[gmx-users] Query regarding all-atom simulations on 980 Ti GPUs

[Suhani Nagpal]

Greetings

I have just recently started to work on this workstation (details as shown
below) and have employed all-atom simulation of 30,000 atoms with this
command (just for 1 micro-sec),

gmx mdrun -deffnm md -v -ntmpi 2 -ntomp 6 -gpu_id 01 -pin on


Running on 1 node with total 8 cores, 16 logical cores, 2 compatible GPUs
Hardware detected:
  CPU info:
Vendor: GenuineIntel
Brand:  Intel(R) Core(TM) i7-5960X CPU @ 3.00GHz
SIMD instructions most likely to fit this hardware: AVX2_256
SIMD instructions selected at GROMACS compile time: AVX2_256
  GPU info:
Number of GPUs detected: 2
#0: NVIDIA GeForce GTX 980 Ti, compute cap.: 5.2, ECC:  no, stat:
compatible
#1: NVIDIA GeForce GTX 980 Ti, compute cap.: 5.2, ECC:  no, stat:
compatible

Reading file md.tpr, VERSION 5.1.2 (single precision)
Using 2 MPI threads
Using 6 OpenMP threads per tMPI thread

2 GPUs user-selected for this run.
Mapping of GPU IDs to the 2 PP ranks in this node: 0,1



I have just given the command,

and it shows:

step 126800, will finish Tue Jul 12 16:11:59 2016imb F  3%


and the imb F increases till 25-28%.

According to the time, this gives a performance of less 100 ns per day.
Because I have read other benchmarking studies, it should give above 200 ns
at least.

On another note (very strange),

when I subject more than 1 simulation (around 30,000 atoms) on this system,
the performance reduces a lot (~ 50 ns per day for the second trajectory). and
pin off when running more than one simulation.

What am I doing wrong? I have not worked on GPUs before.


Kindly help

Thanks a lot

Suhani
-- 
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] md with multiple ligands

[Billy Williams-Noonan]

Hi Thanh,

Maybe try to check if the numbering of your atoms in the topology files is
consistent with the .pdb or .gro files they describe.

Billy

On 1 July 2016 at 09:20, Thanh Le  wrote:

>
> > On Jun 30, 2016, at 3:48 PM, Thanh Le  wrote:
> >
> > My name is Thanh Le, a graduate student in chemistry. Currently, I am
> using gromacs to do a dynamics simulation for my RNA-protein complex. I saw
> you posted a question titled “Atoms in the .top are not numbered
> consecutively from 1” on gromacs forum. I know it has been 3 years since
> you asked the question. I would like to know if you have solved the problem
> and how to fix this error?
> > Hope to hear from you soon,
> > Thanks,
> > Thanh Le
>
> --
> 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.




-- 
Billy Noonan*|*PhD Student*|*Bsci ( *Adv* ), IA Hon

*LinkedIn Profile

**|*   +61420 382 557

Monash Institute for Pharmaceutical Sciences ( *MIPS* )
Royal Parade, Parkville, 3052
-- 
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] Failed to lock: pre.log

[OuyangYanhua]

Hi,
I am running a 100ns-REMD with 50 replicas. when continuing the REMD by 
appending to the old output files, such error happened: "Fatal error:Failed to 
lock: md1.log. Already running simulation?”
Then I had a test of 5 replicas. I limit the run time for 10 minutes 
for every short REMD and continue the REMD for at least 4 times. The test went 
sccessfully. 
why the log files are locked when I run 50 replicas REMD,how to solve 
it?  And why is OK to continue the test REMD of 5 replicas on the same system?

Best regards,
Ouyang.
-- 
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] md with multiple ligands

[Thanh Le]

> On Jun 30, 2016, at 3:48 PM, Thanh Le  wrote:
> 
> My name is Thanh Le, a graduate student in chemistry. Currently, I am using 
> gromacs to do a dynamics simulation for my RNA-protein complex. I saw you 
> posted a question titled “Atoms in the .top are not numbered consecutively 
> from 1” on gromacs forum. I know it has been 3 years since you asked the 
> question. I would like to know if you have solved the problem and how to fix 
> this error?
> Hope to hear from you soon,
> Thanks,
> Thanh Le

-- 
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] Atoms in the .top are not numbered consecutively from 1

[Justin Lemkul]

On 6/30/16 6:47 PM, Thanh Le wrote:

Hi Gandhi,
My name is Thanh Le, a graduate student in chemistry. Currently, I am using 
gromacs to do a dynamics simulation for my RNA-protein complex. I saw you 
posted a question titled “Atoms in the .top are not numbered consecutively from 
1” on gromacs forum. I know it has been 4 years since you asked the question. I 
would like to know if you have solved the problem and how to fix this error?


The atoms of each [moleculetype] directive must be numbered consecutively and 
start from 1.  Global atom numbering plays no part in topological numbering.


-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] Water molecule can not be settled

[Mark Abraham]

Hi,

Run with one rank, e.g. gmx mdrun -ntmpi 1

Mark

On Thu, Jun 30, 2016 at 6:11 PM zeineb SI CHAIB 
wrote:

> Dear Mark,
>
> Thank you for your orientation. I looked in Google and based on what you
> advised me, I prepared the following mdp file
>
> integrator = steep
> nsteps = 1000
> emtol = 1000.0
> emstep = 0.01
> energygrps= system
>
> pbc = no
> nstlist = 0
> ns-type = simple
> rlist   = 0
>
> cutoff-scheme = group
> coulombtype  = cut-off
> rcoulomb = 0
> vdw-type = Cut-off
> rvdw = 0
>
>
> However, I'm working with GROMACS 5.1.2 and apparently some combinaisons
> doesn't work. When I run EM with these mdp parameters I got the following
> error:
>
> Fatal error:
> Domain decomposition does not support simple neighbour searching, use grid
> searching or run with one MPI rank
>
> Is there a problem with these parameters?
>
> Zeineb
>
> --
> 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] Water molecule can not be settled

[zeineb SI CHAIB]

Dear Mark,

Thank you for your orientation. I looked in Google and based on what you 
advised me, I prepared the following mdp file

integrator = steep  
nsteps = 1000  
emtol = 1000.0   
emstep = 0.01   
energygrps= system

pbc = no
nstlist = 0
ns-type = simple
rlist   = 0

cutoff-scheme = group
coulombtype  = cut-off
rcoulomb = 0
vdw-type = Cut-off
rvdw = 0


However, I'm working with GROMACS 5.1.2 and apparently some combinaisons 
doesn't work. When I run EM with these mdp parameters I got the following error:

Fatal error:
Domain decomposition does not support simple neighbour searching, use grid 
searching or run with one MPI rank

Is there a problem with these parameters? 

Zeineb
  
-- 
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] REMD - replicas sampling in temperatures beyond the assigned range

[Mark Abraham]

Hi,

Best practice is to read and learn others practice from publications that
are similar to what you want to do, rather than making ad-hoc changes. In
this case, the GROMACS defaults are pretty close to the de facto standard,
and supported by analysis work done by other members of the community.

Mark

On Thu, Jun 30, 2016 at 4:16 PM NISHA Prakash 
wrote:

> Dear Justin,
>
> Thanks a lot for pointing out the issues. I now understand why there were
> such high oscillations.
>
> Could you please also tell me if there are any ideal values for pme_order
> and fourier spacing with respect to the cut offs' value of 1.4.
>
> Does the following Note imply I can raise the fourier grid spacing to 0.25?
>
> NOTE 2 [file sim-new.mdp]:
>   The optimal PME mesh load for parallel simulations is below 0.5
>   and for highly parallel simulations between 0.25 and 0.33,
>   for higher performance, increase the cut-off and the PME grid spacing
>
> Thank you again,
>
> Nisha
>
>
> On Thu, Jun 30, 2016 at 6:55 PM, Justin Lemkul  wrote:
>
> >
> >
> > On 6/30/16 9:16 AM, NISHA Prakash wrote:
> >
> >> Dear Justin,
> >>
> >> Thank you for your reply.
> >> It is a protein carbohydrate system.  Including the solvent, the number
> of
> >> atoms is 43499.
> >> I have minimized the system for 200 ps followed by NPT and NVT
> simulations
> >> for 200 ps respectively
> >>
> >>
> > Given that your temperature output started from 0 K, then you did not
> > continue from the equilibration properly by supplying the checkpoint file
> > to grompp -t. This is important to get right, otherwise you're basically
> > starting over from some random point (an equilibrated structure without
> any
> > velocities likely isn't a physically realistic state).
> >
> > Below is the .mdp file.
> >>
> >>
> >> ; VARIOUS PREPROCESSING OPTIONS
> >> title= REMD Simulation
> >> define   = -DPOSRES
> >>
> >>
> >> ; RUN CONTROL PARAMETERS
> >> integrator   = md-vv  ; velocity verlet algorithm -
> >> tinit= 0 ;
> >> dt   = 0.002; timestep in ps
> >> nsteps  = 500;
> >> simulation-part  = 1 ; Part index is updated automatically on
> >> checkpointing
> >> comm-mode= Linear ; mode for center of mass motion
> removal
> >> nstcomm  = 100 ; number of steps for center of mass
> motion
> >> removal
> >> comm-grps= Protein_Carb  Water_and_Ions ; group(s)
> for
> >> center of mass motion removal
> >>
> >>
> > In a solvated system, you should not be separating these groups.  This
> > could explain the sudden jump in temperature - you could have things
> > clashing badly over the course of the simulation.
> >
> >
> >
> >> ; ENERGY MINIMIZATION OPTIONS
> >> emtol= 10 ; Force tolerance
> >> emstep   = 0.01 ; initial step-size
> >> niter= 20 ; Max number of iterations in relax-shells
> >> fcstep   = 0 ; Step size (ps^2) for minimization of
> >> flexible constraints
> >> nstcgsteep   = 1000 ; Frequency of steepest descents steps
> >> when
> >> doing CG
> >> nbfgscorr= 10
> >>
> >>
> >> ; OUTPUT CONTROL OPTIONS
> >> nstxout  = 5 ; Writing full precision coordinates
> >> every
> >> ns
> >> nstvout  = 5 ; Writing velocities every nanosecond
> >> nstfout  = 0 ; Not writing forces
> >> nstlog   = 5000  ; Writing to the log file every step
> 10ps
> >> nstcalcenergy= 100
> >> nstenergy= 5000  ; Writing out energy information every
> >> step 10ps
> >> nstxtcout= 2500  ; Writing coordinates every 5 ps
> >> xtc-precision= 1000
> >> xtc-grps = Protein_Carb  Water_and_Ions ; subset of
> >> atoms for the .xtc file.
> >> energygrps   = Protein_Carb  Water_and_Ions ; Selection
> of
> >> energy groups
> >>
> >>
> >> ; NEIGHBORSEARCHING PARAMETERS
> >> nstlist  = 10 ; nblist update frequency-
> >> ns-type  = Grid ; ns algorithm (simple or grid)
> >> pbc  = xyz ; Periodic boundary conditions: xyz,
> >> no,
> >> xy
> >> periodic-molecules   = no
> >> rlist= 1.4 ;  nblist cut-off
> >> rlistlong= -1 ; long-range cut-off for switched
> >> potentials
> >>
> >>
> >> ; OPTIONS FOR ELECTROSTATICS
> >> coulombtype  = PME ; Method for doing electrostatics
> >> rcoulomb = 1.4 ;
> >> epsilon-r= 1 ; Relative dielectric constant for the
> >> medium
> >> pme_order= 10;
> >>
> >>
> >> ; OPTIONS FOR VDW
> >> vdw-type = Cut-off  ; Method for doing Van der Waals
> >> rvdw-switch  = 0 ; cut-off lengths
> >> rvdw = 1.4 ;

Re: [gmx-users] REMD - replicas sampling in temperatures beyond the assigned range

[NISHA Prakash]

Dear Justin,

Thanks a lot for pointing out the issues. I now understand why there were
such high oscillations.

Could you please also tell me if there are any ideal values for pme_order
and fourier spacing with respect to the cut offs' value of 1.4.

Does the following Note imply I can raise the fourier grid spacing to 0.25?

NOTE 2 [file sim-new.mdp]:
  The optimal PME mesh load for parallel simulations is below 0.5
  and for highly parallel simulations between 0.25 and 0.33,
  for higher performance, increase the cut-off and the PME grid spacing

Thank you again,

Nisha


On Thu, Jun 30, 2016 at 6:55 PM, Justin Lemkul  wrote:

>
>
> On 6/30/16 9:16 AM, NISHA Prakash wrote:
>
>> Dear Justin,
>>
>> Thank you for your reply.
>> It is a protein carbohydrate system.  Including the solvent, the number of
>> atoms is 43499.
>> I have minimized the system for 200 ps followed by NPT and NVT simulations
>> for 200 ps respectively
>>
>>
> Given that your temperature output started from 0 K, then you did not
> continue from the equilibration properly by supplying the checkpoint file
> to grompp -t. This is important to get right, otherwise you're basically
> starting over from some random point (an equilibrated structure without any
> velocities likely isn't a physically realistic state).
>
> Below is the .mdp file.
>>
>>
>> ; VARIOUS PREPROCESSING OPTIONS
>> title= REMD Simulation
>> define   = -DPOSRES
>>
>>
>> ; RUN CONTROL PARAMETERS
>> integrator   = md-vv  ; velocity verlet algorithm -
>> tinit= 0 ;
>> dt   = 0.002; timestep in ps
>> nsteps  = 500;
>> simulation-part  = 1 ; Part index is updated automatically on
>> checkpointing
>> comm-mode= Linear ; mode for center of mass motion removal
>> nstcomm  = 100 ; number of steps for center of mass motion
>> removal
>> comm-grps= Protein_Carb  Water_and_Ions ; group(s) for
>> center of mass motion removal
>>
>>
> In a solvated system, you should not be separating these groups.  This
> could explain the sudden jump in temperature - you could have things
> clashing badly over the course of the simulation.
>
>
>
>> ; ENERGY MINIMIZATION OPTIONS
>> emtol= 10 ; Force tolerance
>> emstep   = 0.01 ; initial step-size
>> niter= 20 ; Max number of iterations in relax-shells
>> fcstep   = 0 ; Step size (ps^2) for minimization of
>> flexible constraints
>> nstcgsteep   = 1000 ; Frequency of steepest descents steps
>> when
>> doing CG
>> nbfgscorr= 10
>>
>>
>> ; OUTPUT CONTROL OPTIONS
>> nstxout  = 5 ; Writing full precision coordinates
>> every
>> ns
>> nstvout  = 5 ; Writing velocities every nanosecond
>> nstfout  = 0 ; Not writing forces
>> nstlog   = 5000  ; Writing to the log file every step 10ps
>> nstcalcenergy= 100
>> nstenergy= 5000  ; Writing out energy information every
>> step 10ps
>> nstxtcout= 2500  ; Writing coordinates every 5 ps
>> xtc-precision= 1000
>> xtc-grps = Protein_Carb  Water_and_Ions ; subset of
>> atoms for the .xtc file.
>> energygrps   = Protein_Carb  Water_and_Ions ; Selection of
>> energy groups
>>
>>
>> ; NEIGHBORSEARCHING PARAMETERS
>> nstlist  = 10 ; nblist update frequency-
>> ns-type  = Grid ; ns algorithm (simple or grid)
>> pbc  = xyz ; Periodic boundary conditions: xyz,
>> no,
>> xy
>> periodic-molecules   = no
>> rlist= 1.4 ;  nblist cut-off
>> rlistlong= -1 ; long-range cut-off for switched
>> potentials
>>
>>
>> ; OPTIONS FOR ELECTROSTATICS
>> coulombtype  = PME ; Method for doing electrostatics
>> rcoulomb = 1.4 ;
>> epsilon-r= 1 ; Relative dielectric constant for the
>> medium
>> pme_order= 10;
>>
>>
>> ; OPTIONS FOR VDW
>> vdw-type = Cut-off  ; Method for doing Van der Waals
>> rvdw-switch  = 0 ; cut-off lengths
>> rvdw = 1.4 ;
>> DispCorr = EnerPres; Apply long range dispersion
>> corrections for Energy and Pressure
>> table-extension  = 1; Extension of the potential lookup tables
>> beyond the cut-off
>> fourierspacing   = 0.08;  Spacing for the PME/PPPM FFT grid
>>
>>
> This small Fourier spacing, coupled with the very high PME order above, is
> going to unnecessarily slow your system down.  Is there some reason you
> have set these this way?
>
>
>> ; GENERALIZED BORN ELECTROSTATICS
>> gb-algorithm = Still; Algorithm for calculating Born radii
>> nstgbradii   = 1; Frequency of 

Re: [gmx-users] REMD - replicas sampling in temperatures beyond the assigned range

[Justin Lemkul]

On 6/30/16 9:16 AM, NISHA Prakash wrote:

Dear Justin,

Thank you for your reply.
It is a protein carbohydrate system.  Including the solvent, the number of
atoms is 43499.
I have minimized the system for 200 ps followed by NPT and NVT simulations
for 200 ps respectively



Given that your temperature output started from 0 K, then you did not continue 
from the equilibration properly by supplying the checkpoint file to grompp -t. 
This is important to get right, otherwise you're basically starting over from 
some random point (an equilibrated structure without any velocities likely isn't 
a physically realistic state).



Below is the .mdp file.


; VARIOUS PREPROCESSING OPTIONS
title= REMD Simulation
define   = -DPOSRES


; RUN CONTROL PARAMETERS
integrator   = md-vv  ; velocity verlet algorithm -
tinit= 0 ;
dt   = 0.002; timestep in ps
nsteps  = 500;
simulation-part  = 1 ; Part index is updated automatically on
checkpointing
comm-mode= Linear ; mode for center of mass motion removal
nstcomm  = 100 ; number of steps for center of mass motion
removal
comm-grps= Protein_Carb  Water_and_Ions ; group(s) for
center of mass motion removal



In a solvated system, you should not be separating these groups.  This could 
explain the sudden jump in temperature - you could have things clashing badly 
over the course of the simulation.




; ENERGY MINIMIZATION OPTIONS
emtol= 10 ; Force tolerance
emstep   = 0.01 ; initial step-size
niter= 20 ; Max number of iterations in relax-shells
fcstep   = 0 ; Step size (ps^2) for minimization of
flexible constraints
nstcgsteep   = 1000 ; Frequency of steepest descents steps when
doing CG
nbfgscorr= 10


; OUTPUT CONTROL OPTIONS
nstxout  = 5 ; Writing full precision coordinates every
ns
nstvout  = 5 ; Writing velocities every nanosecond
nstfout  = 0 ; Not writing forces
nstlog   = 5000  ; Writing to the log file every step 10ps
nstcalcenergy= 100
nstenergy= 5000  ; Writing out energy information every
step 10ps
nstxtcout= 2500  ; Writing coordinates every 5 ps
xtc-precision= 1000
xtc-grps = Protein_Carb  Water_and_Ions ; subset of
atoms for the .xtc file.
energygrps   = Protein_Carb  Water_and_Ions ; Selection of
energy groups


; NEIGHBORSEARCHING PARAMETERS
nstlist  = 10 ; nblist update frequency-
ns-type  = Grid ; ns algorithm (simple or grid)
pbc  = xyz ; Periodic boundary conditions: xyz, no,
xy
periodic-molecules   = no
rlist= 1.4 ;  nblist cut-off
rlistlong= -1 ; long-range cut-off for switched
potentials


; OPTIONS FOR ELECTROSTATICS
coulombtype  = PME ; Method for doing electrostatics
rcoulomb = 1.4 ;
epsilon-r= 1 ; Relative dielectric constant for the
medium
pme_order= 10;


; OPTIONS FOR VDW
vdw-type = Cut-off  ; Method for doing Van der Waals
rvdw-switch  = 0 ; cut-off lengths
rvdw = 1.4 ;
DispCorr = EnerPres; Apply long range dispersion
corrections for Energy and Pressure
table-extension  = 1; Extension of the potential lookup tables
beyond the cut-off
fourierspacing   = 0.08;  Spacing for the PME/PPPM FFT grid



This small Fourier spacing, coupled with the very high PME order above, is going 
to unnecessarily slow your system down.  Is there some reason you have set these 
this way?




; GENERALIZED BORN ELECTROSTATICS
gb-algorithm = Still; Algorithm for calculating Born radii
nstgbradii   = 1; Frequency of calculating the Born radii
inside rlist
rgbradii = 1; Cutoff for Born radii calculation
gb-epsilon-solvent   = 80; Dielectric coefficient of the implicit
solvent
gb-saltconc  = 0; Salt concentration in M for Generalized
Born models


; Scaling factors used in the OBC GB model. Default values are OBC(II)
gb-obc-alpha = 1
gb-obc-beta  = 0.8
gb-obc-gamma = 4.85
gb-dielectric-offset = 0.009
sa-algorithm = Ace-approximation
sa-surface-tension   = -1; Surface tension (kJ/mol/nm^2) for the SA
(nonpolar surface) part of GBSA - default -1



Implicit solvent should not be used if you have explicit solvent, though it 
looks like these options are probably off since the default for the 
implicit-solvent keyword is "no," but be aware that these are extraneous.





; Temperature coupling
tcoupl = nose-hoover
nsttcouple  

Re: [gmx-users] REMD - replicas sampling in temperatures beyond the assigned range

[NISHA Prakash]

Dear Justin,

Thank you for your reply.
It is a protein carbohydrate system.  Including the solvent, the number of
atoms is 43499.
I have minimized the system for 200 ps followed by NPT and NVT simulations
for 200 ps respectively

Below is the .mdp file.


; VARIOUS PREPROCESSING OPTIONS
title= REMD Simulation
define   = -DPOSRES


; RUN CONTROL PARAMETERS
integrator   = md-vv  ; velocity verlet algorithm -
tinit= 0 ;
dt   = 0.002; timestep in ps
nsteps  = 500;
simulation-part  = 1 ; Part index is updated automatically on
checkpointing
comm-mode= Linear ; mode for center of mass motion removal
nstcomm  = 100 ; number of steps for center of mass motion
removal
comm-grps= Protein_Carb  Water_and_Ions ; group(s) for
center of mass motion removal


; ENERGY MINIMIZATION OPTIONS
emtol= 10 ; Force tolerance
emstep   = 0.01 ; initial step-size
niter= 20 ; Max number of iterations in relax-shells
fcstep   = 0 ; Step size (ps^2) for minimization of
flexible constraints
nstcgsteep   = 1000 ; Frequency of steepest descents steps when
doing CG
nbfgscorr= 10


; OUTPUT CONTROL OPTIONS
nstxout  = 5 ; Writing full precision coordinates every
ns
nstvout  = 5 ; Writing velocities every nanosecond
nstfout  = 0 ; Not writing forces
nstlog   = 5000  ; Writing to the log file every step 10ps
nstcalcenergy= 100
nstenergy= 5000  ; Writing out energy information every
step 10ps
nstxtcout= 2500  ; Writing coordinates every 5 ps
xtc-precision= 1000
xtc-grps = Protein_Carb  Water_and_Ions ; subset of
atoms for the .xtc file.
energygrps   = Protein_Carb  Water_and_Ions ; Selection of
energy groups


; NEIGHBORSEARCHING PARAMETERS
nstlist  = 10 ; nblist update frequency-
ns-type  = Grid ; ns algorithm (simple or grid)
pbc  = xyz ; Periodic boundary conditions: xyz, no,
xy
periodic-molecules   = no
rlist= 1.4 ;  nblist cut-off
rlistlong= -1 ; long-range cut-off for switched
potentials


; OPTIONS FOR ELECTROSTATICS
coulombtype  = PME ; Method for doing electrostatics
rcoulomb = 1.4 ;
epsilon-r= 1 ; Relative dielectric constant for the
medium
pme_order= 10;


; OPTIONS FOR VDW
vdw-type = Cut-off  ; Method for doing Van der Waals
rvdw-switch  = 0 ; cut-off lengths
rvdw = 1.4 ;
DispCorr = EnerPres; Apply long range dispersion
corrections for Energy and Pressure
table-extension  = 1; Extension of the potential lookup tables
beyond the cut-off
fourierspacing   = 0.08;  Spacing for the PME/PPPM FFT grid


; GENERALIZED BORN ELECTROSTATICS
gb-algorithm = Still; Algorithm for calculating Born radii
nstgbradii   = 1; Frequency of calculating the Born radii
inside rlist
rgbradii = 1; Cutoff for Born radii calculation
gb-epsilon-solvent   = 80; Dielectric coefficient of the implicit
solvent
gb-saltconc  = 0; Salt concentration in M for Generalized
Born models


; Scaling factors used in the OBC GB model. Default values are OBC(II)
gb-obc-alpha = 1
gb-obc-beta  = 0.8
gb-obc-gamma = 4.85
gb-dielectric-offset = 0.009
sa-algorithm = Ace-approximation
sa-surface-tension   = -1; Surface tension (kJ/mol/nm^2) for the SA
(nonpolar surface) part of GBSA - default -1



; Temperature coupling
tcoupl = nose-hoover
nsttcouple   = 10 ;
nh-chain-length  = 10
tc-grps  = Protein_Carb  Water_and_Ions ; Groups to
couple separately
tau-t= 1010; Time constant (ps)-
ref-t  = 270.0 270.0; reference temperature (K)


; pressure coupling
pcoupl   = no  ;-


; GENERATE VELOCITIES FOR STARTUP RUN
gen-vel  = no
gen-temp  = 270.0
gen-seed = 173529


; OPTIONS FOR BONDS
continuation = yes ;  constrain the start configuration

constraints  = all-bonds
constraint-algorithm = lincs ; Type of constraint algorithm-
lincs-order  = 4
lincs-iter   = 1
lincs-warnangle  = 30


Thank you for your help.

Nisha



On Thu, Jun 30, 2016 at 6:21 PM, Justin Lemkul  wrote:

>
>
> On 6/30/16 8:46 AM, NISHA Prakash wrote:
>
>> Dear all,
>>
>> I have conducted a 10ns REMD simulation for a protein ligand complex with
>> the temperature range - 270 

Re: [gmx-users] rmsdist

[Justin Lemkul]

On 6/30/16 8:58 AM, Sanja Zivanovic wrote:

Dear users,Could you please tell me if it is possible to make gmx rmsdist
without hydrogens, comparing just heavy atoms?


Supply an index file with a group that contains the atoms you want to use.

-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.

[gmx-users] rmsdist

[Sanja Zivanovic]

Dear users,Could you please tell me if it is possible to make gmx rmsdist
without hydrogens, comparing just heavy atoms?
Thanks in advance a lot

Best Regards, Sanja
-- 
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] REMD - replicas sampling in temperatures beyond the assigned range

[Justin Lemkul]

On 6/30/16 8:46 AM, NISHA Prakash wrote:

Dear all,

I have conducted a 10ns REMD simulation for a protein ligand complex with
the temperature range - 270 to 350 K, however the temperature distribution
plot of the replicas show that the sampling has occurred at higher
temperatures as well that is beyond 350K -
Below is an excerpt from the temperature xvg file


@title "Gromacs Energies"
@xaxis  label "Time (ps)"
@yaxis  label "(K)"
@TYPE xy
@ view 0.15, 0.15, 0.75, 0.85
@ legend on
@ legend box on
@ legend loctype view
@ legend 0.78, 0.8
@ legend length 2
@ s0 legend "Temperature"
0.000.00
   10.00  350.997864
   20.00  353.618927
   30.00  350.068481
   40.00  353.921753
   50.00  359.485565
   60.00  353.463654
   70.00  352.015778
   80.00  350.657898
   90.00  351.927155
  100.00  354.539429
  110.00  354.287720
  120.00  349.436096
  130.00  352.960541
  140.00  351.631317
  150.00  354.217407
  160.00  350.185852
  170.00  350.294434
  180.00  350.980194
  190.00  350.914429
   
   
 470.00  349.224060
  480.00  350.819458
  490.00  348.541748
  500.00  350.393127
  510.00  398.775208
  520.00  444.802856
  530.00  470.899323
  540.00  466.652740
  550.00  465.600677
  560.00  469.22
  570.00  470.548370
  580.00  470.011566
  590.00  470.643951
  600.00  472.433197
  610.00  470.451172
  620.00  469.991699
  630.00  469.073090
  640.00  467.259521
  650.00  464.561798
  660.00  468.416901
  670.00  468.754913
  680.00  469.259613
  690.00  467.641144
  700.00  468.542328


Temperature coupling was done using Nose hoover algorithm.

Does this imply the sampling is wrong or insufficent?
Any help / suggestion is appreciated.



How large is your system, and what is it?  What were your (full) .mdp settings? 
The fact that your temperature started at 0 K and ramped up suggests that you 
did not equilibrate prior to the run, did not generate appropriate velocities, 
or did not continue properly.  The sudden jump in temperature later suggests 
instability, and could be due to incorrect settings.  N-H allows for large 
oscillations, but I wouldn't expect a stable system to that degree.


-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.

[gmx-users] REMD - replicas sampling in temperatures beyond the assigned range

[NISHA Prakash]

Dear all,

I have conducted a 10ns REMD simulation for a protein ligand complex with
the temperature range - 270 to 350 K, however the temperature distribution
plot of the replicas show that the sampling has occurred at higher
temperatures as well that is beyond 350K -
Below is an excerpt from the temperature xvg file


@title "Gromacs Energies"
@xaxis  label "Time (ps)"
@yaxis  label "(K)"
@TYPE xy
@ view 0.15, 0.15, 0.75, 0.85
@ legend on
@ legend box on
@ legend loctype view
@ legend 0.78, 0.8
@ legend length 2
@ s0 legend "Temperature"
0.000.00
   10.00  350.997864
   20.00  353.618927
   30.00  350.068481
   40.00  353.921753
   50.00  359.485565
   60.00  353.463654
   70.00  352.015778
   80.00  350.657898
   90.00  351.927155
  100.00  354.539429
  110.00  354.287720
  120.00  349.436096
  130.00  352.960541
  140.00  351.631317
  150.00  354.217407
  160.00  350.185852
  170.00  350.294434
  180.00  350.980194
  190.00  350.914429
   
   
 470.00  349.224060
  480.00  350.819458
  490.00  348.541748
  500.00  350.393127
  510.00  398.775208
  520.00  444.802856
  530.00  470.899323
  540.00  466.652740
  550.00  465.600677
  560.00  469.22
  570.00  470.548370
  580.00  470.011566
  590.00  470.643951
  600.00  472.433197
  610.00  470.451172
  620.00  469.991699
  630.00  469.073090
  640.00  467.259521
  650.00  464.561798
  660.00  468.416901
  670.00  468.754913
  680.00  469.259613
  690.00  467.641144
  700.00  468.542328


Temperature coupling was done using Nose hoover algorithm.

Does this imply the sampling is wrong or insufficent?
Any help / suggestion is appreciated.

Thanking you in anticipation.

Nisha
-- 
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] Tesla K80 vs GeForce GTX 980-repost

[Mark Abraham]

Hi,

Yes, but it runs the same code.

Mark

On Thu, Jun 30, 2016 at 2:01 PM Nikhil Maroli  wrote:

> Thanks, MARK,
>
> but I saw it is "Pascal." is that new Archi. ?
>
>
>
> --
> 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] Tesla K80 vs GeForce GTX 980-repost

[Nikhil Maroli]

Thanks, MARK,

but I saw it is "Pascal." is that new Archi. ?



-- 
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] Treatment of HIE, HID, HIP with the Amber force field

[Mark Abraham]

Hi,

Only if that makes a sensible model physics... But probably this is a
problem in the (user-contributed) port of amber14sb (and there are others,
including that they haven't updated the documentation files in it, so it
claims to be something else). The alpha carbon has a new type, so probably
a dihedral for HID needs further consideration for whichever N atom type.
But you'll need to look at the literature and/or consult the people who did
the port.

Mark

On Thu, Jun 30, 2016 at 12:46 PM Julian  wrote:

> Dear Gromacs users,
>
> I am running Gromacs using the amber14sb force field. Amber distinguishes
> histamine residues according to their protonation state into HIE, HID, and
> HIP.
> But then Gromacs complains for HID residues: No default Improper Dih. types
> Do I have to do something additional or should I just discard the different
> protonation states and rename all HID to HIE?
>
> gmx pdb2gmx -f xxx.pdb -p xxx.top -ff amber14sb -water tip3p -ignh
> gmx grompp -f xxx.mdp -c xxx.pdb -p xxx.top -o xxx.tpr
>
> Thanks for your answers!
> --
> 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] Treatment of HIE, HID, HIP with the Amber force field

[Julian]

Dear Gromacs users,

I am running Gromacs using the amber14sb force field. Amber distinguishes
histamine residues according to their protonation state into HIE, HID, and
HIP.
But then Gromacs complains for HID residues: No default Improper Dih. types
Do I have to do something additional or should I just discard the different
protonation states and rename all HID to HIE?

gmx pdb2gmx -f xxx.pdb -p xxx.top -ff amber14sb -water tip3p -ignh
gmx grompp -f xxx.mdp -c xxx.pdb -p xxx.top -o xxx.tpr

Thanks for your answers!
-- 
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] Water molecule can not be settled

[Mark Abraham]

Hi,

Google has some good suggestions for examples for GROMACS mimization in
vacuo. You want no PBC, simple grid and maybe no cutoffs.

EM has no time, so time step is not a relevant concept. The .mdp option
documentation is quite useful...

Mark

On Thu, Jun 30, 2016 at 12:15 PM zeineb SI CHAIB 
wrote:

> Dear Mark,
>
> Thank you very much for your answer.
>
> I'm trying to minimize my protein in vacuo but I'm havig problem to
> undrstand the concept of "in vacuo" so I'm enable to construct an mdp file
> and I can't found valid examples in forums and internet. Could you help me
> with that please?
>
> Also, the "time step of the minimisation is 'emstep', is it correct?
>
> Thank you very much for your help.
>
> Zeineb.
>
>
>
> --
> 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] Tesla K80 vs GeForce GTX 980-repost

[Mark Abraham]

Hi,

Yes, it works already. There's no architectural change that would break
things.

Mark

On Thu, Jun 30, 2016 at 12:01 PM Nikhil Maroli  wrote:

> HI,
> Thanks
>
> I have seen 1070 is much better than gtx 980ti and cheap also, released ?
> but it will support GROMACS or MD? Since it released a few months ago?
>
>
> Thanks
>
> --
> 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] Water molecule can not be settled

[zeineb SI CHAIB]

Dear Mark, 

Thank you very much for your answer. 

I'm trying to minimize my protein in vacuo but I'm havig problem to undrstand 
the concept of "in vacuo" so I'm enable to construct an mdp file and I can't 
found valid examples in forums and internet. Could you help me with that please?

Also, the "time step of the minimisation is 'emstep', is it correct? 

Thank you very much for your help. 

Zeineb. 


  
-- 
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] Tesla K80 vs GeForce GTX 980-repost

[Nikhil Maroli]

HI,
Thanks

I have seen 1070 is much better than gtx 980ti and cheap also, released ?
but it will support GROMACS or MD? Since it released a few months ago?


Thanks

-- 
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.