Re: [gmx-users] Dihedral restraints in 4.6 vs 4.5.X

Thanks - I have filed issue #1194 and assigned it to David. http://redmine.gromacs.org/issues/1194 Let me know if there is anything else that is needed. I could take a shot at myself, and will play around with it, but I don't know the details of the symtab in any detail, unfortunately. 15

[gmx-users] Dihedral restraints in 4.6 vs 4.5.X

Hi I anyone aware of any issue starting simulations with 4.6 (or 4.6.1) from a 4.5.X-tpr file with dihedral restraints? I'm unsuccessful in making them start. To investigate further, I created a small dialanine peptide in vacuum, with a dihedral restraint. Here's the details: Making a 4.5.5

Re: [gmx-users] cmaptypes format

Hi Francesco That number corresponds to a type for each cmap-entry. The idea (iirc, it was some time ago) was that it could be useful to be able to have multiple cmap-types (other grid values, different grid spacing etc) for the same cmap-dihedral (much like different torsions etc...), but

Re: [gmx-users] comparing gmx GB energy with Amber11

0 8 nov 2012 kl. 23:42 skrev Sandeep Somani: Hi Per I tried with single precision gmx as well. No change. Will send you input files soon. Best Sandeep On Thu, Nov 8, 2012 at 3:55 PM, Per Larsson per.lars...@sbc.su.se wrote: Hi Thanks for doing those test. They are all

Re: [gmx-users] comparing gmx GB energy with Amber11

Hi A few things: 1. If I recall correctly, the GB-energy in gromacs is split into two parts, GB-polarization and non-polar solvation. Can you check whether this is the case and if the value you report is the sum of those two terms. 2. Try setting all cut-offs to 0 (infinite cutoffs). That

Re: [gmx-users] comparing gmx GB energy with Amber11

Hi Thanks for doing those test. They are all reassuring, I think. Could you maybe send me your input-files off list and I'll take a look. I suspect the issue is that different radii are being used, as Gromacs does not use the Bondi radii. Cheers /Per 8 nov 2012 kl. 16:35 skrev Sandeep

Re: [gmx-users] MM/PBSA Gromacs

Hi! There is support in gromacs for gb calculations. However, I'm currently investigating some recent reports about unstable simulations (unfolding proteins), so my advice is to use the code with caution. Thanks /Per 30 nov 2011 kl. 20:04 skrev R.S.K.Vijayan biovija...@gmail.com: Dear

Re: [gmx-users] System heating up during MD run? (Implicit solvent)

Hi There has been a number of reports lately about ill-behaving simulations using implicit solvent. I'm currently trying to investigate the cause of this, as such simulations used to work very nice in our hands. In the meantime I would advice to use implicit solvent with caution. Thanks /Per

Re: [gmx-users] Energy Drift

Hi Justin! Do you maybe have an example of such a protein (preferably not too large :-), that I could run some tests on? I'd be interested in seeing if there has been any bugs introduced in the cutoff code that destabilises proteins that way. Thanks /Per 18 okt 2011 kl. 23:41 skrev Justin A.

Re: [gmx-users] Simulation runs on iMac but explodes on cluster

Hi! Could you try running with GMX_NOOPTIMIZEDKERNELS=1, to see if this could potentially be a bug in the sse code (I noticed nb_kernel410_x86_64_sse in your error message. Thanks /Per 13 jul 2011 kl. 09.35 skrev Luke Goodsell: As the subject suggests, I have a simulation that runs

Re: [gmx-users] Questions about GB parameters

should be minor. Of course this needs to be properly documented, and it was very good that some light was shed on this. /Per 6 jul 2011 kl. 04:29 skrev Justin A. Lemkul jalem...@vt.edu: Per Larsson wrote: Hi! I did some digging and think I can clarify at least the first question. Sorry

Re: [gmx-users] Questions about GB parameters

Hi! I did some digging and think I can clarify at least the first question. Sorry for the confusion with regard to the earlier post. What is specified in the gbsa.itp file under the gbr column is indeed vdW-radii. These are used to compute Born radii, as the manual says. The dielectric offset

Re: [gmx-users] Questions about GB parameters

Hi! Hmm.. Let me see if I can shed some more light on this. It's been a while though since I visited the literature here, and also my laptop broke down today, so I need to take of that first before I can check the code! Thanks /Per 14 jun 2011 kl. 20:24 skrev Justin A. Lemkul

Re: [gmx-users] Is it possible to run implicit solvent simulations in parallel?

Hi! If you are running implicit solvent with no cutoffs, ie using the special all-vs-all kernels, then particle decomposition will be used. This exact combination (gb, all-vs-all, dd) is quite tricky to implement, and is not supported at the moment, IIRC. This could be documented better,

Re: [gmx-users] implicit solvent GB force table

Hi! The tabulated function is applied to all interactions, both bonded and non-bonded. The function is initialized in make_gb_table(), and accessed in the non-bonded kernels the same way as other tabulated functions are accessed. See GBtab[nnn] in the kernels for more details. In

Re: [gmx-users] implicit solvent model: formulation of non-polar free energy

Hi! I know about this difference. The only reason for it is that, when I first started with this I used Tinker as a reference implementation, which has the 6-th power in the code. The differences should be small, though. The non-polar part only amounts to a few percent of the total force

Re: [gmx-users] IMPLICIT SOLVENT IN AMBER

Hi! Starting an implicit solvent simulation works just as starting a normal, explicit solvent simulation, except there is no solvent molecules. You should use version 4.5.3 for this though (4.0.5 will not work). Then you specify in the mdp-file implicit_solvent = GBSA and use pdb2gmx, grompp,

Re: [gmx-users] IMPLICIT SOLVENT IN AMBER

is not implemented currently? Or it is not critically? 2011/3/10 Per Larsson per.lars...@sbc.su.se Hi! Yes, except that in point 2, I'm not sure about the effects of explicit ions in an implicit solvent. Do deal with that properly one should use an implicit salt concentration

Re: [gmx-users] RE: g_energy inconsistent results

Hi! When you are computing your zero-step energies, do you then start from the gro-file that you got from em? If so, maybe the energies changes because gro-files have a fixed precision format (with three decimals), while the em-calculations are done using either full single or double

Re: [gmx-users] question about GBSA and opls-aa

Hi! The treatment of 1-2, 1-3 and 1-4 interactions with gbsa differ between the different born radii models, but not between force fields. For the Still model, polarisation energies are computed once for 1-2, 1-3 and 1-4 interactions, and never updated for these. For the HCT/OBC models, all

Re: [gmx-users] Target implementation date for gb_saltconc?

Skickat från min iPhone 18 feb 2011 kl. 21:10 skrev Matthew Zwier mczw...@gmail.com: Dear GROMACS developers and users, Our research group is interested in performing GBSA simulations with GROMACS, but we would need to perform them with a nonzero salt concentration. I was wondering if

Re: [gmx-users] OPLS implicit solvent problems

Hi! Thanks for pointing this out. It is unfortunate that they are missing from the gbsa.itp-file. I will add them, but I do not have time today. If you want to get going, you can add them yourself. The gb-parameters are essentially based on atom hybridization, so for each missing atom you find

Re: [gmx-users] van der Walls radius and implicit solvation

Hi! Currently there is no support in Gromacs for implicit solvation using a united-atom force field. It is definitely possible though, but you need to consult the literature for values for the radii. For united-atoms, I guess the hydrogens you mention are the polar hydrogens? All radii in the

Re: [gmx-users] Re: gb_saltconc in implicit water simulations

of on the same topic, no matter what option I choose for sa_algorithm with Gromacs 4.5.3, it always uses sa_algorithm=Still. Thanks for your help, Rogan On Dec 14, 2010, at 12:13 PM, Per Larsson wrote: Hi! Sorry that I missed this discussion, we recently moved the lab

Re: [gmx-users] Re: gb_saltconc in implicit water simulations

Hi! Sorry that I missed this discussion, we recently moved the lab and there where quite many things to attend to. When we did the implementation of implicit solvent into Gromacs, one of the strong points was to see how fast we could make it, to speed up conformational sampling as much as

Re: [gmx-users] Thermostat for REMD simulations in implicit solvent

Hi! Have never tried remd with implicit solvent, but note that the unit of tau-t in the mdp-file is ps, not ps-1. This means you should set tau-t = 0.0109 rather than 91. Try this and see if the problem goes away! /Per 26 nov 2010 kl. 15:55 skrev César Ávila clav...@gmail.com: Dear all,

Re: [gmx-users] Using Charmm FF with implicit solvent on gromacs

Hi! Currently it is not possible to use GBSW in Gromacs. As you noted, there are three GB-models available (Still,HCT,OBC), that can be run using plain cutoffs (or no cutoffs at all, using the optimized all-vs-all kernels). Currently there is unfortunately no work being done to extend the range

Re: [gmx-users] Role of epsilon_rf in implicit solvation

Hi! It would be nice to have reaction-field for GB as well, but currently this is not the case. The polarisation energy is scaled using (1.0/epsilon_r - 1.0/gb_epsilon_solvent) always, epsilon_rf is never used. Cheers /Per 19 nov 2010 kl. 12.30 skrev Christian Mücksch: So is this

Re: [gmx-users] Re: GB-polarization

Hi! Sorry for the confusion here. Justin is right. In versions before 4.5.2, the non-polar solvation calculation was done regardless of the sa_algorithm setting. In 4.5.2, this has to be specified explicitly using the setting Ace-approximation, to get the complete solvation free energy. Also,

Re: [gmx-users] GB-polarization

Hi! Yes, the GB-polarization energy corresponds to the solvent-solute electrostatics polarization energy. The non-polar part of the solvation energy (the solvent-solvent cavity term and the solute-solvent vdw-term) are named Non. polar solvation (or something like that) in the log-file. /Per

Re: [gmx-users] about sa_surface_tension

Hi! This is unfortunately a bit confusing, but prior to the release of 4.5.2 gromacs did not use the value of the sa_surface_tension that was specified in the mdp-file. It was always set to 2.092, and then hardcode to take a certain value (other than 2.092) depending on the GB-model. 2.25936

Re: [gmx-users] best parameters for energy determination with GBSA

Hi Ehud! Your settings are correct. However, you can also use domain decomposition with GB, and I at least usually bypass the editconf stage, and do pdb2gmx-grompp-mdrun, when you do not want any pbc. /Per 7 okt 2010 kl. 12.29 skrev Ehud Schreiber: Dear GROMACS users, I try to compute

Re: [gmx-users] potential energy in implicit solvent simulations

Hi It seems that the non-polar energy term has gone missing, and does not get included in the print-out. While adding it back again properly I found some other issues that I need to check before committing. The total potential energy should be correct. /Per 16 sep 2010 kl. 17:47 skrev

Re: [gmx-users] About implicit water simulations

Hi! This is not something that is planned for the immediate future, as useful as it might be. However, for small to medium sized systems one can get good performance by using the all-vs-all kernels. Then the problem of poor energy conservation disappears. Cheers /Per Skickat från min

Re: [gmx-users] Generalized Born segfaults with v4.5.1

Hi! I am currently working on another issue with the GB-double precision loops. I'll include this as well. But your pdb-file only contains 1 residue with the name UNK (by openBabel). Could you please send me another pdb-file that reproduces this error off-list, and I'll get to it. Cheers /Per

Re: [gmx-users] grompp_d and mdrun_d issues with GBSA and normal mode analysis

This has been fixed with Berk's commit c06ee471... The error was not related to GB, but rather a combination of domain decomp.+nm+cut-offs. Cheers /Per 6 sep 2010 kl. 11.38 skrev Ehud Schreiber: Dear GROMACS users, I am encountering a couple of issues when trying to perform normal mode

Re: [gmx-users] grompp_d and mdrun_d issues with GBSA and normal mode analysis

Hi! This should work. I checked the code but did not find anything obvious. Could you please file a bugzilla and I'll have a look as soon as possible. Thanks! /Per 6 sep 2010 kl. 11.38 skrev Ehud Schreiber: Dear GROMACS users, I am encountering a couple of issues when trying to perform

Re: [gmx-users] ffamber03 H0 atomtype in 4.5.1

Hi! This seems to be an unfortunate mistake. It should not be commented out. Thank you for reporting this, I will fix it for the next release. Cheers /Per 6 sep 2010 kl. 17.08 skrev William Joseph Allen: Hello Gromacs users, I have recently set up some simulations of a peptide using the

Re: [gmx-users] GB Parameters

replying, please edit your Subject line so it is more specific than Re: Contents of gmx-users digest... Today's Topics: 1. Re: GB Parameters (Per Larsson) 2. Re: Charmm to Gromacs: Polyols force field [Justin] (Bruce D. Ray) 3. Re: Re: Re: [gmx-users] Charmm to Gromacs: Polyols force

Re: [gmx-users] GB Parameters

Hi! Values in columns 1-3 are not currently used in version beta3. They pertain to more elaborate surface area algorithms, the one from Still et al in particular. Column 4 is the Vdw-radii for atom types that are used for computing the Born radii. The dielectric offset is specified in the

Re: [gmx-users] GB Parameters

Also, again, as David said, if you find any errors in the parameters or in the code, please let me know! Cheers /Per 23 aug 2010 kl. 20:09 skrev Per Larsson per.lars...@sbc.su.se: Hi! Values in columns 1-3 are not currently used in version beta3. They pertain to more elaborate surface

Re: [gmx-users] cutoff for implicit solvent in gromacs4.5

Hi, for implicit solvent, people generally use longer cut-offs, as you noted. There can be many reasons for your seg. fault, however. Did you do energy minimization before starting md? Please also test the same system with using a cut-off to see if that works. /Per 8 aug 2010 kl. 22.36

Re: [gmx-users] new beta release - gromacs-4.5-beta2

Hi! Is this with GBSA implicit solvent, and in that case what Born radii model? Could you mail me the tpr off-list, and I'll have a look at it. Cheers /Per 5 aug 2010 kl. 20.26 skrev Elio Cino: Since the charmm force field has some instances with large charge groups (grompp warns you

Re: [gmx-users] Re: gromacs with CMAP

Hi, I do not fully understand what you are trying to do, but currently CMAP is only available for the standard amino acid residues present in the rtp-file for the Charmm-forcefield, and the values for the grid are specified in the cmap.itp-file. Do you use something else? /Per 29 jun 2010

Re: [gmx-users] Charmm without CMAP

Hi! You can choose cmap/no cmap when you run pdb2gmx. /Per Skickat från min iPhone 22 jun 2010 kl. 19:24 skrev Sai Pooja saipo...@gmail.com: Hi, I am using gromacs from the git repository which I downloaded 2 days back. Charmm has been implemented in this version. Is it possible to use

Re: [gmx-users] protein structure prediction

23 feb 2009 kl. 12.55 skrev Mark Abraham: José Carlos Calvo Tudela wrote: can I use the GROMACS software to get the energy of a protein conformation? Yes, but your question is not well-formed. Only energy differences can have meaning. Up to now I am using TINKER library (analyze

Re: [gmx-users] Error in pdb2gmx.....Atom CE2 not found in residue PHE270 while adding hydrogens

If atoms are missing in the sidechains of your protein, you can try a program like scwrl3 to rebuild all or some of the sidechains. This will give you complete sidechains in sort of reasonable positions. Cheers /Per 12 aug 2008 kl. 15.04 skrev vivek sharma: Hi David, Thanx a lot again.

Re: [gmx-users] (no subject)

Check the information on the gromacs wiki about this: http://wiki.gromacs.org/index.php/Errors#T-Coupling_group_XXX_has_fewer_than_10.25_of_the_atoms Cheers /Per 17 jul 2008 kl. 14.38 skrev [EMAIL PROTECTED]: Hi all I have installed gromacs 3.3.3 in one of my 32 machine fedora core 2. so

Re: [gmx-users] regarding rmsd !!

Check your input/output! In this case it seems you are missing/forgetting the -s flag to specify the name of the tpr-file. /Per 17 jun 2008 kl. 17.07 skrev Anamika Awasthi: Dear All, My protein is simulating for 20 ns and 16 ns has already over, but I want to analyze RMSD

Re: [gmx-users] Implicit Solvent Simulations

luck we should be able to commit this to cvs in the next month or so :-) Cheers /Per Larsson, PhD student in the Lindahl group 22 maj 2008 kl. 16.13 skrev Una Bjarnadottir: To the Gromacs user community, As of April 2007 it was not possible to do implicit solvent MDs with Gromacs (http

Re: [gmx-users] Help for GNUPlot

Hello! Please note that xmgrace and gunplot are two different programs and as such work in different ways, for example in terms of what types of input and output they expect. The xvg-files that Gromacs produces are indended for xmgrace, and will not work out-of-the-box with gnuplot.

Re: [gmx-users] results of energy minimization

Hello! Check out http://wiki.gromacs.org/index.php/Errors#Stepsize_too_small.2C_or_no_change_in_energy._Converged_to_machine_precision.2C_but_not_to_the_requested_precision Cheers /Per 12 mar 2008 kl. 11.58 skrev s lal badshah: Dear gromacs experts, I have run the energy minimization of a

Re: [gmx-users] Step size too small

Hello! Check out http://wiki.gromacs.org/index.php/Errors#Stepsize_too_small. 2C_or_no_change_in_energy._Converged_to_machine_precision. 2C_but_not_to_the_requested_precision Cheers /Per 17 aug 2007 kl. 18.28 skrev Sheyore Omovie: Dear gmx-users, I have 2 molecules in a box, as usual

Re: [gmx-users] stepsize too small

Hello! There are many discussions on the maillist about this... In short, if all you want to do is normal md, it should be ok to just proceed with that. Just check that your potential energy is negative and everything should fine. Otherwise, recompiling Gromacs to use double precision

Re: [gmx-users] problem with writing energies

Hello! Are you out of disk-space? Cheers /Per 2 aug 2007 kl. 08.49 skrev aneesh chandran: Hello Users I am using gromacs 3.3 with g43a1 force field. I am running a MD without any position restraints. After around 770 ps system stoped running with an error could not write energies.

Re: [gmx-users] error in pdb2gmx step

or send it to [EMAIL PROTECTED] Can't post? Read http://www.gromacs.org/mailing_lists/users.php -- Per Larsson PhD Student Stockholm Bioinformatics Center +46 (0) 8 55378577 +46 (0) 733 461467 [EMAIL PROTECTED] www.sbc.su.se