[gmx-users] Simulation on multi nodes performance

Hello GMX users I'm trying to do simulations on multi nodes so I used 8 nodes (128 processors). AS I see there is no difference between 1 node and 8 nodes. I used these commands: --- load apps/gromacs-2016 load apps/openmpi-2.1.1 gmx grompp -f EMZ.mdp -c BOX.gro -p TOP.top -o EMZ.tpr

Re: [gmx-users] pH dependent simulation

pKa of amino acid residue is a constant quantity. Depending on the pH you can protonate and deprotonate amino acid according to the pKa of particular residue. Virus-free. www.avast.com

[gmx-users] pH dependent simulation

Hello Everyone I am trying to simulate a protein at two pH (7.4 and 9.0) using gromacs 5.1.4. The protein has no missing residues. Optimal pH range is 6.0 to 9.0 and pI of the protein is 11.4. It has 9 negatively and 17 positively charged residues. However upon predicting the pKa values using

Re: [gmx-users] cmap

On 11/26/17 6:33 PM, MD wrote: Hi Justin, I added the CSO to residuetypes.dat in my local computer and it worked now. I then tried to copy the modified "residuetypes.dat" to the ff folder on the server computer, but the server computer gromacs wouldn't give option to choose which

Re: [gmx-users] cmap

Hi Justin, I added the CSO to residuetypes.dat in my local computer and it worked now. I then tried to copy the modified "residuetypes.dat" to the ff folder on the server computer, but the server computer gromacs wouldn't give option to choose which residuetypes.dat to use. Do you know how to let

Re: [gmx-users] .cpt file

On 11/26/17 6:20 PM, rose rahmani wrote: Hi; Why .cpt files are not as much important as other files (since grompp works without it)? Can you please provide some context for this question? I'd call a checkpoint file essential, as it is the only true way to get an exact continuation of a

[gmx-users] .cpt file

Hi; Why .cpt files are not as much important as other files (since grompp works without it)? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] cmap

On 11/26/17 6:06 PM, MD wrote: Hi, I had to repost this cause of the size of the previous thread. I had another error after grompp, "ERROR 1 [file topol_Protein_chain_E.itp, line 52165]: Unknown cmap torsion between atoms 3022 3024 3026 3046 3049" The 3022 3024 3026 3046 3049 in the itp

[gmx-users] cmap

Hi, I had to repost this cause of the size of the previous thread. I had another error after grompp, "ERROR 1 [file topol_Protein_chain_E.itp, line 52165]: Unknown cmap torsion between atoms 3022 3024 3026 3046 3049" The 3022 3024 3026 3046 3049 in the itp files referred to the backbone

Re: [gmx-users] umbrella sampling

On Mon, Nov 27, 2017 at 1:55 AM, Alex wrote: > Hi, > > The answer is clearly in the error, just read it please: > 2.848793 > 0.49 * (4/2) where 4 is the box length. > > You can either keep the “pull_coord1_geometry= distance” and increase > the box length or change

Re: [gmx-users] umbrella sampling

Hi, The answer is clearly in the error, just read it please: 2.848793 > 0.49 * (4/2) where 4 is the box length. You can either keep the “pull_coord1_geometry= distance” and increase the box length or change the “pull_coord1_geometry” to “= direction-periodic“. I prefer the former though.

[gmx-users] umbrella sampling

Hi; This is md_pull.mdp integrator = md dt = 0.001 nsteps = 200 nstxout = 0 nstvout = 0 nstfout = 0 nstlog = 500 nstenergy= 1000 nstxtcout

Re: [gmx-users] Build thin film.

You can do whatever you want, and there are many tools to help you do that. Couple of options are using gmx insert-molecules (constrain the box size to make the film, then enlarge afterwards to insert other molecules), gmx genconf (just stack as required, some initial care with runs will be

Re: [gmx-users] gmx angle memory consumption

Matthias, GROMACS only writes to a file when the amount of data reaches a certain threshold, prior to that it is just stored in memory. I don't know what that chunk size is. It may be the manner in which the analysis script stores the trajectory data in memory? Hopefully a developer will be

Re: [gmx-users] virtual site connected to dummy massive site

Hi, Your linear version has two rotation axes (and thus DOF) because it has mass distribution. Your point version has no mass distribution. Mark On Sun, Nov 26, 2017 at 7:52 PM Faezeh Pousaneh wrote: > Please note that the carbon is a virtual site, so has not mass. So the

Re: [gmx-users] virtual site connected to dummy massive site

Please note that the carbon is a virtual site, so has not mass. So the DOF does not influence of its configurations on average. Besides the angle of M--C--M is always 180. Mark, I compare the results of average distributions not a snapshot. Justin, the only LJ are from atom C with next

[gmx-users] Extending multidir simulations

Dear gmx community, I would like to extend a simulation I ran using the mdrun -multidir option. Ideally, it would be good to extend .cpt files in all subdirectories simultaneously to make efficient use of a GPU node. I looked here:

Re: [gmx-users] version 16 of check point file

On Sun, Nov 26, 2017 at 2:05 PM rose rahmani wrote: > Hello; > > Which version of GROMACS use version 16 of check point file? > 5.0 and all since. > Is version 4.5.x of GROMACS use version 12 of check point file? > Yes Mark > Best Regards > Rose > -- > Gromacs Users

Re: [gmx-users] Gromacs simulation with PEG

FYI, starting from GROMOS 53a6 parameters, we have refined them against the free energy of solvation of dimethoxyethane in water (and cyclohexane) (DOI: 10.1021/ct300245h). These new parameters are now part of the new 2016H66 force field (DOI: 10.1021/acs.jctc.6b00187). Best, Patrick Le

Re: [gmx-users] virtual site connected to dummy massive site

Hi, On Sun, Nov 26, 2017 at 6:35 PM Faezeh Pousaneh wrote: > Hi Mark, sorry I am not convinced yet. I hope you did not misunderstand my > question, I have no charges in systems on any atoms. So the extra DOF do > not effect on potential. > Charge has nothing to do with it.

Re: [gmx-users] Umbrella Sampling

On 11/26/17 1:08 PM, Roja wrote: Would you please tell me what is the difference between using umbrella in 4.5.x version and 5.1.4 version?are they different in ACCURACY of sampling? Probably. There have been thousands of bug fixes and features added since 4.5.4, which was released 6 years

Re: [gmx-users] Umbrella Sampling

Would you please tell me what is the difference between using umbrella in 4.5.x version and 5.1.4 version?are they different in ACCURACY of sampling? Sent from my iPhone > On Nov 26, 2017, at 21:14, Justin Lemkul wrote: > > > >> On 11/26/17 12:40 PM, rose rahmani wrote: >>

Re: [gmx-users] Umbrella Sampling

On 11/26/17 12:40 PM, rose rahmani wrote: Hello; what re pull_ncoords , pull_coord1_type , pull_coord1_groups (in V5.1.4) called in version 4.5.4? Why use deprecated software? We don't officially support the 4.5.x series any more. Use the newest version. Some things don't have exact

Re: [gmx-users] virtual site connected to dummy massive site

On 11/26/17 12:34 PM, Faezeh Pousaneh wrote: Hi Mark, sorry I am not convinced yet. I hope you did not misunderstand my question, I have no charges in systems on any atoms. So the extra DOF do not effect on potential. Maybe I (and others) are missing something here. We understand there are

[gmx-users] Umbrella Sampling

Hello; what re pull_ncoords , pull_coord1_type , pull_coord1_groups (in V5.1.4) called in version 4.5.4? Thank you so much Rose -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read

Re: [gmx-users] virtual site connected to dummy massive site

Hi Mark, sorry I am not convinced yet. I hope you did not misunderstand my question, I have no charges in systems on any atoms. So the extra DOF do not effect on potential. Best regards On Fri, Nov 24, 2017 at 11:03 AM, Mark Abraham wrote: > Hi, > > Potential energy

Re: [gmx-users] my modified residue falls apart after MD

On 11/26/17 11:13 AM, MD wrote: ​Thank you Justin. At the mean time I had another error after grompp, "ERROR 1 [file topol_Protein_chain_E.itp, line 52165]: Unknown cmap torsion between atoms 3022 3024 3026 3046 3049" the 3022 3024 3026 3046 3049 in the itp files referred to the backbone

Re: [gmx-users] my modified residue falls apart after MD

​Thank you Justin. At the mean time I had another error after grompp, "ERROR 1 [file topol_Protein_chain_E.itp, line 52165]: Unknown cmap torsion between atoms 3022 3024 3026 3046 3049" the 3022 3024 3026 3046 3049 in the itp files referred to the backbone (C-N-CA-C-N) of three connected

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Dear Alex, But g-x2top doesn't work till we define all atom types in .n2t file. Thanks On 26-Nov-2017 9:34 PM, "Alexandr Nasedkin" wrote: Hi, You will need to have both ffbonded.itp and ffnonbonded.itp . atomname2type.n2t is needed for g_x2top to make it work

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Hi, You will need to have both ffbonded.itp and ffnonbonded.itp . atomname2type.n2t is needed for g_x2top to make it work correctly. Good luck. -Alex On 26/11/2017 10:52, Rakesh Pant wrote: Dear, Alex Does it mean, we do not need to define all the atomtypes in the .nt2 file. It is enough

Re: [gmx-users] my modified residue falls apart after MD

On 11/26/17 9:32 AM, MD wrote: Thank you Mark. Do you think if the atom types of HE, OE and SG are correct in my merged.rtp? These are the three atoms which have name changing cause of the inserted sulfur. I modified their atom types as well. I need to make sure the type names are correct

Re: [gmx-users] my modified residue falls apart after MD

Thank you Mark. Do you think if the atom types of HE, OE and SG are correct in my merged.rtp? These are the three atoms which have name changing cause of the inserted sulfur. I modified their atom types as well. I need to make sure the type names are correct before I add the bond types in

[gmx-users] version 16 of check point file

Hello; Which version of GROMACS use version 16 of check point file? Is version 4.5.x of GROMACS use version 12 of check point file? Best Regards Rose -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! *

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Dear, Alex Does it mean, we do not need to define all the atomtypes in the .nt2 file. It is enough to define all types in bonded and non-bonded itp. Thanks, Rakesh Sent with Mailtrack

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Edit the content of you ffnonbonded.itp file approximately this way: [ atomtypes ] ;name    at.num    mass        charge    ptype    sigma epsilon C    14    12.00    1.08    A    0.44000        0.63      ; OCE    8    15.999400    -0.70    A    0.35400        0.159705  ; C-O-C Bridging Oxygen;

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Dear Alex, If I use O atoms to define the respective functional groups, then how to include the C atoms present in the molecule. Thanks, Rakesh Sent with Mailtrack On Sun, Nov

Re: [gmx-users] Regarding use of g_x2top utility in Gromacs

Hi, As I understood the problem, you need to distinguish between surface and buried carbons? You could define two atomtypes that differ in name and use it to process with x2top. Below is an example of *.n2t lines. You would need to change the charges and bondlengthes. ... O    OCH