Re: [gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Mark Abraham
Hi, You're running 2016.x which had a bug, not the 2018.x you thought you were using. Use GMXRC or your cluster's modules to select the version you want to use in the terminal or script that you want to use. Mark On Mon, 26 Aug 2019 at 20:34, Navneet Kumar Singh wrote: > What kind of error is

Re: [gmx-users] (no subject)

2019-08-26 Thread Justin Lemkul
On 8/26/19 2:33 PM, Navneet Kumar Singh wrote: I can't understand meaning of this "For all other GROMACS versions, you will have to manually edit the topology to use "3fad" construction and appropriate atom numbers." If I am using version other than the Gromacs2016.x. Can I get example of

Re: [gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Navneet Kumar Singh
This is make check reuslt 100% tests passed, 0 tests failed out of 27 Label Time Summary: GTest = 1.24 sec*proc (18 tests) IntegrationTest = 5.93 sec*proc (2 tests) MpiIntegrationTest

Re: [gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Mark Abraham
Hi, If you have a reason to have to use 2016.x, then please get the latest version (which is always advisable when starting new work) because this issue is fixed there. You also don't have to run energy minimization on GPUs, so you could just append -nb cpu to your mdrun command line to avoid the

Re: [gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Navneet Kumar Singh
Currently I am using only gromacs16.5. As this have *"PLEASE NOTE* that the current versions do support lone pair construction on halogens, however the current construction is only compatible with GROMACS-2016.x and by using gmx grompp -maxwarn 1 to override the warning about lone pair

Re: [gmx-users] (no subject)

2019-08-26 Thread Navneet Kumar Singh
I have seen this in Gromacs manual. [ virtual_sites2 ] ; Site fromfunct a 5 1 2 1 0.7439756 for type 3 like this: [ virtual_sites3 ] ; Site from funct a b 5

Re: [gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Navneet Kumar Singh
Thank You Sir. Again installed gromacs2016.6 (final version from gromacs2016.x). Throwing same error(HtoD cudaMemcpyAsync failed: invalid argument) . Its running on CPU, but not on GPU. Work of 3 days will take 15 days. Already completed simulation for other complexes. This complex creating

Re: [gmx-users] (no subject)

2019-08-26 Thread Justin Lemkul
On 8/26/19 1:14 PM, Navneet Kumar Singh wrote: Do I have to add vsite3 information from unk.itp to system.top files? That directive belongs in the topology to which it corresponds. If it is in unk.itp, then it is already in system.top. You don't need to include anything else. -Justin

Re: [gmx-users] Question about Gromacs

2019-08-26 Thread David van der Spoel
Den 2019-08-26 kl. 20:53, skrev Najla Hosseini: Dear David, Hope you are doing well. I am Gromacs user and I need to change the partial charge of molecules in force field or itp file as a function of distance during the run in Gromacs. Is it possible? How I should do that? Please pose your

Re: [gmx-users] CHOL still not recognised

2019-08-26 Thread Mark Abraham
Hi, You should follow the error message instructions "... they should not have the same chain ID as the adjacent protein chain" which you know is that X. Make the protein have a different chain ID from the rest. Mark On Mon, 26 Aug 2019 at 19:27, Ayesha Fatima wrote: > Dear Justin, > Thank

Re: [gmx-users] (no subject)

2019-08-26 Thread Navneet Kumar Singh
I can't understand meaning of this "For all other GROMACS versions, you will have to manually edit the topology to use "3fad" construction and appropriate atom numbers." If I am using version other than the Gromacs2016.x. Can I get example of any topology file where these kind of construction've

[gmx-users] CHOL still not recognised

2019-08-26 Thread Ayesha Fatima
Dear Justin, Thank you for your earlier response. I opened the file in vmd and it returned me with a 3 letter residue name and an X identifier for the whole molecule. CRYST10.0000.0000.000 90.00 90.00 90.00 P 1 1 ATOM 1 C3 CHL X 1 -1.678 2.702 17.328

[gmx-users] HtoD cudaMemcpyAsync failed: invalid argument

2019-08-26 Thread Navneet Kumar Singh
What kind of error is this. Previously Gromacs 2018.4 version was running fine using GPU. But Now this error I am getting. _ gmx mdrun -v -deffnm em :-) GROMACS - gmx mdrun, 2016.5 (-:

[gmx-users] Residue XXX not found in residue topology database

2019-08-26 Thread Neena Susan Eappen
Hello gromacs users, I saw the following error for a modified residue I added though I edited all necessary files including PDB. I also read the gromacs documentation on this error. Residue XXX not found in residue topology database. Any hints on what might be happening? Many thanks, Neena

Re: [gmx-users] Residue XXX not found in residue topology database

2019-08-26 Thread Justin Lemkul
On 8/26/19 8:27 PM, Neena Susan Eappen wrote: Hello gromacs users, I saw the following error for a modified residue I added though I edited all necessary files including PDB. I also read the gromacs documentation on this error. Residue XXX not found in residue topology database. Any

[gmx-users] [HELP] Residue 'NI' not found in residue topology database

2019-08-26 Thread Edjan Silva
Dear users, I am trying to perform a simulation with a protein which contains two nickel atoms in the active site. When using the pdb2gmx command the following error appears: 'NI' not found in residue topology database I have edited the ions.itp file in the force field directory used (opls)

Re: [gmx-users] Gromacs 2019.3 compilation with GPU support

2019-08-26 Thread Mark Abraham
Hi, All versions of icc requires a standard library from an installation of gcc. There are various dependencies between them, and your system admins should have an idea which one is known to work well in your case. If you need to help the GROMACS build find the right one, do check out the GROMACS

[gmx-users] gromacs is not recognising opls ff

2019-08-26 Thread Ayesha Fatima
Dear All, I have come across another issue When i want to use opls itp for cholesterol, it gives me this error " Fatal error: Residue 'OL' not found in residue topology database" It does not take CHOL as the residue name as given below [ atoms ] ; nr type resnr residue atom cgnr

Re: [gmx-users] wham analysis

2019-08-26 Thread John Whittaker
Hi, > >  Dear all,I used Justin's tutorial(Tutorial 3: Umbrella Sampling: GROMACS > Tutorial ) for my file which is protein-ligand complex. > The pulling force was in Y direction. when Umbrella sampling finished, > "Wham" couldn't analysis the data because wham is in z direction.what > should I

[gmx-users] wham analysis

2019-08-26 Thread Negar Parvizi
 Dear all,I used Justin's tutorial(Tutorial 3: Umbrella Sampling: GROMACS Tutorial ) for my file which is protein-ligand complex. The pulling force was in Y direction. when Umbrella sampling finished, "Wham" couldn't analysis the data because wham is in z direction.what should I do now for

[gmx-users] How to restrain small molecules in a region where the small molecules can move freely?

2019-08-26 Thread wtzou
Dear all, I want to restrain some water in a big cluster where the water can move freely in the cluster but they cannot move out of the cluster. How can I do to implement the restrain? Thank you very much! Sincerely, Wentian -- Institute of Theoretical and Computational

Re: [gmx-users] How to restrain small molecules in a region where the small molecules can move freely?

2019-08-26 Thread Justin Lemkul
On 8/26/19 5:41 AM, wtzou wrote: Dear all, I want to restrain some water in a big cluster where the water can move freely in the cluster but they cannot move out of the cluster. How can I do to implement the restrain? Check the manual for flat-bottom restraints. -Justin --

Re: [gmx-users] gromacs is not recognising opls ff

2019-08-26 Thread Justin Lemkul
On 8/26/19 7:04 AM, Ayesha Fatima wrote: Dear All, I have come across another issue When i want to use opls itp for cholesterol, it gives me this error " Fatal error: Residue 'OL' not found in residue topology database" It does not take CHOL as the residue name as given below That suggests

[gmx-users] Gromacs 2019.3 compilation with GPU support

2019-08-26 Thread Prithwish Nandi
Hi, I am trying to compile Gromacs-2019.3 at our HPC cluster. I successfully compiled the single and double precision versions, but it’s producing error for GPU support. (The error message is pasted below) I am using Intel/2018 update 4 and CUDA/10.0. The base gcc version is 4.8.5. Using MKL

Re: [gmx-users] wham analysis

2019-08-26 Thread Jochen Hub
Hi, you can also use the pullf output for WHAM (option -if), this may be easier. Cheers, Jochen Am 26.08.19 um 12:59 schrieb Negar Parvizi:  Dear all,I used Justin's tutorial(Tutorial 3: Umbrella Sampling: GROMACS Tutorial ) for my file which is protein-ligand complex. The pulling force

Re: [gmx-users] [HELP] Residue 'NI' not found in residue topology database

2019-08-26 Thread Justin Lemkul
On 8/26/19 12:04 PM, Edjan Silva wrote: Dear users, I am trying to perform a simulation with a protein which contains two nickel atoms in the active site. When using the pdb2gmx command the following error appears: 'NI' not found in residue topology database I have edited the ions.itp

Re: [gmx-users] (no subject)

2019-08-26 Thread Navneet Kumar Singh
Do I have to add vsite3 information from unk.itp to system.top files? -maxwarn 1 flag I checked in gromacs 16 version and it's running there. On Mon, 26 Aug 2019, 00:58 Justin Lemkul, wrote: > > > On 8/25/19 2:17 PM, Navneet Kumar Singh wrote: > > Yeah! I have read that. > > > > uses -maxwarn