Re: [gmx-users] add Mn in itp file
Hi, have look here in the supp mat. of this paper: http://onlinelibrary.wiley.com/doi/10.1002/anie.201202032/abstract Mn2+ for OPLS ff in case you needed it best and On 22/10/2014 16:13, Justin Lemkul wrote: On 10/22/14 5:09 AM, yaser wrote: hi i tried add Mn in .itp file but its need some information such as sigma and epsilon and i search for find this but i cant find this parameters .anyone knows any things about how can i add Mn in .itp file and if you know any tutorials articles about add ions . Parameters compatible with the AMBER force fields (and the associated reference) can be found at http://www.pharmacy.manchester.ac.uk/bryce/amber. -Justin -- --- Andrea Spitaleri PhD Principal Investigator AIRC D3 - Drug Discovery Development Istituto Italiano di Tecnologia Via Morego, 30 16163 Genova cell: +39 3485188790 http://www.iit.it/en/d3-people/andrea-spitaleri.html ORCID: http://orcid.org/-0003-3012-3557 -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Implementation of SHAKE
Hi Mark, There was something in the matrix calculations that was not totally clear to me. That is why I started to think in a different approach. Now everything is fine. Thank you very much, Mario Fernández-Pendás El 23/10/2014 01:15, Mark Abraham mark.j.abra...@gmail.com escribió: Hi, No, it's the original, as cited in the manual. Mark On Wed, Oct 22, 2014 at 6:06 PM, Mario Fernández Pendás mariof...@gmail.com wrote: Dear all, Am I right if I say that the SHAKE algorithm is implemented in GROMACS following the ideas presented in A fast SHAKE algorithm to solve distance constraint equations for small molecules in molecular dynamics simulations by Kräutler,van Gunsteren and Hünenberger? http://onlinelibrary.wiley.com/doi/10.1002/1096-987X%2820010415%2922:5%3C501::AID-JCC1021%3E3.0.CO;2-V/abstract Thank you very much. Cheers, Mario Fernández-Pendás -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Accelerated MD
Hi Doug, I recently read a good review on ABF http://pubs.acs.org/doi/abs/10.1021/jp506633n. It's focused on ABF but there's a section which compares ABF to other techniques. While I'm at it, are there any plans from gromacs devs to implement ABF in gromacs? Best, Patrick Le 20/10/2014 22:33, Douglas Houston a écrit : Hi all, Can anyone recommend a book or (preferably) review article summarising the various accelerated MD methods commonly used in protein/peptide simulation these days? e.g. ABF, PLUMED, REMD, aMD etc. etc. I can read about each individually but I could use a comprehensive but concise comparison of their pros/cons and most popular applications. cheers, Doug _ Dr. Douglas R. Houston Lecturer Institute of Structural and Molecular Biology Room 3.23, Michael Swann Building King's Buildings University of Edinburgh Edinburgh, EH9 3JR, UK Tel. 0131 650 7358 http://tinyurl.com/douglasrhouston -- ___ Patrick FUCHS Dynamique des membranes et trafic intracellulaire Institut Jacques Monod, CNRS UMR 7592, Université Paris Diderot Bâtiment Buffon, 15 rue Hélène Brion, 75013 Paris Tel : +33 (0)1 57 27 80 05 - Fax : +33 (0)1 57 27 81 35 E-mail address: patrick.fu...@univ-paris-diderot.fr Web Site: http://www.dsimb.inserm.fr/~fuchs -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] how can i install sse4.1 and avx_256
hi Reading file md.tpr, VERSION 4.6.5 (single precision) Using 4 MPI threads Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine, which is better) how can i use this features for better mdrun speed. thanke you. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] how to generate executable file g_energy_mpi ?
Hi everyone. I am new to Gromacs and have some question in the building part. I can only genarate mdrun_mpi when set the DGMX_BUILD_MDRUN_ONLY to ON and when I set it to OFF, it can only produce gmx_mpi and template in the bin directory. What is the different between mdrun_mpi and gmx_mpi ? Another question, I want to produce .xvg file to do some analysis and have to use g_energy_mpi to read the .edr file, so how can I generate g_energy_mpi ? Thanks:) Vincent -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how can i install sse4.1 and avx_256
Hello, you need to recompile Gromacs with -DGMX_SIMD=AVX_256 flag. Or not set it and cmake should detect and use the best version automatically. http://www.gromacs.org/Documentation/Installation_Instructions_4.6 Cheers. Milan On Thursday 23 October 2014 13:32:31 Yaser Hosseini wrote: hi Reading file md.tpr, VERSION 4.6.5 (single precision) Using 4 MPI threads Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine, which is better) how can i use this features for better mdrun speed. thanke you. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
On 10/23/14 6:37 AM, Vinson Leung wrote: Hi everyone. I am new to Gromacs and have some question in the building part. I can only genarate mdrun_mpi when set the DGMX_BUILD_MDRUN_ONLY to ON and when I set it to OFF, it can only produce gmx_mpi and template in the bin directory. What is the different between mdrun_mpi and gmx_mpi ? Another question, I want to produce .xvg file to do some analysis and have to use g_energy_mpi to read the .edr file, so how can I generate g_energy_mpi ? Thanks:) The only program that benefits from MPI is mdrun. None of the analysis tools are parallelized via MPI. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] charge group moved too far between two domain decomposition step
On 10/23/14 1:47 AM, Kester Wong wrote: Thanks for the input Mark, Hi Mark, Thanks for the input, I thought a time step of 2 fs is small enough? Not always for unconstrained bonds. The largest stable time step is determined by the size of the period of the fastest oscillation, which is vibrations of bonds to hydrogen. 2fs stretches the friendship there, and you would certainly want to equilibrate thoroughly at a smaller time step if you are doing messy things like freezing nanotubes and using pressure coupling (which would not be my go-to method without a clear demonstration that it produces valid results). More generally, equilibration of a not-quite-right structure works better with a smaller timestep, because that copes better with the artificially large forces you get... Thanks for the very informative input, this should serve me well as a rule of thumb for other systems as well. Should I change my constraint to constraints = h-bonds instead? That would be normal. Or, Should I use: dt = 0.001; 1 fs constraints = none constraint-algorithm = shake Plausible, but you want to run the final simulation with constraints for the higher ns/day you obtain... I think I got a little confused here. For energy minimisation, I should impose no constraints to obtain the most realistic starting structure for NVT. As for NVT, one should use constraints to achieve a much quicker equilibration, is that correct? Constraints generally do not have any real effect on how fast the target properties reach an equilibrated state. One thing to note - if you're going to run with constraints, you should minimize with constraints. If you have problems during EM, setting the water to flexible can help, but you should minimize again with constraints, otherwise you often end up with distorted geometries that cannot be constrained during dynamics. For instance (TIPS3P.itp as shown below), to achieve a higher ns/day in my NVT run, I should have constraints = none in my mdp setting (using [bonds] and [angles]); If you want greater performance in terms of ns/day, you need a larger time step, in which case you do want constraints. Now, the thing you have to consider is that if there is a [settles] block, constraints = none has no real effect since the SETTLE algorithm will be used (constraints present in the topology always override constraints = none in the .mdp file). or have define = -DFLEXIBLE in my .mdp setting to achieve a better energy minimisation (using SETTLES)? See above. [ moleculetype ] ; molname nrexcl SOL 2 [ atoms ] ; id at type res nr residu name at name cg nr charge 1 OT 1 SOL OW 1 -0.834 2 HT 1 SOL HW1 10.417 3 HT 1 SOL HW2 10.417 #ifndef FLEXIBLE [ settles ] ; i j funct length 1 1 0.09572 0.15139 [ exclusions ] 1 2 3 2 1 3 3 1 2 #else [ bonds ] ; i j funct length force.c. 1 2 1 0.09572 376560.0 0.09572376560.0 1 3 1 0.09572 376560.0 0.09572376560.0 [ angles ] ; i j k funct angle force.c. 21 3 1 104.52 460.24 104.52 460.24 #endif I would appreciate it if anyone could point me to the right direction, once and for all. For constraints in a system with tips3p.itp, and hydronium.itp, how does one define FLEXIBLE or CONSTRAINTS? define = -DFLEXIBLE in the .mdp file. See the manual. Below is the .itp for the hydronium model: [ moleculetype ] ; molname nrexcl H3O 2 [ atoms ] ; idat type res nr residu name at name cg nr charge mass 1 H3O-O1 H3O OW 1 -0.59 15.9994 2 H3O-H1 H3O H31 1 0.53 1.008 3 H3O-H1 H3O H32 1 0.53 1.008 4 H3O-H1 H3O H33 1 0.53 1.008 ; use #ifdef FLEXIBLE or #ifdef CONSTRAINTS #ifdef FLEXIBLE [ bonds ] ;i j funct length (nm) force.c. (kJ/mol) 1 2 1 0.102 ; Hydronium OH bond 1 3 1 0.102 ; Wolf used 1000? check this 1 4 1 0.102 ; Hub used 400,000 [ angles ] ; i j k funct angle force.c. 21 3 1 112 ; Hub used 400 force 41 3 1 112 21 4 1 112 #else [ constraints ] ; ifunct dohdhh 1 2 2 0.102 1 3 2 0.102 1 4 2 0.102 2 3 2 0.169124 3 4 2 0.169124 2 4 2 0.169124 #endif [ exclusions ] 1 2 3 4 2 1 3 4 3 1 2 4 4 1 2 3 I tried the two approaches, by either having define = -DFLEXIBLE in
Re: [gmx-users] how can i install sse4.1 and avx_256
On 10/23/14 6:02 AM, Yaser Hosseini wrote: hi Reading file md.tpr, VERSION 4.6.5 (single precision) Using 4 MPI threads Compiled acceleration: SSE4.1 (Gromacs could use AVX_256 on this machine, which is better) how can i use this features for better mdrun speed. The level of optimization is automatically detected by cmake, but if you're compiling for a cluster that may have compute nodes with different configurations from the head node, you need to explicitly set -DGMX_CPU_ACCELERATION (note this has been changed to -DGMX_SIMD in 5.0) to the desired level of optimization. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
and how can I get those analysis tools? On Thu, Oct 23, 2014 at 7:31 PM, Justin Lemkul jalem...@vt.edu wrote: On 10/23/14 6:37 AM, Vinson Leung wrote: Hi everyone. I am new to Gromacs and have some question in the building part. I can only genarate mdrun_mpi when set the DGMX_BUILD_MDRUN_ONLY to ON and when I set it to OFF, it can only produce gmx_mpi and template in the bin directory. What is the different between mdrun_mpi and gmx_mpi ? Another question, I want to produce .xvg file to do some analysis and have to use g_energy_mpi to read the .edr file, so how can I generate g_energy_mpi ? Thanks:) The only program that benefits from MPI is mdrun. None of the analysis tools are parallelized via MPI. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
On 10/23/14 7:48 AM, Vinson Leung wrote: and how can I get those analysis tools? http://www.gromacs.org/Documentation/Installation_Instructions Everything gets built by default unless you use -DGMX_BUILD_MDRUN_ONLY=ON -Justin On Thu, Oct 23, 2014 at 7:31 PM, Justin Lemkul jalem...@vt.edu wrote: On 10/23/14 6:37 AM, Vinson Leung wrote: Hi everyone. I am new to Gromacs and have some question in the building part. I can only genarate mdrun_mpi when set the DGMX_BUILD_MDRUN_ONLY to ON and when I set it to OFF, it can only produce gmx_mpi and template in the bin directory. What is the different between mdrun_mpi and gmx_mpi ? Another question, I want to produce .xvg file to do some analysis and have to use g_energy_mpi to read the .edr file, so how can I generate g_energy_mpi ? Thanks:) The only program that benefits from MPI is mdrun. None of the analysis tools are parallelized via MPI. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
On 10/23/14 8:03 AM, Vinson Leung wrote:
But I remove the option -DGMX_BUILD_MDRUN_ONLY=ON and can only get
gmx_mpi and template. There is no those analysis tools and my Gromacs
version is 5.0.2.
My build congfiure is like below:
=
export CCDIR=/opt/intel/composer_xe_2013_sp1.2.144/bin/intel64
export CMAKE_PREFIX_PATH=/opt/mpich
export ZLIB_DIR=/usr
../../cmake-3.0.2/bin/cmake .. \
-DBUILD_SHARED_LIBS=OFF \
-DGMX_PREFER_STATIC_LIBS=ON \
-DGMX_FFT_LIBRARY=mkl \
-DCMAKE_INSTALL_PREFIX=/home/vincent/gromacs-5.0.2/build-all \
-DGMX_MPI=ON \
-DGMX_GPU=OFF \
-DGMX_XML=OFF \
-DGMX_SOFTWARE_INVSQRT=OFF \
-DGMX_SKIP_DEFAULT_CFLAGS=ON \
-DCMAKE_EXE_LINKER_FLAGS=-L${ZLIB_DIR}/lib64 -mkl=sequential \
-DCMAKE_CXX_COMPILER=icpc \
-DCMAKE_C_COMPILER=icc \
-DCMAKE_C_FLAGS= -xAVX -O3 -vec-report1 -fno-alias -g -DNDEBUG -ip
-funroll-all-loops -fimf-domain-exclusion=15 -I${ZLIB_DIR}/include \
-DCMAKE_CXX_FLAGS= -xAVX -O3 -vec-report1 -fno-alias -g -DNDEBUG -ip
-funroll-all-loops -fimf-domain-exclusion=15 -I${ZLIB_DIR}/include
==
Is there anything I miss?
In version 5.0, every Gromacs program is now a module of the gmx binary, so
g_energy is just a symlink to gmx energy, which is now the actual command.
There is no point in compiling anything other than mdrun with MPI support, so
set -DGMX_MPI=OFF unless doing an mdrun-only build.
-Justin
--
==
Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow
Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201
jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul
==
--
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
* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a
mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
Thanks. It seem to much more clear for me:).
I have compiled again with -DGMX_MPI=OFF. Now I have only two executable
file (gmx and template) in the bin dir. If I want to generate .xvg file
and I already have .edr file. What is the comand line exactly ?
Vincent
On Thu, Oct 23, 2014 at 8:05 PM, Justin Lemkul jalem...@vt.edu wrote:
On 10/23/14 8:03 AM, Vinson Leung wrote:
But I remove the option -DGMX_BUILD_MDRUN_ONLY=ON and can only get
gmx_mpi and template. There is no those analysis tools and my Gromacs
version is 5.0.2.
My build congfiure is like below:
=
export CCDIR=/opt/intel/composer_xe_2013_sp1.2.144/bin/intel64
export CMAKE_PREFIX_PATH=/opt/mpich
export ZLIB_DIR=/usr
../../cmake-3.0.2/bin/cmake .. \
-DBUILD_SHARED_LIBS=OFF \
-DGMX_PREFER_STATIC_LIBS=ON \
-DGMX_FFT_LIBRARY=mkl \
-DCMAKE_INSTALL_PREFIX=/home/vincent/gromacs-5.0.2/build-all \
-DGMX_MPI=ON \
-DGMX_GPU=OFF \
-DGMX_XML=OFF \
-DGMX_SOFTWARE_INVSQRT=OFF \
-DGMX_SKIP_DEFAULT_CFLAGS=ON \
-DCMAKE_EXE_LINKER_FLAGS=-L${ZLIB_DIR}/lib64 -mkl=sequential \
-DCMAKE_CXX_COMPILER=icpc \
-DCMAKE_C_COMPILER=icc \
-DCMAKE_C_FLAGS= -xAVX -O3 -vec-report1 -fno-alias -g -DNDEBUG -ip
-funroll-all-loops -fimf-domain-exclusion=15 -I${ZLIB_DIR}/include \
-DCMAKE_CXX_FLAGS= -xAVX -O3 -vec-report1 -fno-alias -g -DNDEBUG -ip
-funroll-all-loops -fimf-domain-exclusion=15 -I${ZLIB_DIR}/include
==
Is there anything I miss?
In version 5.0, every Gromacs program is now a module of the gmx binary,
so g_energy is just a symlink to gmx energy, which is now the actual
command. There is no point in compiling anything other than mdrun with MPI
support, so set -DGMX_MPI=OFF unless doing an mdrun-only build.
-Justin
--
==
Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow
Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201
jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul
==
--
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
* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
send a mail to gmx-users-requ...@gromacs.org.
--
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
* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a
mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
On 10/23/14 8:26 AM, Vinson Leung wrote: Thanks. It seem to much more clear for me:). I have compiled again with -DGMX_MPI=OFF. Now I have only two executable file (gmx and template) in the bin dir. If I want to generate .xvg file and I already have .edr file. What is the comand line exactly ? gmx energy -f *.edr Note that you can get the help information (syntax, required input, etc) for any program with gmx help [program] Everything is also in the printed manual and simple tasks are well covered in most tutorials available online. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] how to generate executable file g_energy_mpi ?
Thank you very much:) It help me a lot ! Best Vincent On Thu, Oct 23, 2014 at 8:28 PM, Justin Lemkul jalem...@vt.edu wrote: On 10/23/14 8:26 AM, Vinson Leung wrote: Thanks. It seem to much more clear for me:). I have compiled again with -DGMX_MPI=OFF. Now I have only two executable file (gmx and template) in the bin dir. If I want to generate .xvg file and I already have .edr file. What is the comand line exactly ? gmx energy -f *.edr Note that you can get the help information (syntax, required input, etc) for any program with gmx help [program] Everything is also in the printed manual and simple tasks are well covered in most tutorials available online. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] binding Free energy
Dear gromacs user, I would like to calculate Binding Free Energy Calculations for my protein-DNA complex (already run for 50ns). Is there any manual or tutorial (for more complex systems) available? best regards Urszula - Ta wiadomość została wysłana z serwera Uniwersytetu Gdańskiego http://www.ug.edu.pl/ -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] binding Free energy
On 10/23/14 10:35 AM, Urszula Uciechowska wrote: Dear gromacs user, I would like to calculate Binding Free Energy Calculations for my protein-DNA complex (already run for 50ns). Is there any manual or tutorial (for more complex systems) available? See published methods like g_mmpbsa (very recent) or GMXAPBS. -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] binding Free energy
Dear Urszula, Apologies for the self promotion, but we've made some attempts to do that for a TF-DNA complex: http://www.pnas.org/content/110/49/19796. Note that the rotational degrees of freedom are hard to sample and you may need some further assumptions about mesoscopic scales for the energies to be meaningful. Kind regards, Erik On 23 Oct 2014, at 15:35, Urszula Uciechowska urszula.uciechow...@biotech.ug.edu.pl wrote: Dear gromacs user, I would like to calculate Binding Free Energy Calculations for my protein-DNA complex (already run for 50ns). Is there any manual or tutorial (for more complex systems) available? best regards Urszula - Ta wiadomość została wysłana z serwera Uniwersytetu Gdańskiego http://www.ug.edu.pl/ -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] binding Free energy
Hi, thanks Justin to point out our tool GMXPBSA (GMXAPBS was the old one) http://www.sciencedirect.com/science/article/pii/S0010465514002240 and new version 2.1: http://www.sciencedirect.com/science/article/pii/S0010465514003154 The link to the program is still not avaliable but you can grab it here: http://gdriv.es/gmxpbsa along with examples. For any help please consider to use our small mailing list here https://groups.google.com/forum/#!forum/gmxpbsa Best and On 23/10/2014 16:50, Justin Lemkul wrote: On 10/23/14 10:35 AM, Urszula Uciechowska wrote: Dear gromacs user, I would like to calculate Binding Free Energy Calculations for my protein-DNA complex (already run for 50ns). Is there any manual or tutorial (for more complex systems) available? See published methods like g_mmpbsa (very recent) or GMXAPBS. -Justin -- --- Andrea Spitaleri PhD Principal Investigator AIRC D3 - Drug Discovery Development Istituto Italiano di Tecnologia Via Morego, 30 16163 Genova cell: +39 3485188790 http://www.iit.it/en/d3-people/andrea-spitaleri.html ORCID: http://orcid.org/-0003-3012-3557 -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Question
Hello Gromacs users, If I have all the files that were used to generate a trajectory file, what is the proper procedure to update the atomnames in the trajectory. My best guess is below. Is this problematic? # remove first frame of trajectory cat EOF tmp.ndx [ FIRSTFRAME ] 1 EOF trjconv -fr tmp.ndx -f old.trr -s old.tpr -o tmp.gro # Edit tmp.gro and rename edit.gro # Edit old.itp to edit.itp # Edit old.top tp edit.top grompp -f old.mdp -c edit.gro -p edit.top -o edit.tpr trjconv -s edit.tpr -f old.trr -o new.trr Thanks for the help! Best, Eric -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Question
On 10/23/14 4:29 PM, Eric Smoll wrote: Hello Gromacs users, If I have all the files that were used to generate a trajectory file, what is the proper procedure to update the atomnames in the trajectory. My best guess is below. Is this problematic? # remove first frame of trajectory cat EOF tmp.ndx [ FIRSTFRAME ] 1 EOF trjconv -fr tmp.ndx -f old.trr -s old.tpr -o tmp.gro # Edit tmp.gro and rename edit.gro # Edit old.itp to edit.itp # Edit old.top tp edit.top grompp -f old.mdp -c edit.gro -p edit.top -o edit.tpr trjconv -s edit.tpr -f old.trr -o new.trr Why do you need new atom names? Since atom names are not stored in the trajectory, it is simple to change the input to have things reinterpreted: 1. Edit topology that contains atom names to be changed. 2. Run grompp with original coordinate file, new topology, and -maxwarn 1 (because it will complain about mismatching atom names, but will override those in the coordinate file with whatever is in the topology) 3. Carry out whatever analysis you wanted with the new atom names 4. Profit? -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Question
Thanks for the guidance, Justin. -Eric On Thu, Oct 23, 2014 at 2:39 PM, Justin Lemkul jalem...@vt.edu wrote: On 10/23/14 4:29 PM, Eric Smoll wrote: Hello Gromacs users, If I have all the files that were used to generate a trajectory file, what is the proper procedure to update the atomnames in the trajectory. My best guess is below. Is this problematic? # remove first frame of trajectory cat EOF tmp.ndx [ FIRSTFRAME ] 1 EOF trjconv -fr tmp.ndx -f old.trr -s old.tpr -o tmp.gro # Edit tmp.gro and rename edit.gro # Edit old.itp to edit.itp # Edit old.top tp edit.top grompp -f old.mdp -c edit.gro -p edit.top -o edit.tpr trjconv -s edit.tpr -f old.trr -o new.trr Why do you need new atom names? Since atom names are not stored in the trajectory, it is simple to change the input to have things reinterpreted: 1. Edit topology that contains atom names to be changed. 2. Run grompp with original coordinate file, new topology, and -maxwarn 1 (because it will complain about mismatching atom names, but will override those in the coordinate file with whatever is in the topology) 3. Carry out whatever analysis you wanted with the new atom names 4. Profit? -Justin -- == Justin A. Lemkul, Ph.D. Ruth L. Kirschstein NRSA Postdoctoral Fellow Department of Pharmaceutical Sciences School of Pharmacy Health Sciences Facility II, Room 629 University of Maryland, Baltimore 20 Penn St. Baltimore, MD 21201 jalem...@outerbanks.umaryland.edu | (410) 706-7441 http://mackerell.umaryland.edu/~jalemkul == -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- 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 * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] error in the middle of running mdrun_mpi
Dear gromacs users, I try simulate protein folding using REMD sampling method in implicit solvent. I run my simulation on MPI-compiled gromacs 5.0.2 on single node. I have succesfully minimized equilibrated (NVT-constrained, and NPT constrained) my system. However, In the middle of mdrun_mpi process, the warning messages appear. *starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.5 steps, 50.0 ps.step 2873100, will finish Sat Nov 1 10:03:07 2014WARNING: Listed nonbonded interaction between particles 192 and 197at distance 16.773 which is larger than the table limit 10.500 nm.This is likely either a 1,4 interaction, or a listed interaction insidea smaller molecule you are decoupling during a free energy calculation.Since interactions at distances beyond the table cannot be computed,they are skipped until they are inside the table limit again. You willonly see this message once, even if it occurs for several interactions.IMPORTANT: This should not happen in a stable simulation, so there isprobably something wrong with your system. Only change the table-extensiondistance in the mdp file if you are really sure that is the reason.* *[nizarPC:07548] *** Process received signal ***[nizarPC:07548] Signal: Segmentation fault (11)[nizarPC:07548] Signal code: Address not mapped (1)[nizarPC:07548] Failing at address: 0x1ef8d90[nizarPC:07548] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7f610bc9fc30][nizarPC:07548] [ 1] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(nb_kernel_ElecGB_VdwLJ_GeomP1P1_F_avx_256_single+0x836) [0x7f610d3a2466][nizarPC:07548] [ 2] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_nonbonded+0x240) [0x7f610d235a30][nizarPC:07548] [ 3] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_lowlevel+0x1d3e) [0x7f610d97bebe][nizarPC:07548] [ 4] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_cutsGROUP+0x1510) [0x7f610d91bbe0][nizarPC:07548] [ 5] mdrun_mpi(do_md+0x57c1) [0x42e5e1][nizarPC:07548] [ 6] mdrun_mpi(mdrunner+0x12a1) [0x413af1][nizarPC:07548] [ 7] mdrun_mpi(_Z9gmx_mdruniPPc+0x18e5) [0x4337b5][nizarPC:07548] [ 8] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(_ZN3gmx24CommandLineModuleManager3runEiPPc+0x92) [0x7f610ce15a42][nizarPC:07548] [ 9] mdrun_mpi(main+0x7c) [0x40cb8c][nizarPC:07548] [10] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7f610bc8aec5][nizarPC:07548] [11] mdrun_mpi() [0x40ccce][nizarPC:07548] *** End of error message ***--mpirun noticed that process rank 5 with PID 7548 on node nizarPC exited on signal 11 (Segmentation fault).* I have increased the table-extension to 500.00 (how much this value should be?), and re-grompp and mdrun again. there were no warning message regarding table-extension anymore, However, this error messages showed: *starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.5 steps, 50.0 ps.step 4142800, will finish Sat Nov 1 10:35:55 2014[nizarPC:09984] *** Process received signal ***[nizarPC:09984] Signal: Segmentation fault (11)[nizarPC:09984] Signal code: Address not mapped (1)[nizarPC:09984] Failing at address: 0x1464040[nizarPC:09984] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7fa764b65c30][nizarPC:09984] [ 1] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(nb_kernel_ElecGB_VdwLJ_GeomP1P1_F_avx_256_single+0x85f) [0x7fa76626848f][nizarPC:09984] [ 2] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_nonbonded+0x240) [0x7fa7660fba30][nizarPC:09984] [ 3] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_lowlevel+0x1d3e) [0x7fa766841ebe][nizarPC:09984] [ 4] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_cutsGROUP+0x1510) [0x7fa7667e1be0][nizarPC:09984] [ 5] mdrun_mpi(do_md+0x57c1) [0x42e5e1][nizarPC:09984] [ 6] mdrun_mpi(mdrunner+0x12a1) [0x413af1][nizarPC:09984] [ 7] mdrun_mpi(_Z9gmx_mdruniPPc+0x18e5) [0x4337b5][nizarPC:09984] [ 8] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(_ZN3gmx24CommandLineModuleManager3runEiPPc+0x92) [0x7fa765cdba42][nizarPC:09984] [ 9] mdrun_mpi(main+0x7c) [0x40cb8c][nizarPC:09984] [10] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7fa764b50ec5][nizarPC:09984] [11] mdrun_mpi()
Re: [gmx-users] error in the middle of running mdrun_mpi
Hi, The warning message told you not to increase the table distance unless you were sure the table distance was the problem. Why were you sure the table distance was the problem, rather than some form of general instability of your system? In addition to all the usual reasons for http://www.gromacs.org/Documentation/Terminology/Blowing_Up, the GB kernels are completely untested, so you might try running with 4.5.7 (last version known to be probably-good for GB) to see whether the problem is in the code or your setup. Mark On Thu, Oct 23, 2014 at 10:38 PM, Nizar Masbukhin nizar.fku...@gmail.com wrote: Dear gromacs users, I try simulate protein folding using REMD sampling method in implicit solvent. I run my simulation on MPI-compiled gromacs 5.0.2 on single node. I have succesfully minimized equilibrated (NVT-constrained, and NPT constrained) my system. However, In the middle of mdrun_mpi process, the warning messages appear. *starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.5 steps, 50.0 ps.step 2873100, will finish Sat Nov 1 10:03:07 2014WARNING: Listed nonbonded interaction between particles 192 and 197at distance 16.773 which is larger than the table limit 10.500 nm.This is likely either a 1,4 interaction, or a listed interaction insidea smaller molecule you are decoupling during a free energy calculation.Since interactions at distances beyond the table cannot be computed,they are skipped until they are inside the table limit again. You willonly see this message once, even if it occurs for several interactions.IMPORTANT: This should not happen in a stable simulation, so there isprobably something wrong with your system. Only change the table-extensiondistance in the mdp file if you are really sure that is the reason.* *[nizarPC:07548] *** Process received signal ***[nizarPC:07548] Signal: Segmentation fault (11)[nizarPC:07548] Signal code: Address not mapped (1)[nizarPC:07548] Failing at address: 0x1ef8d90[nizarPC:07548] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7f610bc9fc30][nizarPC:07548] [ 1] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(nb_kernel_ElecGB_VdwLJ_GeomP1P1_F_avx_256_single+0x836) [0x7f610d3a2466][nizarPC:07548] [ 2] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_nonbonded+0x240) [0x7f610d235a30][nizarPC:07548] [ 3] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_lowlevel+0x1d3e) [0x7f610d97bebe][nizarPC:07548] [ 4] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_force_cutsGROUP+0x1510) [0x7f610d91bbe0][nizarPC:07548] [ 5] mdrun_mpi(do_md+0x57c1) [0x42e5e1][nizarPC:07548] [ 6] mdrun_mpi(mdrunner+0x12a1) [0x413af1][nizarPC:07548] [ 7] mdrun_mpi(_Z9gmx_mdruniPPc+0x18e5) [0x4337b5][nizarPC:07548] [ 8] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(_ZN3gmx24CommandLineModuleManager3runEiPPc+0x92) [0x7f610ce15a42][nizarPC:07548] [ 9] mdrun_mpi(main+0x7c) [0x40cb8c][nizarPC:07548] [10] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7f610bc8aec5][nizarPC:07548] [11] mdrun_mpi() [0x40ccce][nizarPC:07548] *** End of error message ***--mpirun noticed that process rank 5 with PID 7548 on node nizarPC exited on signal 11 (Segmentation fault).* I have increased the table-extension to 500.00 (how much this value should be?), and re-grompp and mdrun again. there were no warning message regarding table-extension anymore, However, this error messages showed: *starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'starting mdrun 'Protein'5 steps, 50.0 ps.starting mdrun 'Protein'5 steps, 50.0 ps.5 steps, 50.0 ps.step 4142800, will finish Sat Nov 1 10:35:55 2014[nizarPC:09984] *** Process received signal ***[nizarPC:09984] Signal: Segmentation fault (11)[nizarPC:09984] Signal code: Address not mapped (1)[nizarPC:09984] Failing at address: 0x1464040[nizarPC:09984] [ 0] /lib/x86_64-linux-gnu/libc.so.6(+0x36c30) [0x7fa764b65c30][nizarPC:09984] [ 1] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(nb_kernel_ElecGB_VdwLJ_GeomP1P1_F_avx_256_single+0x85f) [0x7fa76626848f][nizarPC:09984] [ 2] /usr/local/gromacs/bin/../lib/libgromacs_mpi.so.0(do_nonbonded+0x240) [0x7fa7660fba30][nizarPC:09984] [ 3]
Re: [gmx-users] charge group moved too far between two domain decomposition step
Hi Justin and all, Thanks for the input, I thought a time step of 2 fs is small enough? Not always for unconstrained bonds. The largest stable time step is determined by the size of the period of the fastest oscillation, which is vibrations of bonds to hydrogen. 2fs stretches the friendship there, and you would certainly want to equilibrate thoroughly at a smaller time step if you are doing messy things like freezing nanotubes and using pressure coupling (which would not be my go-to method without a clear demonstration that it produces valid results). More generally, equilibration of a not-quite-right structure works better with a smaller timestep, because that copes better with the artificially large forces you get... Thanks for the very informative input, this should serve me well as a rule of thumb for other systems as well. Should I change my constraint to constraints = h-bonds instead? That would be normal. Or, Should I use: dt = 0.001; 1 fs constraints = none constraint-algorithm = shake Plausible, but you want to run the final simulation with constraints for the higher ns/day you obtain... I think I got a little confused here. For energy minimisation, I should impose no constraints to obtain the most realistic starting structure for NVT. As for NVT, one should use constraints to achieve a much quicker equilibration, is that correct? Constraints generally do not have any real effect on how fast the target properties reach an equilibrated state. One thing to note - if you're going to run with constraints, you should minimize with constraints. If you have problems during EM, setting the water to flexible can help, but you should minimize again with constraints, otherwise you often end up with distorted geometries that cannot be constrained during dynamics.Yes, I do want to use constraints in later calculations, so I will do the two-step minimisation as mentioned.Isn't setting water to flexible effectively calls for [ settles ], which means the calculation will be run with constraints, shouldn't the [ bonds ] and [ angles ] restraints be used instead? For this reason, I asked "For constraints in a system with tips3p.ipt, and hydronium.itp, how does one define "FLEXIBLE" or "CONSTRAINTS"" in the previous email.Always a good thing to be cautious about the distorted geo. that cannot be constrained during MD can be the root of all evil. For instance (TIPS3P.itp as shown below), to achieve a higher ns/day in my NVT run, I should have constraints = none in my mdp setting (using [bonds] and [angles]); If you want greater performance in terms of ns/day, you need a larger time step, in which case you do want constraints. Now, the thing you have to consider is that if there is a [settles] block, "constraints = none" has no real effect since the SETTLE algorithm will be used (constraints present in the topology always override "constraints = none" in the .mdp file). This is exactly what I wanted to find out, it is clear to me now that constraints = none will have no effect since GROMACS will use what ever constraints defined in the topology (if any), as stated in the manual. Thank you for that.According to the manual, SHAKE cannot be used with energy minimisation. How does one minimise the water (settles) and ions system without using SHAKE, given that the maxwarn-angle for LINCS is 90 degree and the H-O-H angle of hydronium (H3O) is 112 degree?Can both LINCS and SHAKE be turned off during minimisation without constraints (constraints = none, i.e. using [ bonds ] and [ angles ] for water and hydronium), and then constraint-algorithm = SHAKE for the second minimisation with constraints (i.e. using [ settles ] and [ constraints ] for water and hydronium, respectively)? or have define = -DFLEXIBLE in my .mdp setting to achieve a better energy minimisation (using SETTLES)? See above. [ moleculetype ] ; molname nrexcl SOL 2 [ atoms ] ; id at type res nr residu name at name cg nr charge 1 OT 1 SOL OW 1 -0.834 2 HT 1 SOL HW1 10.417 3 HT 1 SOL HW2 10.417 #ifndef FLEXIBLE [ settles ] ; i j funct length 1 1 0.09572 0.15139 [ exclusions ] 1 2 3 2 1 3 3 1 2 #else [ bonds ] ; i j funct length force.c. 1 2 1 0.09572 376560.0 0.09572376560.0 1 3 1 0.09572 376560.0 0.09572376560.0 [ angles ] ; i j k funct angle force.c. 21 3 1 104.52 460.24 104.52 460.24 #endif I would appreciate it if anyone
Re: [gmx-users] charge group moved too far between two domain decomposition step
On 10/23/14 10:19 PM, Kester Wong wrote: Hi Justin and all, Thanks for the input, I thought a time step of 2 fs is small enough? Not always for unconstrained bonds. The largest stable time step is determined by the size of the period of the fastest oscillation, which is vibrations of bonds to hydrogen. 2fs stretches the friendship there, and you would certainly want to equilibrate thoroughly at a smaller time step if you are doing messy things like freezing nanotubes and using pressure coupling (which would not be my go-to method without a clear demonstration that it produces valid results). More generally, equilibration of a not-quite-right structure works better with a smaller timestep, because that copes better with the artificially large forces you get... Thanks for the very informative input, this should serve me well as a rule of thumb for other systems as well. Should I change my constraint to constraints = h-bonds instead? That would be normal. Or, Should I use: dt = 0.001; 1 fs constraints = none constraint-algorithm = shake Plausible, but you want to run the final simulation with constraints for the higher ns/day you obtain... I think I got a little confused here. For energy minimisation, I should impose no constraints to obtain the most realistic starting structure for NVT. As for NVT, one should use constraints to achieve a much quicker equilibration, is that correct? Constraints generally do not have any real effect on how fast the target properties reach an equilibrated state. One thing to note - if you're going to run with constraints, you should minimize with constraints. If you have problems during EM, setting the water to flexible can help, but you should minimize again with constraints, otherwise you often end up with distorted geometries that cannot be constrained during dynamics. Yes, I do want to use constraints in later calculations, so I will do the two-step minimisation as mentioned. Isn't setting water to flexible effectively calls for [ settles ], which means the calculation will be run with constraints, shouldn't the [ bonds ] and [ angles ] restraints be used instead? For this reason, I asked For constraints in a system with tips3p.ipt, and hydronium.itp, how does one define FLEXIBLE or CONSTRAINTS in the previous email. Note that the topology uses different constructs. One is #ifndef, the other is #ifdef. With #ifndef, it says if FLEXIBLE is not defined, use SETTLE, in contrast to the more direct use of #ifdef. Always a good thing to be cautious about the distorted geo. that cannot be constrained during MD can be the root of all evil. For instance (TIPS3P.itp as shown below), to achieve a higher ns/day in my NVT run, I should have constraints = none in my mdp setting (using [bonds] and [angles]); If you want greater performance in terms of ns/day, you need a larger time step, in which case you do want constraints. Now, the thing you have to consider is that if there is a [settles] block, constraints = none has no real effect since the SETTLE algorithm will be used (constraints present in the topology always override constraints = none in the .mdp file). This is exactly what I wanted to find out, it is clear to me now that constraints = none will have no effect since GROMACS will use what ever constraints defined in the topology (if any), as stated in the manual. Thank you for that. LINCS is more stable than SHAKE, anyway, so it is preferable. According to the manual, SHAKE cannot be used with energy minimisation. How does one minimise the water (settles) and ions system without using SHAKE, given that the maxwarn-angle for LINCS is 90 degree and the H-O-H angle of hydronium (H3O) is 112 degree? Bond rotations and topology angles are unrelated. Can both LINCS and SHAKE be turned off during minimisation without constraints (constraints = none, i.e. using [ bonds ] and [ angles ] for water and hydronium), and then constraint-algorithm = SHAKE for the second minimisation with constraints (i.e. using [ settles ] and [ constraints ] for water and hydronium, respectively)? Not if the statement in the manual is accurate, but you can constrain with LINCS. or have define = -DFLEXIBLE in my .mdp setting to achieve a better energy minimisation (using SETTLES)? See above. [ moleculetype ] ; molname nrexcl SOL 2 [ atoms ] ; id at type res nr residu name at name cg nr charge 1 OT 1 SOL OW 1
