Re: [gmx-users] Regarding masses in Martini force field

[Lutz Maibaum]

On Nov 11, 2010, at 6:35 PM, George Khelashvili wrote:
> I am relatively new to Martini force fields and the fact that masses of all 
> the coarse grained beads are set to a same value (72amu) somewhat bothers me. 
> I was wondering what influence may this have on the process of spontaneous 
> lipid aggregation into different phases? I read that it will generally 
> influence the kinetics of the process, but how about the actual final outcome?

As long as by "final outcome" you mean equilibrium properties, you don't have 
to worry about masses too much. Masses usually appear only in the kinetic 
energy part of the energy function, which decouples from the configurational 
part of the Boltzmann weight. Of course, if you study a system where the masses 
couple to positional degrees of freedom (e.g., gravity), you have to be more 
careful.

> Should I expect that the correct phase of lipid/water mixture will be formed 
> in the appropriate region of the phase space using default parameters?

I haven't used the Martini force field, but like all other force fields I would 
expect it to work well only for those situations for which it has been 
parametrized. Everything beyond that would require careful validation.

Hope this helps,

  Lutz

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] Regarding masses in Martini force field

[George Khelashvili]

Dear users,

I am relatively new to Martini force fields and the fact that masses of 
all the coarse grained beads are set to a same value (72amu) somewhat 
bothers me. I was wondering what influence may this have on the process 
of spontaneous lipid aggregation into different phases? I read that it 
will generally influence the kinetics of the process, but how about the 
actual final outcome? Should I expect that the correct phase of 
lipid/water mixture will be formed in the appropriate region of the 
phase space using default parameters? Or some tuning of the parameters 
is still needed?

I would appreciate your comments on this.

Thank you,
George

--
George Khelashvili, Ph.D.
Department of Physiology and Biophysics
Weill Medical College of Cornell University
1300 York Avenue, Room LC501
New York, NY, 10065, USA
gek2...@med.cornell.edu
Phone: 1-212-746-6539
Fax:   1-212-746-6226

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

RE: [gmx-users] ./mdrun

[lin hen]

 
yeah，Thanks a lot :)
 
YY
 
 
> Date: Thu, 11 Nov 2010 20:32:55 -0500
> From: jalem...@vt.edu
> To: gmx-users@gromacs.org
> Subject: Re: [gmx-users] ./mdrun
> 
> 
> 
> lin hen wrote:
> > It doesn't work, even I deleted this line
> > it stills shows the same error...
> > 
> 
> Did you re-generate the .tpr file? I see no reason that a suitable setting 
> for 
> nsteps would be magically substituted by either grompp or mdrun.
> 
> > I use ./mdrun-gpu without any flags. Did I miss something?
> > 
> 
> Is the correct .tpr file being read in? Without any flags, mdrun will look 
> for 
> "topol.tpr" by default. If this is not the file you've created, then you're 
> just re-running the same malformed input.
> 
> -Justin
> 
> > YY 
> > > Date: Thu, 11 Nov 2010 20:25:45 -0500
> > > From: jalem...@vt.edu
> > > To: gmx-users@gromacs.org
> > > Subject: Re: [gmx-users] ./mdrun
> > >
> > >
> > >
> > > lin hen wrote:
> > > >
> > > > constraints = all-bonds
> > > > integrator = md
> > > > dt = 0.002 ; ps !
> > > > #nsteps = -1
> > > > nsteps = 100
> > > > nstlist = 0
> > > > ns_type = grid
> > > > rlist = 0
> > > > coulombtype = cut-off
> > > > vdwtype = cut-off
> > > > rcoulomb = 0
> > > > rvdw = 0
> > > > pbc = no
> > > > epsilon_rf = 0
> > > > rgbradii = 0
> > > > comm_mode = angular
> > > > implicit_solvent = GBSA
> > > > gb_algorithm = OBC
> > > > gb_epsilon_solvent = 78.3
> > > > sa_surface_tension = 2.25936
> > > > nstxout = 0
> > > > nstfout = 0
> > > > nstvout = 0
> > > > nstxtcout = 0
> > > > nstlog = 0
> > > > nstcalcenergy = -1
> > > > nstenergy = 0
> > > > tcoupl = berendsen
> > > > tc-grps = system
> > > > tau-t = 0.1
> > > > ref-t = 300
> > > >
> > > > This is the .mdf file, even I modified the nsteps, it still comes with
> > > > the same error:
> > > >
> > > > Back Off! I just backed up ener.edr to ./#ener.edr.1#
> > > >
> > > > WARNING: OpenMM does not support leap-frog, will use velocity-verlet
> > > > integrator.
> > > >
> > > >
> > > > WARNING: OpenMM supports only Andersen thermostat with the
> > > > md/md-vv/md-vv-avek integrators.
> > > >
> > > >
> > > > WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE
> > > > and CCMA. Accuracy is based on the SHAKE tolerance set by the
> > > > "shake_tol" option.
> > > >
> > > >
> > > > WARNING: The OBC scale factors alpha, beta and gamma are hardcoded in
> > > > OpenMM with the default Gromacs values.
> > > >
> > > >
> > > > Pre-simulation ~15s memtest in progress...done, no errors detected
> > > > starting mdrun 'Protein'
> > > > -1 steps, infinite ps.
> > > >
> > > >
> > > > the log file:
> > > > Input Parameters:
> > > > integrator = md
> > > > nsteps = -1
> > > > init_step = 0
> > > > ns_type = Grid
> > > > nstlist = 0
> > > > ndelta = 2
> > > > nstcomm = 10
> > > > comm_mode = Angular
> > > > nstlog = 0
> > > > nstxout = 0
> > > > nstvout = 0
> > > > nstfout = 0
> > > > nstcalcenergy = 10
> > > > nstenergy = 0
> > > >
> > > >
> > > > the nsteps is still -1, did I do something wrong?
> > >
> > > You haven't properly commented out the "nsteps = -1" line. The 
> > comment sign is
> > > a ';' not '#' though I'm surprised grompp didn't simply abort with 
> > this input file.
> > >
> > > -Justin
> > >
> > > >
> > > > Thanks a lot
> > > >
> > > > YY
> > >
> > > --
> > > 
> > >
> > > Justin A. Lemkul
> > > Ph.D. Candidate
> > > ICTAS Doctoral Scholar
> > > MILES-IGERT Trainee
> > > Department of Biochemistry
> > > Virginia Tech
> > > Blacksburg, VA
> > > jalemkul[at]vt.edu | (540) 231-9080
> > > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> > >
> > > 
> > > --
> > > gmx-users mailing list gmx-users@gromacs.org
> > > http://lists.gromacs.org/mailman/listinfo/gmx-users
> > > Please search the archive at 
> > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> > > Please don't post (un)subscribe requests to the list. Use the
> > > www interface or send it to gmx-users-requ...@gromacs.org.
> > > Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> -- 
> 
> 
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> 
> 
> -- 
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at 
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
  -- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please

Re: [gmx-users] ./mdrun

[Justin A. Lemkul]

lin hen wrote:

It doesn't work, even I deleted this line
it stills shows the same error...
 


Did you re-generate the .tpr file?  I see no reason that a suitable setting for 
nsteps would be magically substituted by either grompp or mdrun.



I use ./mdrun-gpu without any flags. Did I miss something?
 


Is the correct .tpr file being read in?  Without any flags, mdrun will look for 
"topol.tpr" by default.  If this is not the file you've created, then you're 
just re-running the same malformed input.


-Justin

YY 
 > Date: Thu, 11 Nov 2010 20:25:45 -0500

 > From: jalem...@vt.edu
 > To: gmx-users@gromacs.org
 > Subject: Re: [gmx-users] ./mdrun
 >
 >
 >
 > lin hen wrote:
 > >
 > > constraints = all-bonds
 > > integrator = md
 > > dt = 0.002 ; ps !
 > > #nsteps = -1
 > > nsteps = 100
 > > nstlist = 0
 > > ns_type = grid
 > > rlist = 0
 > > coulombtype = cut-off
 > > vdwtype = cut-off
 > > rcoulomb = 0
 > > rvdw = 0
 > > pbc = no
 > > epsilon_rf = 0
 > > rgbradii = 0
 > > comm_mode = angular
 > > implicit_solvent = GBSA
 > > gb_algorithm = OBC
 > > gb_epsilon_solvent = 78.3
 > > sa_surface_tension = 2.25936
 > > nstxout = 0
 > > nstfout = 0
 > > nstvout = 0
 > > nstxtcout = 0
 > > nstlog = 0
 > > nstcalcenergy = -1
 > > nstenergy = 0
 > > tcoupl = berendsen
 > > tc-grps = system
 > > tau-t = 0.1
 > > ref-t = 300
 > >
 > > This is the .mdf file, even I modified the nsteps, it still comes with
 > > the same error:
 > >
 > > Back Off! I just backed up ener.edr to ./#ener.edr.1#
 > >
 > > WARNING: OpenMM does not support leap-frog, will use velocity-verlet
 > > integrator.
 > >
 > >
 > > WARNING: OpenMM supports only Andersen thermostat with the
 > > md/md-vv/md-vv-avek integrators.
 > >
 > >
 > > WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE
 > > and CCMA. Accuracy is based on the SHAKE tolerance set by the
 > > "shake_tol" option.
 > >
 > >
 > > WARNING: The OBC scale factors alpha, beta and gamma are hardcoded in
 > > OpenMM with the default Gromacs values.
 > >
 > >
 > > Pre-simulation ~15s memtest in progress...done, no errors detected
 > > starting mdrun 'Protein'
 > > -1 steps, infinite ps.
 > >
 > >
 > > the log file:
 > > Input Parameters:
 > > integrator = md
 > > nsteps = -1
 > > init_step = 0
 > > ns_type = Grid
 > > nstlist = 0
 > > ndelta = 2
 > > nstcomm = 10
 > > comm_mode = Angular
 > > nstlog = 0
 > > nstxout = 0
 > > nstvout = 0
 > > nstfout = 0
 > > nstcalcenergy = 10
 > > nstenergy = 0
 > >
 > >
 > > the nsteps is still -1, did I do something wrong?
 >
 > You haven't properly commented out the "nsteps = -1" line. The 
comment sign is
 > a ';' not '#' though I'm surprised grompp didn't simply abort with 
this input file.

 >
 > -Justin
 >
 > >
 > > Thanks a lot
 > >
 > > YY
 >
 > --
 > 
 >
 > Justin A. Lemkul
 > Ph.D. Candidate
 > ICTAS Doctoral Scholar
 > MILES-IGERT Trainee
 > Department of Biochemistry
 > Virginia Tech
 > Blacksburg, VA
 > jalemkul[at]vt.edu | (540) 231-9080
 > http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
 >
 > 
 > --
 > gmx-users mailing list gmx-users@gromacs.org
 > http://lists.gromacs.org/mailman/listinfo/gmx-users
 > Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!

 > Please don't post (un)subscribe requests to the list. Use the
 > www interface or send it to gmx-users-requ...@gromacs.org.
 > Can't post? Read http://www.gromacs.org/Support/Mailing_Lists


--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

RE: [gmx-users] ./mdrun

[lin hen]

It doesn't work, even I deleted this line
it stills shows the same error...
 
I use ./mdrun-gpu without any flags. Did I miss something?
 
YY 
> Date: Thu, 11 Nov 2010 20:25:45 -0500
> From: jalem...@vt.edu
> To: gmx-users@gromacs.org
> Subject: Re: [gmx-users] ./mdrun
> 
> 
> 
> lin hen wrote:
> > 
> > constraints = all-bonds
> > integrator = md
> > dt = 0.002 ; ps !
> > #nsteps = -1
> > nsteps = 100
> > nstlist = 0
> > ns_type = grid
> > rlist = 0
> > coulombtype = cut-off
> > vdwtype = cut-off
> > rcoulomb = 0
> > rvdw = 0
> > pbc = no
> > epsilon_rf = 0
> > rgbradii = 0
> > comm_mode = angular
> > implicit_solvent = GBSA
> > gb_algorithm = OBC
> > gb_epsilon_solvent = 78.3
> > sa_surface_tension = 2.25936
> > nstxout = 0
> > nstfout = 0
> > nstvout = 0
> > nstxtcout = 0
> > nstlog = 0
> > nstcalcenergy = -1
> > nstenergy = 0
> > tcoupl = berendsen
> > tc-grps = system
> > tau-t = 0.1
> > ref-t = 300
> > 
> > This is the .mdf file, even I modified the nsteps, it still comes with 
> > the same error:
> > 
> > Back Off! I just backed up ener.edr to ./#ener.edr.1#
> > 
> > WARNING: OpenMM does not support leap-frog, will use velocity-verlet 
> > integrator.
> > 
> > 
> > WARNING: OpenMM supports only Andersen thermostat with the 
> > md/md-vv/md-vv-avek integrators.
> > 
> > 
> > WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE 
> > and CCMA. Accuracy is based on the SHAKE tolerance set by the 
> > "shake_tol" option.
> > 
> > 
> > WARNING: The OBC scale factors alpha, beta and gamma are hardcoded in 
> > OpenMM with the default Gromacs values.
> > 
> > 
> > Pre-simulation ~15s memtest in progress...done, no errors detected
> > starting mdrun 'Protein'
> > -1 steps, infinite ps.
> > 
> > 
> > the log file:
> > Input Parameters:
> > integrator = md
> > nsteps = -1
> > init_step = 0
> > ns_type = Grid
> > nstlist = 0
> > ndelta = 2
> > nstcomm = 10
> > comm_mode = Angular
> > nstlog = 0
> > nstxout = 0
> > nstvout = 0
> > nstfout = 0
> > nstcalcenergy = 10
> > nstenergy = 0
> > 
> > 
> > the nsteps is still -1, did I do something wrong?
> 
> You haven't properly commented out the "nsteps = -1" line. The comment sign 
> is 
> a ';' not '#' though I'm surprised grompp didn't simply abort with this input 
> file.
> 
> -Justin
> 
> > 
> > Thanks a lot
> > 
> > YY
> 
> -- 
> 
> 
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> 
> 
> -- 
> gmx-users mailing list gmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at 
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
  -- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] ./mdrun

[Justin A. Lemkul]

lin hen wrote:


 constraints =  all-bonds
integrator  =  md
dt  =  0.002; ps !
#nsteps  =  -1
nsteps  =  100
nstlist =  0
ns_type =  grid
rlist   =  0
coulombtype =  cut-off
vdwtype =  cut-off
rcoulomb=  0
rvdw=  0
pbc =  no
epsilon_rf  =  0
rgbradii=  0
comm_mode   = angular
implicit_solvent= GBSA
gb_algorithm= OBC
gb_epsilon_solvent  = 78.3
sa_surface_tension  = 2.25936
nstxout = 0
nstfout = 0
nstvout = 0
nstxtcout   = 0
nstlog  = 0
nstcalcenergy   = -1
nstenergy   = 0
tcoupl  = berendsen
tc-grps = system
tau-t   = 0.1
ref-t   = 300

This is the .mdf file, even I modified the nsteps, it still comes with 
the same error:
 
Back Off! I just backed up ener.edr to ./#ener.edr.1#
 
WARNING: OpenMM does not support leap-frog, will use velocity-verlet 
integrator.
 

WARNING: OpenMM supports only Andersen thermostat with the 
md/md-vv/md-vv-avek integrators.
 

WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE 
and CCMA. Accuracy is based on the SHAKE tolerance set by the 
"shake_tol" option.
 

WARNING: The OBC scale factors alpha, beta and gamma are hardcoded in 
OpenMM with the default Gromacs values.
 


Pre-simulation ~15s memtest in progress...done, no errors detected
starting mdrun 'Protein'
-1 steps, infinite ps.

 
the log file:

Input Parameters:
   integrator   = md
   nsteps   = -1
   init_step= 0
   ns_type  = Grid
   nstlist  = 0
   ndelta   = 2
   nstcomm  = 10
   comm_mode= Angular
   nstlog   = 0
   nstxout  = 0
   nstvout  = 0
   nstfout  = 0
   nstcalcenergy= 10
   nstenergy= 0

 
the nsteps is still -1, did I do something wrong?


You haven't properly commented out the "nsteps = -1" line.  The comment sign is 
a ';' not '#' though I'm surprised grompp didn't simply abort with this input file.


-Justin

 
Thanks a lot
 
YY


--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

RE: [gmx-users] ./mdrun

[lin hen]

 constraints =  all-bonds
integrator  =  md
dt  =  0.002; ps !
#nsteps  =  -1
nsteps  =  100
nstlist =  0
ns_type =  grid
rlist   =  0
coulombtype =  cut-off
vdwtype =  cut-off
rcoulomb=  0
rvdw=  0
pbc =  no
epsilon_rf  =  0
rgbradii=  0
comm_mode   = angular
implicit_solvent= GBSA
gb_algorithm= OBC
gb_epsilon_solvent  = 78.3
sa_surface_tension  = 2.25936
nstxout = 0
nstfout = 0
nstvout = 0
nstxtcout   = 0
nstlog  = 0
nstcalcenergy   = -1
nstenergy   = 0
tcoupl  = berendsen
tc-grps = system
tau-t   = 0.1
ref-t   = 300

This is the .mdf file, even I modified the nsteps, it still comes with the same 
error:
 

Back Off! I just backed up ener.edr to ./#ener.edr.1#
 
WARNING: OpenMM does not support leap-frog, will use velocity-verlet integrator.
 

WARNING: OpenMM supports only Andersen thermostat with the md/md-vv/md-vv-avek 
integrators.
 

WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE and CCMA. 
Accuracy is based on the SHAKE tolerance set by the "shake_tol" option.
 

WARNING: The OBC scale factors alpha, beta and gamma are hardcoded in OpenMM 
with the default Gromacs values.
 

Pre-simulation ~15s memtest in progress...done, no errors detected
starting mdrun 'Protein'
-1 steps, infinite ps.

 
the log file:
Input Parameters:
   integrator   = md
   nsteps   = -1
   init_step= 0
   ns_type  = Grid
   nstlist  = 0
   ndelta   = 2
   nstcomm  = 10
   comm_mode= Angular
   nstlog   = 0
   nstxout  = 0
   nstvout  = 0
   nstfout  = 0
   nstcalcenergy= 10
   nstenergy= 0

 
the nsteps is still -1, did I do something wrong?
 
Thanks a lot
 
YY



From: rol...@utk.edu
Date: Tue, 9 Nov 2010 21:34:34 -0500
Subject: Re: [gmx-users] ./mdrun
To: gmx-users@gromacs.org

Hi,


as you output says. It infinite long trajectory. Either set a number of steps 
(in mdp file or with tpbconf) or set a maximum time with (mdrun -maxh)


Roland


On Tue, Nov 9, 2010 at 9:03 PM, lin hen  wrote:


Hi,
 
I am using the dhfr gpu benchmark with mdrun-gpu:dhfr-solv-RF-2nm.bench.
export LD_LIBRARY_PATH=/path to OPENMM lib and cuda lib/
export OPENMM_PLUGIN=/path to OPENMM lib plugins/
it shows:
Back Off! I just backed up md.log to ./#md.log.1#
Reading file topol.tpr, VERSION 4.5.1-dev-20100917-b1d66 (single precision)
Back Off! I just backed up ener.edr to ./#ener.edr.1#
WARNING: OpenMM does not support leap-frog, will use velocity-verlet integrator.

WARNING: OpenMM supports only Andersen thermostat with the md/md-vv/md-vv-avek 
integrators.

WARNING: OpenMM provides contraints as a combination of SHAKE, SETTLE and CCMA. 
Accuracy is based on the SHAKE tolerance set by the "shake_tol" option.

Pre-simulation ~15s memtest in progress...done, no errors detected
starting mdrun 'Protein in water'
-1 steps, infinite ps.

 
and keeps this status for at least one hour, did I miss something? or do 
somthing wrong?
 
thanks,
 
yy

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists


-- 
ORNL/UT Center for Molecular Biophysics cmb.ornl.gov
865-241-1537, ORNL PO BOX 2008 MS6309

-- gmx-users mailing list gmx-users@gromacs.org 
http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive 
at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! Please 
don't post (un)subscribe requests to the list. Use the www interface or send it 
to gmx-users-requ...@gromacs.org. Can't post? Read 
http://www.gromacs.org/Support/Mailing_Lists
   -- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] What can we do with gromacs?

[Justin A. Lemkul]

mustafa bilsel wrote:

Hi,
I would like to ask questions about what we can do with Gromacs.
3. For example I have some different molecules and I would like to see 
their binding affinity to carbon nanotubes. Is it possible to do this 
simulation with Gromacs? If possible, I should have attractive and 
repulsive potentials. Which potential(s) in Gromacs should I use for 
this purpose? 


This question is better posed as a parameterization issue.  I don't know which 
of the force field Gromacs has that would accommodate such a venture, but if 
there is literature precedent, I would start there.  Be advised that 
parameterization is an advanced topic, and you should be prepared to invest more 
time in deriving parameters than in actually collecting data.


http://www.gromacs.org/Documentation/How-tos/Parameterization

2. Is it possible to simulate folding mechanism of a protein in closed 
geometry? closed geometry maybe in cylindrical or cubic shape.


Anything is possible, but the time frame is probably a limiting factor.  Protein 
folding occurs on a time scale that exceeds most simulations.  Small model 
peptides can be done, but even then a significant amount of data needs to be 
collected.  Whether or not you can achieve an exact configuration depends 
entirely upon the quality of the force field model, simulation parameters, and 
extent of sampling.



3. Can we check if a protein adhesed or not a to a surface? surface 
consists of atoms.




Sure.

Gromacs is a very versatile program capable of doing pretty much anything you 
tell it.  Therein lies the problem - whether or not self-consistent force fields 
exist to describe heterogeneous and/or non-biological systems as you've 
described.  Certainly people have worked on such areas, so I would advise you to 
do some extensive background reading and planning.  It will save you a lot of 
time down the road.


-Justin



Best wishes
Mustafa Bilsel



--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] What can we do with gromacs?

[mustafa bilsel]

Hi,
I would like to ask questions about what we can do with Gromacs.
3. For example I have some different molecules and I would like to see their
binding affinity to carbon nanotubes. Is it possible to do this simulation
with Gromacs? If possible, I should have attractive and repulsive
potentials. Which potential(s) in Gromacs should I use for this purpose?
2. Is it possible to simulate folding mechanism of a protein in closed
geometry? closed geometry maybe in cylindrical or cubic shape.
3. Can we check if a protein adhesed or not a to a surface? surface consists
of atoms.


Best wishes
Mustafa Bilsel
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] virtual site error

[Ting Zhou]

Hi,

when I run
pdb2gmx -v -f 2KO3.pdb -o initial-vacuum.gro -ff oplsaa -water tip4p
-vsite hydrogens -ignh ,
I got
Fatal error:
Invalid directive HISD in vsite database
/home/XXX/gromacs/g/share/gromacs/top/oplsaa.ff/aminoacids.vsd .
What is the problem? 2KO3.pdb was downloaded from PDB.

Best regards,
Ting
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] mdrun -rerun: bonded interactions of protein

[Mark Abraham]

On 11/11/2010 7:46 PM, NG HUI WEN wrote:


Thanks a lot Justin and Mark for your useful input.

Indeed Justin was right, the quest to dissect the total energy of the 
system to get that contributed by the protein alone was not trivial at 
all!




Indeed. If I'd observed you were doing PME, I'd have made the same point.

I missed out this thread yesterday 
http://www.mail-archive.com/gmx-users@gromacs.org/msg34610.html as 
well as Mark's trick to decompose bonded terms by hacking the .top 
file. (However, I read from 
http://www.gromacs.org/Documentation/Gromacs_Utilities/g_energy that 
the Coul-recip term is not decomposable regardless of the tricks used...)


Mark, following your advice, I added --nsteps 0 (-1 didn't work) to 
tpbconv. I managed to get the interactive prompt succesfully this time.




Yes, nsteps=-1 is a GROMACS 4.5-ism, sorry.


Reading toplogy and shit from rerun1.tpr
Reading file rerun1.tpr, VERSION 4.0.7 (single precision)
Setting nsteps to 0
Group 0 (  System) has 100581 elements
Group 1 ( Protein) has  4002 elements
Group 2 (   Protein-H) has  3145 elements
Group 3 ( C-alpha) has   412 elements
Group 4 (Backbone) has  1236 elements
Group 5 (   MainChain) has  1649 elements
Group 6 (MainChain+Cb) has  2027 elements
Group 7 ( MainChain+H) has  2039 elements
Group 8 (   SideChain) has  1963 elements
Group 9 ( SideChain-H) has  1496 elements
Group10 ( Prot-Masses) has  4002 elements
Group11 ( Non-Protein) has 96579 elements
Group12 (POPE) has 13052 elements
Group13 ( SOL) has 83523 elements
Group14 ( CL-) has 4 elements
Group15 (   Other) has 96579 elements
Group16 ( SOL_CL-) has 83527 elements
Group17 (Protein_POPE) has 17054 elements
Select a group:

As suggested, I have also tried to compare the (average) values of 
subset.tpr and fullsystem.tpr, these are the total energies of:


1)System=  -1.28209 e+06

2)Protein=  -9571.86

3)Lipid=-296121

4)SOL_CL=  -821353

5)Other= -1.22684e+06

It was found that the:

1)Total energies (Protein + lipid + SOL_CL-) not equal to total energy 
of the system


2)Total energies of (Lipid + SOL_CL-) not equal to total energy of other

3)Total energies of (others + protein) not equal to total energy of 
the system


I have yet to work out the reasons for the discrepancies, will spend 
some time pondering and trying to make sense of these (and other) values.




Aren't you omitting the inter-group non-bonded terms? Anyway, I was 
suggesting this comparison for seeing whether the bonded components can 
be decomposed group-wise. It is already known that the non-bonded 
decomposition works.


Mark


Thanks a bunch again!

HW

*From:*gmx-users-boun...@gromacs.org 
[mailto:gmx-users-boun...@gromacs.org] *On Behalf Of *Mark Abraham

*Sent:* Wednesday, November 10, 2010 9:08 PM
*To:* Discussion list for GROMACS users
*Subject:* Re: [gmx-users] mdrun -rerun: bonded interactions of protein

On 10/11/2010 11:29 PM, NG HUI WEN wrote:

Hi Gmxusers,

I have been trying to run mdrun --rerun to get the energy of the 
protein in my protein-lipid system. I know similar questions have been 
raised on this topic before, I have tried to glean useful information 
from them to solve my problem but unfortunately to no avail. Thanks 
for your patience!


As I did not have energygrps in the initial .mdp file, I included it 
this time


integrator  = md
nsteps  = 0
dt  = 0.002
nstxout = 5
nstvout = 5
nstenergy   = 500
nstlog  = 500
Continuation  = yes
constraint_algorithm = lincs
constraints = all-bonds
lincs_iter  = 1
lincs_order = 4
ns_type   = grid
nstlist   = 5

rlist   = 1.2
rcoulomb= 1.2
rvdw= 1.2
coulombtype = PME
pme_order   = 4
fourierspacing= 0.16
tcoupl= Nose-Hoover
tc-grps   = Protein POPESOL_CL-

tau_t   = 0.1 0.1   0.1
ref_t   = 323   323   323
pcoupl= Parrinello-Rahman
pcoupltype  = semiisotropic

tau_p   = 5.0
ref_p   = 1.0 1.0
compressibility = 4.5e-5  4.5e-5
pbc = xyz
DispCorr= EnerPres
gen_vel   = no
nstcomm = 1
comm-mode   = Linear
comm-grps   = Protein_POPE SOL_CL-
*energygrps  = Protein SOL POPE*

I then did

1)grompp --f new.mdp --n index.ndx --c old.tpr --o rerun.tpr --p topol.top

2)trjconv --f old.trr --n index.ndx --s rerun.tpr --o rerun.trr (when 
prompted, I selected "0" system)


3)mdrun --s rerun.tpr --rerun rerun.trr

I notice that the previous post 
http://oldwww.gromacs.org/pipermail/gmx-users/2009-January/038968.html 
suggested to use tpbconv (on rerun.tpr) and trjconv (on rerun.trr) to 
extract the protein only. While it was possible to do so with trjconv, 
it wasn't feasible with tpbconv (I'm using gromacs 4.0.7) -- I might 
have missed out something as I did not get any prompt/out

Re: [gmx-users] Translation motion during MD

[Justin A. Lemkul]

nahren manuel wrote:

Dear Gromacs Users,

Yes Justin I totally agree with you and your point is well taken.  But 
given the fact that if i dont restrain the protein, they rotate and I 
dont get the formation of clusters.




If you know the manner in which the clusters should form, then isn't this just a 
problem of poor initial orientation?


Yes, i do want to protein to rotate in the plane of the membrane, but 
not the rotation that membrane-proximal become membrane -distal (this 
flipping rotation, if I may call like that)




This sounds like a problem.  If you have an ectodomain with a known structure, 
then the transmembrane portion shouldn't be too complex, right?  If there's an 
anchoring transmembrane helix, it is trivial to construct and add onto the 
structure.


Another alternate I was thinking was to do a steered MD, but since there 
is no clue which part (which domain in the protein) form cluster, this 
method doesn't help either.




I guess this rebuts my comment above...

Is  applying semiisotropic pressure a good idea ? similar to what one 
would do in membrane simulation.




I don't see why not.  Membranes should deform independently in x/y and z, so 
semiisotropic is fine.


-Justin


Best,
nahren

--- On *Thu, 11/11/10, Justin A. Lemkul //* wrote:


From: Justin A. Lemkul 
Subject: Re: [gmx-users] Translation motion during MD
To: "Discussion list for GROMACS users" 
Date: Thursday, November 11, 2010, 4:14 PM



nahren manuel wrote:
 > Dear Gromacs Users,
 >
 > I am simulating a Membrane protein, the extracellular domain
alone (since the structure of only extracellular domain is solved).
So I will have to simulate the protein in such a way that only the
translational motion is allowed but the rotational motions are
prevented (which would mimic the behavior of the protein attached to
the membrane).
 >

I don't understand how you conclude that your protein shouldn't
rotate.  Even if the transmembrane portion was there, the protein
could certainly rotate in the plane of the membrane.

 > The protein is expected to former dimer and multimers on the cell
surface, so I want to restrict motion only to 2D. This would give an
idea as to how the protein clusters on the membrane surface
(studying this multimerization looks too ambitious, but want to make
an attempt).
 >

I suppose you could place a weak position restraint on all atoms in
the z-dimension only.  I would seriously question the validity of
doing so, though.  If you force a system into a preconceived
behavior that masks other missing information, I'd say you're just
causing the events to happen, not allowing them to occur naturally. 
You're also causing unnatural forces between the membrane and the

protein.  If the membrane deforms or undulates, the protein cannot
accommodate this change.  This is true no matter how you restrict
this 2-D motion, and I think it would be a serious problem.

-Justin

 > Best,
 > nahren
 >
 >
 >
 >
 >
 >
 >

-- 

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


-- gmx-users mailing listgmx-users@gromacs.org

http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the www
interface or send it to gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists




--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] g_hbond

[Erik Marklund]

Hi,

I have looked at the files to sent me, and I can't spot any errors. 
There *is* a hydrogen bond at y=1. It only occurs in 5 frames though, so 
they're really hard to find when isnpecting the xpm in e.g. Gimp, since 
the image is 6 pixels wide. So, I got tired of sqinting in front of 
Gimp and went for a bit of comandline work instead. Here's what I did:


>> tail -1 map_concatenated.xpm | tr 'o' '\n' > lastline
>> head -1 lastline | wc -c
   16749
>> head -2 lastline | wc -c
   16936
>> head -3 lastline | wc -c
   40226
>> head -4 lastline | wc -c
   45019
>> head -5 lastline | wc -c
   58565
>> head -6 lastline | wc -c
   60001

The numbers I got (except for the last one) are the frames where the 
hbond manifests. I verified the first one in gimp. You can check the 
other's if you wish. I also found two frames in map1.xpm and three in 
map3.xpm using the same method.


Erik


Carla Jamous skrev 2010-11-04 11.32:

Hi Erik,

I tried what you said: I made index groups containing only the atoms 
involved in that hbond and ran g_hbond again. The problem didn't persist.
So I wanted to check if I misinterpreted the map: for that, I compared 
my 4 index files: the one of my concatenated trajectory and the three 
of the separate trajectories.
When I look at this comparison table and then at my concatenated hbond 
map, the results don't match: for example, the Hbond that has the 
number 1 in my concatenated trajectory doesn't appear in my 
concatenated map (it means that in my map, next to y=1, nothing 
appears), however, this same H-bond (same number of atoms involved) 
exists in my first and my third trajectories' index files. So please 
could you explain how to interpret the map and if next to some y value 
in my map, nothing appears, is this value taken into account in my 
index file?


Thank you,
Carla

On Tue, Nov 2, 2010 at 8:36 PM, Erik Marklund > wrote:


Erik Marklund skrev 2010-11-02 20.33:

Carla Jamous skrev 2010-11-02 17.35:

Hi to all,

I'm sorry I'm asking again a question I asked a week ago
but I still haven't found my answer:

I concatenated 3 trajectories of 3 different molecules
(that have the same number of atoms) with trjcat: trjcat
-settime

Then I ran g_hbond on the concatenated trajectory, I got
an index.ndx file that contains an H-bond between the
Thr121 of my protein and an atom N2 of my ligand. This
H_bond  figures in the 3 trajectories when I look at the
hbmap of my concatenated trajectory.

On the other hand, I ran g_hbond on each of the 3
different trajectories. This H_bond doesn't exist in the
index files of two of my trajectories. So it doesn't match
the result I get with my concatenated trajectory.

To be sure that this result is right, I calculated the
angle and the r of my H-bond in VMD during each of the 3
trajectories and the results indicate that this H-bond
doesn't exist (I consider that r must be < or equal to
0.35 nm and angle < or equal to 30 degrees).

Please can anyone tell me why the results of g_hbond in my
concatenated trajectory don't match the results of g_hbond
on each of my trajectories?

Thank you

Carla

Hi,

This sounds strange. Could you file a bugzilla and either
upload the trajectories+tpr, or, if the files are huge, make
them available my other means? I need to have a closer look,
perhaps through a debugger.

A simple test you could do yourself is to make index groups
containing only the atoms involved in that hbond and run g_hbond
again. If the problem persists, then it looks like a bug. If not,
then I still can't rule out misinterpretation of the matrix.

Erik


-- 
---

Erik Marklund, PhD student
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,75124 Uppsala, Sweden
phone:+46 18 471 4537fax: +46 18 511 755
er...@xray.bmc.uu.se 
http://folding.bmc.uu.se/

-- 
gmx-users mailing list gmx-users@gromacs.org


http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the www
interface or send it to gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists





--
---
Erik Marklund, PhD student
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,751

Re: [gmx-users] g_hbond

[Carla Jamous]

Hi everyone,

actually Erik is right. My problem is that I wasn't using the "-nomerge"
option.
So g-hbond was printing the wrong number of hydrogen atom and so all my
results were not matching.
Now that I used "-nomerge", all my H-bonds match even in my concatenated
trajectory.

Thanks to all,
It took me a long time to figure it out, thanks for your patience,
Carla

On Thu, Nov 11, 2010 at 4:26 PM, Erik Marklund  wrote:

> Hi,
>
> I'm thinking it *may* have something to do with how search_donors() and the
> -merge flag work. The hydrogen must be part of the index group you provide.
> Nothing strange there. But the hydrogen that is printed to the index file
> may in fact *not* be the hydrogen bonding one if the -merge flag is set
> (default). Would the vanishing h-bonds happen to have donors with multiple
> hydrogens? There are two things you could try: either add all the donor's
> hydrogen to the index group when validating single h-bonds, or use -nomerge
> to generate a complete list of h-bonds and do your validation on that list.
>
> Erik
>
> Carla Jamous skrev 2010-11-09 15.08:
>
>  Hi everyone,
>>
>> I ran g_hbond on a trajectory. When g_hbond asks for two groups in the
>> index file, I give:
>> group1: "Protein"
>> group2: "ligand"
>> The output .ndx file contains 44 Hbonds.
>>
>> In order to verify my result on each and every Hbond, I ran g_hbond on the
>> same trajectory but this time in my index file, I gave atom triplets, so
>> when g_hbond asks for two groups, I give for example:
>> group1: "a4810_a4811_a1857"
>> group2: "a4810_a4811_a1857"
>> Surprisingly, 10 of the 44 Hbonds are not found using this method.
>>
>> Am I doing something wrong or is there a problem with g_hbond?
>>
>> Thank you,
>>
>> Carla
>>
>
>
> --
> ---
> Erik Marklund, PhD student
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,75124 Uppsala, Sweden
> phone:+46 18 471 4537fax: +46 18 511 755
> er...@xray.bmc.uu.sehttp://folding.bmc.uu.se/
>
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Translation motion during MD

[nahren manuel]

Dear Gromacs Users,

Yes Justin I totally agree with you and your point is well taken.  But given 
the fact that if i dont restrain the protein, they rotate and I dont get the 
formation of clusters.

Yes, i do want to protein to rotate in the plane of the membrane, but 
not the rotation that membrane-proximal become membrane -distal (this flipping 
rotation, if I may call like that)

Another alternate I was thinking was to do a steered MD, but since there is no 
clue which part (which domain in the protein) form cluster, this method doesn't 
help either.

Is  applying semiisotropic pressure a good idea ? similar to what one would do 
in membrane simulation.

Best,
nahren

--- On Thu, 11/11/10, Justin A. Lemkul  wrote:

From: Justin A. Lemkul 
Subject: Re: [gmx-users] Translation motion during MD
To: "Discussion list for GROMACS users" 
Date: Thursday, November 11, 2010, 4:14 PM



nahren manuel wrote:
> Dear Gromacs Users,
> 
> I am simulating a Membrane protein, the extracellular domain alone (since the 
> structure of only extracellular domain is solved). So I will have to simulate 
> the protein in such a way that only the translational motion is allowed but 
> the rotational motions are prevented (which would mimic the behavior of the 
> protein attached to the membrane).
> 

I don't understand how you conclude that your protein shouldn't rotate.  Even 
if the transmembrane portion was there, the protein could certainly rotate in 
the plane of the membrane.

> The protein is expected to former dimer and multimers on the cell surface, so 
> I want to restrict motion only to 2D. This would give an idea as to how the 
> protein clusters on the membrane surface (studying this multimerization looks 
> too ambitious, but want to make an attempt).
> 

I suppose you could place a weak position restraint on all atoms in the 
z-dimension only.  I would seriously question the validity of doing so, 
though.  If you force a system into a preconceived behavior that masks other 
missing information, I'd say you're just causing the events to happen, not 
allowing them to occur naturally.  You're also causing unnatural forces between 
the membrane and the protein.  If the membrane deforms or undulates, the 
protein cannot accommodate this change.  This is true no matter how you 
restrict this 2-D motion, and I think it would be a serious problem.

-Justin

> Best,
> nahren
> 
> 
> 
> 
> 
> 
> 

-- 

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


-- gmx-users mailing list    gmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the www interface or 
send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists



  -- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] When will FEP be on GPU?

[TJ Mustard]

  

  
Hi all,

 

Our group is doing FEP calculations and finding that we need to run alot of simulations for a long time to get accurate results. We would love to see FEP on GPU as this would help us increase our computational power without buying a new expensive cluster. I would be happy to be a/the guinea pig on this one.

 

Thank you,

TJ Mustard
Email: musta...@onid.orst.edu
  

-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] g_hbond

[Erik Marklund]

Hi,

I'm thinking it *may* have something to do with how search_donors() and 
the -merge flag work. The hydrogen must be part of the index group you 
provide. Nothing strange there. But the hydrogen that is printed to the 
index file may in fact *not* be the hydrogen bonding one if the -merge 
flag is set (default). Would the vanishing h-bonds happen to have donors 
with multiple hydrogens? There are two things you could try: either add 
all the donor's hydrogen to the index group when validating single 
h-bonds, or use -nomerge to generate a complete list of h-bonds and do 
your validation on that list.


Erik

Carla Jamous skrev 2010-11-09 15.08:

Hi everyone,

I ran g_hbond on a trajectory. When g_hbond asks for two groups in the 
index file, I give:

group1: "Protein"
group2: "ligand"
The output .ndx file contains 44 Hbonds.

In order to verify my result on each and every Hbond, I ran g_hbond on 
the same trajectory but this time in my index file, I gave atom 
triplets, so when g_hbond asks for two groups, I give for example:

group1: "a4810_a4811_a1857"
group2: "a4810_a4811_a1857"
Surprisingly, 10 of the 44 Hbonds are not found using this method.

Am I doing something wrong or is there a problem with g_hbond?

Thank you,

Carla



--
---
Erik Marklund, PhD student
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,75124 Uppsala, Sweden
phone:+46 18 471 4537fax: +46 18 511 755
er...@xray.bmc.uu.sehttp://folding.bmc.uu.se/

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Translation motion during MD

[Justin A. Lemkul]

nahren manuel wrote:

Dear Gromacs Users,

I am simulating a Membrane protein, the extracellular domain alone 
(since the structure of only extracellular domain is solved). So I will 
have to simulate the protein in such a way that only the translational 
motion is allowed but the rotational motions are prevented (which would 
mimic the behavior of the protein attached to the membrane).




I don't understand how you conclude that your protein shouldn't rotate.  Even if 
the transmembrane portion was there, the protein could certainly rotate in the 
plane of the membrane.


The protein is expected to former dimer and multimers on the cell 
surface, so I want to restrict motion only to 2D. This would give an 
idea as to how the protein clusters on the membrane surface (studying 
this multimerization looks too ambitious, but want to make an attempt).




I suppose you could place a weak position restraint on all atoms in the 
z-dimension only.  I would seriously question the validity of doing so, though. 
 If you force a system into a preconceived behavior that masks other missing 
information, I'd say you're just causing the events to happen, not allowing them 
to occur naturally.  You're also causing unnatural forces between the membrane 
and the protein.  If the membrane deforms or undulates, the protein cannot 
accommodate this change.  This is true no matter how you restrict this 2-D 
motion, and I think it would be a serious problem.


-Justin


Best,
nahren









--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] g_hbond

[Erik Marklund]

Oliver Grant skrev 2010-11-11 11.45:

Hi Carla,

I've seen this behavior too but I didn't look at the code. I figured 
with the hydrogen included it checks the angle too or if the hydrogen 
is between the two heavy atoms. So the first pass just looks at 
distances between donor and acceptor atoms and doesn't consider that 
the hydrogen may be completely on the wrong side of the donor atom.



It always looks at the angle unless you use the -contact flag.

/Erik

Did you look at any of the bonds that were left out in the 2nd pass?

Oliver

On 9 November 2010 14:08, Carla Jamous > wrote:


Hi everyone,

I ran g_hbond on a trajectory. When g_hbond asks for two groups in
the index file, I give:
group1: "Protein"
group2: "ligand"
The output .ndx file contains 44 Hbonds.

In order to verify my result on each and every Hbond, I ran
g_hbond on the same trajectory but this time in my index file, I
gave atom triplets, so when g_hbond asks for two groups, I give
for example:
group1: "a4810_a4811_a1857"
group2: "a4810_a4811_a1857"
Surprisingly, 10 of the 44 Hbonds are not found using this method.

Am I doing something wrong or is there a problem with g_hbond?

Thank you,

Carla

--
gmx-users mailing list gmx-users@gromacs.org

http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists





--
---
Erik Marklund, PhD student
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,75124 Uppsala, Sweden
phone:+46 18 471 4537fax: +46 18 511 755
er...@xray.bmc.uu.sehttp://folding.bmc.uu.se/

-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] Translation motion during MD

[nahren manuel]

Dear Gromacs Users,

I am simulating a Membrane protein, the extracellular domain alone (since the 
structure of only extracellular domain is solved). So I will have to simulate 
the protein in such a way that only the translational motion is allowed but the 
rotational motions are prevented (which would mimic the behavior of the protein 
attached to the membrane). 

The protein is expected to former dimer and multimers on the cell surface, so I 
want to restrict motion only to 2D. This would give an idea as to how the 
protein clusters on the membrane surface (studying this multimerization looks 
too ambitious, but want to make an attempt).

Best,
nahren









  -- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] New release of Wordom, molecular analysis program

[michele seeber]

Hi all,

a new version of Wordom, a program for the analysis of molecular structures,
trajectories, and free energy surfaces, has been released. You can find all
about it at:
http://wordom.sf.net
Wordom can deal with Gromacs and Charmm trajectories and, besides analysis
modules, has many features to make handling xtc, dcd, crd and pdb files
easier and faster.
Wordom is released under the GPL, so it is free to use, modify and
redistribute.

Best regards,

--
Michele Seeber
Dulbecco Telethon Institute c/o
Univ. of Modena and Reggio Emilia
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Re: which force file has parameters for creatine md simulations?

[Justin A. Lemkul]

Olga Ivchenko wrote:

Dear All,

I just want to know if the programm PRODRG is able to generate for my 
ligand files with .top extension.




No, but it is trivial to interconvert .itp and .top manually:

http://www.gromacs.org/Documentation/File_Formats/.itp_File

-Justin


I have two files one is itp format and other is gro using PRODRG.
If it is possible what is the name here 
:http://davapc1.bioch.dundee.ac.uk/cgi-bin/prodrg_beta#DRGFIN.GRO


Yours sincerely,
Olga

2010/11/10 Justin A. Lemkul mailto:jalem...@vt.edu>>



Olga Ivchenko wrote:

Dear All,

I have created creating topology file using  *prodrg_beta (using
gromos 96 ff)
*. When I
am trying to run energy minimization it writes an error:


Please consider carefully the caveats about PRODRG noted here
(second paragraph):

http://www.gromacs.org/Downloads/Related_Software/PRODRG#Tips


Syntax error - File creatine.top, line 19
Last line read:
'[ moleculetype ]'
Invalid order for directive moleculetype

I looked at previoust mailing lists according this problem. I
can not find direct answer.



http://www.gromacs.org/Documentation/Errors#Invalid_order_for_directive_defaults

Somehow you've cut and pasted such that you've got a [moleculetype]
directive out of order.


I tried to add this line in the topology file

#include "ffG43a1.ff/DRGGMX.itp"

An again an error:Fatal error:
Topology include file "ffG43a1.ff/DRGGMX.itp" not found


Then either the file doesn't exist or isn't within the search path
grompp is trying.

-Justin



Please can you advice me on this.


Yours sincerely,

Olga





-- 



Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu  | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


-- 
gmx-users mailing listgmx-users@gromacs.org


http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the www
interface or send it to gmx-users-requ...@gromacs.org
.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists




--


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] hydration maps

[babu gokul]

Dear all
I would like to analysis the hydration site near the proteins the one done by 
the following paper

"Biophysical Journal Volume 79 December 2000 2966–2974"

is there any tool available in gromacs to do this kind of analysis could anyone 
help me in this regard. what tool will be useful to analysis this kind of 
results.

Regards 
E R Azhagiya singam 


-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Re: which force file has parameters for creatine md simulations?

[Olga Ivchenko]

Dear All,

I just want to know if the programm PRODRG is able to generate for my ligand
files with .top extension.

I have two files one is itp format and other is gro using PRODRG.
If it is possible what is the name here :
http://davapc1.bioch.dundee.ac.uk/cgi-bin/prodrg_beta#DRGFIN.GRO

Yours sincerely,
Olga

2010/11/10 Justin A. Lemkul 

>
>
> Olga Ivchenko wrote:
>
>> Dear All,
>>
>> I have created creating topology file using  *prodrg_beta (using gromos 96
>> ff) *. When I am
>> trying to run energy minimization it writes an error:
>>
>>
> Please consider carefully the caveats about PRODRG noted here (second
> paragraph):
>
> http://www.gromacs.org/Downloads/Related_Software/PRODRG#Tips
>
>
>  Syntax error - File creatine.top, line 19
>> Last line read:
>> '[ moleculetype ]'
>> Invalid order for directive moleculetype
>>
>> I looked at previoust mailing lists according this problem. I can not find
>> direct answer.
>>
>>
>
> http://www.gromacs.org/Documentation/Errors#Invalid_order_for_directive_defaults
>
> Somehow you've cut and pasted such that you've got a [moleculetype]
> directive out of order.
>
>
>  I tried to add this line in the topology file
>>
>> #include "ffG43a1.ff/DRGGMX.itp"
>>
>> An again an error:Fatal error:
>> Topology include file "ffG43a1.ff/DRGGMX.itp" not found
>>
>>
> Then either the file doesn't exist or isn't within the search path grompp
> is trying.
>
> -Justin
>
>
>
>> Please can you advice me on this.
>>
>>
>> Yours sincerely,
>>
>> Olga
>>
>>
>>
>>
>>
> --
> 
>
> Justin A. Lemkul
> Ph.D. Candidate
> ICTAS Doctoral Scholar
> MILES-IGERT Trainee
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>
> 
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the www interface
> or send it to gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] g_hbond

[Oliver Grant]

Hi Carla,

I've seen this behavior too but I didn't look at the code. I figured with
the hydrogen included it checks the angle too or if the hydrogen is between
the two heavy atoms. So the first pass just looks at distances between donor
and acceptor atoms and doesn't consider that the hydrogen may be completely
on the wrong side of the donor atom.

Did you look at any of the bonds that were left out in the 2nd pass?

Oliver

On 9 November 2010 14:08, Carla Jamous  wrote:

> Hi everyone,
>
> I ran g_hbond on a trajectory. When g_hbond asks for two groups in the
> index file, I give:
> group1: "Protein"
> group2: "ligand"
> The output .ndx file contains 44 Hbonds.
>
> In order to verify my result on each and every Hbond, I ran g_hbond on the
> same trajectory but this time in my index file, I gave atom triplets, so
> when g_hbond asks for two groups, I give for example:
> group1: "a4810_a4811_a1857"
> group2: "a4810_a4811_a1857"
> Surprisingly, 10 of the 44 Hbonds are not found using this method.
>
> Am I doing something wrong or is there a problem with g_hbond?
>
> Thank you,
>
> Carla
>
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] hblife.xvg and hbac.xvg files

[shahab shariati]

Dear gromacs users

I used g_hbond -f .trr -s .tpr -n .ndx -ac -life
can anyone clarify last 2 columns in hblife.xvg and last 4 columns hbac.xvg
files by details.
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] difference between p(t) and c(t)

[shahab shariati]

Dear all

I want to know what is difference between p(t) in hblife.xvg and c(t) in
hbac.xvg file?

which of them is more suitable for HB lifetime?
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] on calculation of the atomic covariance

[Tsjerk Wassenaar]

Hi Vignesh,

If your covariances show different ranges, isn't that a difference
between your systems, wild-type and mutated? Then again, there's also
noise in the covariances (noise in the fluctuations, ergo noise in the
noise ;)). The rest might be comparable, making scaling based on the
extremes seem like a bad idea. You might be better of calculating
correlations, rather than covariances. There's a modified g_covar in
the contributions section of the gromacs site for calculating those.

Cheers,

Tsjerk

On Thu, Nov 11, 2010 at 3:53 AM, Vigneshwar Ramakrishnan
 wrote:
> Dear All,
> I tried using the xpm2ps -combine option to plot the two matrices in the
> same plot.
> "xpm2ps -f covara_cognate.xpm -f2 covara_G5C.xpm -diag none -combine halves
> -cmin -0.5 -cmax 0.8"
>
> However, I still get two legends and each of the matrices are scaled
> differently. That is, the output range is NOT combined, as the -combine
> option is supposed to do. I tried different options (add, sub, div ; with
> and without the -cmin and -cmax options etc)
> I am sure I am missing something here. May I please know if anybody got it
> worked, and if so, can help me out?
> Thanks very much,
> Vignesh
> On Thu, Nov 11, 2010 at 10:12 AM, Vigneshwar Ramakrishnan
>  wrote:
>>
>> Thanks very much, Justin. Somehow this thread did not come up during my
>> search.
>> Really appreciate your help.
>> Sincerely,
>> Vignesh
>>
>> On Wed, Nov 10, 2010 at 10:23 PM, Justin A. Lemkul 
>> wrote:
>>>
>>> There are two relevant threads on this same topic that will likely
>>> provide some insight (particularly the second):
>>>
>>> http://lists.gromacs.org/pipermail/gmx-users/2009-November/046846.html
>>> http://lists.gromacs.org/pipermail/gmx-users/2009-November/046854.html
>>>
>>> -Justin
>>>
>>> Vigneshwar Ramakrishnan wrote:

 Dear All,
 I am trying to study the effect of single-point mutations on correlated
 motions in a protein-DNA system. I am able to calculate the atomic
 covariance matrix using the g_covar -xpma option. However, when I try to
 compare the covariance matrices for the two systems (to study the effect of
 the mutation), I find that the output is not scaled identically. That is, 
 in
 one of the systems the atomic covariance varies between -0.04 and +0.5
 whereas in the other it varies between -0.1 and +0.4. Now, this means that 
 I
 cannot compare the two systems immediately from the eps file output
 (obtained after xpm2ps).
 Could anybody please tell me if there is a way to plot the output on
 identical scales (say, -1 to +1, or any other scale) using GROMACS?
 The other way, I understand is to use the ascii output of g_covar and
 use the values to create the covariance plot using softwares like MATLAB
 which can rescale the image colors. However, for this, one needs to
 calculate the atomic covariance from the ascii output (which is
 x1x1,x1y1,x1z1 etc). From the manual, I understand that the way to 
 calculate
 atomic covariance is "for each atom pair the sum of the xx, yy and zz
 covariances". Am I right if I understand that this means:
 atomic cov (X1) = x1x1 + y1y1 + z1z1 atomic cov (X2) = x2x2 + y2y2 +
 z2z2 ...
 I greatly appreciate any help or pointers.
 Thanks very much, Sincerely, Vignesh


 --
 R.Vigneshwar
 Graduate Student,
 Dept. of Chemical & Biomolecular Engg,
 National University of Singapore,
 Singapore

 "Strive for Excellence, Never be satisfied with the second Best!!"

 I arise in the morning torn between a desire to improve the world and a
 desire to enjoy the world. This makes it hard to plan the day. (E.B. White)

>>>
>>> --
>>> 
>>>
>>> Justin A. Lemkul
>>> Ph.D. Candidate
>>> ICTAS Doctoral Scholar
>>> MILES-IGERT Trainee
>>> Department of Biochemistry
>>> Virginia Tech
>>> Blacksburg, VA
>>> jalemkul[at]vt.edu | (540) 231-9080
>>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>>
>>> 
>>> --
>>> gmx-users mailing list    gmx-us...@gromacs.org
>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>> Please don't post (un)subscribe requests to the list. Use the www
>>> interface or send it to gmx-users-requ...@gromacs.org.
>>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>>
>>
>> --
>> R.Vigneshwar
>> Graduate Student,
>> Dept. of Chemical & Biomolecular Engg,
>> National University of Singapore,
>> Singapore
>>
>> "Strive for Excellence, Never be satisfied with the second Best!!"
>>
>> I arise in the morning torn between a desire to improve the world and a
>> desire to enjoy the world. This makes it hard to plan the day. (E.B. White)
>
>
>
> --
> R.Vigneshwar
> Graduate Student,
> Dept. of Chemical & Biomolecular Engg,
> Nation

RE: [gmx-users] mdrun -rerun: bonded interactions of protein

[NG HUI WEN]

Thanks a lot Justin and Mark for your useful input.

 

Indeed Justin was right, the quest to dissect the total energy of the
system to get that contributed by the protein alone was not trivial at
all! 

 

I missed out this thread yesterday
http://www.mail-archive.com/gmx-users@gromacs.org/msg34610.html as well
as Mark's trick to decompose bonded terms by hacking the .top file.
(However, I read from
http://www.gromacs.org/Documentation/Gromacs_Utilities/g_energy that the
Coul-recip term is not decomposable regardless of the tricks used...)

 

Mark, following your advice, I added -nsteps 0 (-1 didn't work) to
tpbconv. I managed to get the interactive prompt succesfully this time. 

 

Reading toplogy and shit from rerun1.tpr
Reading file rerun1.tpr, VERSION 4.0.7 (single precision)
Setting nsteps to 0
Group 0 (  System) has 100581 elements
Group 1 ( Protein) has  4002 elements
Group 2 (   Protein-H) has  3145 elements
Group 3 ( C-alpha) has   412 elements
Group 4 (Backbone) has  1236 elements
Group 5 (   MainChain) has  1649 elements
Group 6 (MainChain+Cb) has  2027 elements
Group 7 ( MainChain+H) has  2039 elements
Group 8 (   SideChain) has  1963 elements
Group 9 ( SideChain-H) has  1496 elements
Group10 ( Prot-Masses) has  4002 elements
Group11 ( Non-Protein) has 96579 elements
Group12 (POPE) has 13052 elements
Group13 ( SOL) has 83523 elements
Group14 ( CL-) has 4 elements
Group15 (   Other) has 96579 elements
Group16 ( SOL_CL-) has 83527 elements
Group17 (Protein_POPE) has 17054 elements
Select a group: 

 

As suggested, I have also tried to compare the (average) values of
subset.tpr and fullsystem.tpr, these are the total energies of:

1)  System=  -1.28209 e+06

2)  Protein=  -9571.86

3)  Lipid=-296121

4)  SOL_CL=  -821353

5)  Other= -1.22684e+06

It was found that the: 

1)  Total energies (Protein + lipid + SOL_CL-) not equal to total
energy of the system

2)  Total energies of (Lipid + SOL_CL-) not equal to total energy of
other

3)  Total energies of (others + protein) not equal to total energy
of the system

 

I have yet to work out the reasons for the discrepancies, will spend
some time pondering and trying to make sense of these (and other)
values.

 

Thanks a bunch again!

 

HW

 

 

From: gmx-users-boun...@gromacs.org
[mailto:gmx-users-boun...@gromacs.org] On Behalf Of Mark Abraham
Sent: Wednesday, November 10, 2010 9:08 PM
To: Discussion list for GROMACS users
Subject: Re: [gmx-users] mdrun -rerun: bonded interactions of protein

 

On 10/11/2010 11:29 PM, NG HUI WEN wrote: 

Hi Gmxusers,

 

I have been trying to run mdrun -rerun to get the energy of the protein
in my protein-lipid system. I know similar questions have been raised on
this topic before, I have tried to glean useful information from them to
solve my problem but unfortunately to no avail. Thanks for your
patience!  

 

As I did not have energygrps in the initial .mdp file, I included it
this time  

 

integrator  = md
nsteps  = 0   
dt  = 0.002   
nstxout = 5   
nstvout = 5   
nstenergy   = 500   
nstlog  = 500   
Continuation  = yes   
constraint_algorithm = lincs  
constraints = all-bonds 
lincs_iter  = 1 
lincs_order = 4 
ns_type   = grid
nstlist   = 5 

rlist   = 1.2   
rcoulomb= 1.2   
rvdw= 1.2   
coulombtype = PME   
pme_order   = 4 
fourierspacing= 0.16
tcoupl= Nose-Hoover   
tc-grps   = Protein POPESOL_CL- 

tau_t   = 0.1 0.1   0.1   
ref_t   = 323   323   323   
pcoupl= Parrinello-Rahman 
pcoupltype  = semiisotropic 

tau_p   = 5.0 
ref_p   = 1.0 1.0 
compressibility = 4.5e-5  4.5e-5  
pbc = xyz   
DispCorr= EnerPres  
gen_vel   = no
nstcomm = 1
comm-mode   = Linear
comm-grps   = Protein_POPE SOL_CL-
energygrps  = Protein SOL POPE

 

I then did 

1)grompp -f new.mdp -n index.ndx -c old.tpr -o rerun.tpr -p topol.top

2)trjconv -f old.trr -n index.ndx -s rerun.tpr -o rerun.trr (when
prompted, I selected "0" system)

3)mdrun -s rerun.tpr -rerun rerun.trr

 

I notice that the previous post
http://oldwww.gromacs.org/pipermail/gmx-users/2009-January/038968.html
suggested to use tpbconv (on rerun.tpr) and trjconv (on rerun.trr) to
extract the protein only. While it was possible to do so with trjconv,
it wasn't feasible with tpbconv (I'm using gromacs 4.0.7) - I might have
missed out something as I did not get any prompt/output (see below)

 

tpbconv -s topol.tpr -n index_P.ndx -o rerun2.tpr

 

Reading toplogy and shit from topol.tpr
Reading file