Q()server.ccl.net> X-Original-From: "Rene Thomsen" Date: Mon, 5 Dec 2011 08:15:09 -0500 Sent to CCL by: "Rene Thomsen" [rt ~~ molegro.com] Aarhus, Denmark, December 5th, 2011 - Molegro is pleased to announce a new major release of Molegro Virtual Docker, an integrated platform for computational drug design available for Windows, Linux, and Mac OS X. Molegro Virtual Docker offers high-quality protein-ligand docking based on novel optimization techniques combined with a user interface experience focusing on usability and productivity. Major new features in version 5.0: * GPU-accelerated docking on CUDA supported hardware making it possible to screen drug-like compounds up to 30 times faster than using conventional CPU-based methods. The GPU implementation builds upon and extends the research described in the paper "GPU-Accelerated High-Accuracy Molecular Docking using Guided Differential Evolution" (http://dl.acm.org/citation.cfm?id=2001576.2001818). * The new 2D Ligand Map provides an easy way to inspect and visualize protein-ligand interactions. For more information, or to download a trial version, please visit our company website at: http://www.molegro.com or contact: Rene Thomsen, CEO Molegro C. F. Moellers Alle 8, Bldg. 1110 DK-8000 Aarhus Denmark E-mail: rt-,-molegro.com Phone: (+45) 8715 5571 About Molegro Molegro is a Danish company founded in 2005. Our company develops high-performance drug discovery solutions leading to a faster drug development process. Our goal is to provide scientifically superior products focusing on both state-of-the-art algorithms and an intuitive graphical user interface experience. From owner-chemistry@ccl.net Mon Dec 5 12:31:00 2011 From: "Arne Dieckmann adieckma+*+googlemail.com" To: CCL Subject: CCL:G: Getting ride of additional imaginary frecuencies Message-Id: <-45985-111205122758-31625-Ytgnc2ok7/q3WsmzsFq7Ug~!~server.ccl.net> X-Original-From: Arne Dieckmann Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=us-ascii Date: Mon, 5 Dec 2011 09:27:30 -0800 Mime-Version: 1.0 (1.0) Sent to CCL by: Arne Dieckmann [adieckma^googlemail.com] Those frquencies are just an artefact as long as they are < 50 cm^-1. Have a look at the Gaussian Reference, it's explained under the "freq" keyword if I remember correctly. Cheers, Arne On Dec 5, 2011, at 12:45 AM, "Carlos T Nieto eneas,+,usal.es" wrote: > > Sent to CCL by: "Carlos T Nieto" [eneas ~ usal.es] > Hi, everybody > > One of the typical problems in a optimization is the appearrance of additional > imaginary frecuencies in optimization or transition state minimizations. > > Im an user of Jaguar. In many frequency calculations (DFT/B3LYP) i obtain small negative > frequencies, e.g. < > -30 cm-1.It's very frequently i obtained it in: > > - Solvent calculations > - Transition states when freezing atoms involved in it. It's to > optimize the different substituents. > > When this frequencies are associated with certain type of easy > identifiable displacement, i modify the geometry along the suspected > vibration and reoptimize. This works well. Nevertheless, the most > difficult cases is when the displacement is difficult (some type of small > apparent rotation where almost all atoms participate) i don't know how to disturb > it. It often appears when i freeze atoms, let's say to block the atoms involved in a > TS to add substituents and reoptimize or in solvent calculations. > > With this type of frequencies i've tried several ways, but inefficiently: > > - Disturbing bulky groups and reoptimizing. > - Some authors suggest it's a problem of the grid build up of the dft > process. I tried to change to fine or ultrafine type grids without success. > - Other opinion is that is a mathematical artefact and such frequencies > could be ignored. I've cheked a few publications and it's indicated. > - In Transition state searchs, the small negative frequency is the > eigenvector n 2. Trying to force to search along eigenvector 1 or lowest > eigenvector doesn't works again. > > So, What else can i do? I'm not sure if i'm doing something wrong... Is > there a way to getting rid of it? If not, May i accept the result and > ignore the small imaginary frequency? Other authors said that, accepting > this, it's not accurate taking the thermodinamic data from the frequency > calculations... > > Will be reportable my results ignoring this additional frecuencies? Shall I indicate > them? > > Ill appreciate all the ideas from you.> >