[gmx-users] Re: gmx-users Digest, Vol 80, Issue 90

[Chris Neale]

On 10-12-13 01:29 PM, gmx-users-request at gromacs.org wrote:
> ubject: [gmx-users] Question about COM-Pulling
> To: MAILINGLIST GROMACS<gmx-users at gromacs.org>
> Message-ID:<4D065C09.4020809 at rhrk.uni-kl.de>
> Content-Type: text/plain; charset=UTF-8; format=flowed
>
> Hi,
>
> although this is not relevant to my previous question:

It is indeed relevant.

> I used mdrun like this:
>
> mdrun -v -s pull.tpr -o pull.trr -cpo pull.cpt -c pull.gro -e pull.edr
> -g pull.log -px pull_dist.xvg -pf pull_force.xvg
>
> With an input-file like this:
>
> ------------------------------------------------
> ; Run-Parameter
> integrator = md
> dt = 0.002
> nsteps = 1000000
>
> ; OUTPUT CONTROL OPTIONS
> nstxout = 5000
> nstvout = 5000
> nstfout = 5000
> nstlog = 5000
> nstenergy = 5000
>
> pull_nstxout = 100 ;
> pull_nstfout = 100 ;
>
> rlist = 1.2
> ns_type = grid
> coulombtype = PME-Switch
> rcoulomb = 1.0
> vdw-type = shift
> rvdw = 1.0
> rvdw_switch = 0.9
> pbc = xyz
>
> tcoupl = V-rescale
> tc-grps = Protein Non-Protein
> tau_t = 0.1 0.1
> ref_t = 300 300
>
> pcoupl = Berendsen
> pcoupltype = isotropic
> tau_p = 2
> ref_p = 1
> compressibility = 4.5e-5
>
> DispCorr = EnerPres
> continuation = yes
>
> constraint_algorithm = lincs
> constraints = all-bonds
>
> pull = umbrella
> pull_geometry = distance
> pull_dim = N N Y
> pull_start = yes
> pull_ngroups = 1
> pull_group0 = reference
> pull_group1 = pull
> pull_rate1 = 0.01
> pull_k1 = 1000
> ------------------------------------------------
>
> ...I get an output for pull_dist.xvg which looks sth. like this:
>
> ------------------------------------------------
> # This file was created Fri Nov 26 16:22:53 2010
> # by the following command:
> # /mnt/nas1/c_muecksch/GROMACS_4.5.3/bin/mdrun_mpi_4.5.3_s -v -s
> pull.tpr -o pull.trr -cpo pull.cpt -c pull.gro -e pull.edr -g pull.log
> -px pull_dist.xvg -pf pull_force.xvg
> #
> # mdrun_mpi_4.5.3_s is part of G R O M A C S:
> #
> # Grunge ROck MAChoS
> #
> @ title "Pull COM"
> @ xaxis label "Time (ps)"
> @ yaxis label "Position (nm)"
> @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 "0 Z"
> @ s1 legend "1 dZ"
> 0.0000 1.1675 3.47701
> 0.2000 1.1675 3.48043
> 0.4000 1.1675 3.4795
> 0.6000 1.1675 3.48214
> 0.8000 1.1675 3.4869
>
> and so on

Here you have time, the z position of your reference group and the 
difference along z between your pull group and your reference group. 
There is no "dummy particle" as the spring extends between the 
"reference" group at z=1.1675 and the "pull" group, which in the first 
frame is located at z=1.1675 + 3.47701.

I have used the pull code since v3.3.1 and I do not recall there ever 
being a dummy particle when using:

pull = umbrella
pull_geometry = distance

It is the distance of the restraint that changes over time in your setup.

Chris.



> ------------------------------------------------
>
> In this pull_dist.xvg there is no position of a "dummy"-particle that is
> connected via a harmonic potential to the pull group (as described in
> Grubmüller, Helmut ; Heymann, Berthold ; Tavan, Paul: Ligand Binding:
> Molecular Mechanics Calculation of the Streptavidin–Biotin Rupture
> Force. In: Science 271 (1996))
>
> In older version of Gromacs I believe that this was the case. Has the
> method changed? How is the force calculated if not by the displacment of the
> "dummy"-particle. I just wanted to know how the afm pulling in gromacs
> is achieved without having to look in the source-code.
>
> Thx,
> Christian
>
>
> ---------------------------------------------------------------------------------------
> Dear Christian:
>
> As per my original comments, please provide a .mdp file, sample
> output, and all the other things that I asked for.
>
> Chris.
>
> --original message--
>
> this is a more general question.
>
> I thought that in analogy to an AFM-experiment a "dummy"-particle is placed
> at a certain distance from the center of mass of the pull group (compare
> figure 6.1 in the manual).
> The force is then calculated as the displacement between the pulled atom
> and the
> "dummy"-particle. So if this is true which I'm not sure of,
> I was wondering why the position of this "dummy"-particle is not plotted.
>
> Or does this umbrella pulling work in a different way? How is the force
> calculated?
>
> Thanks a lot,
> Christian
>
> ----------------------------------------------------------------------
> The position of the reference group and the displacement of the pulled
> group from the reference group should be printed (although perhaps
> without the reference position if you are using absolute coordinate
> pulling, which is not recommended anyhow).
>
> If you want some better advice, please be more specific and include
> cut and paste sections saying "I expected X at Y, but it was not in
> the file Z.xvg". you should also include your pull-code .mdp options
> and your grompp and mdrun commands.
>
> Chris.
>
> -- original message --
>
> Dear users,
>
> I've got a short question regarding com-pulling. From what I understand
> Umbrella pulling is the same as AFM pulling using a harmonic potential.
>
> Why is the position of the spring in the pullx.xvg not printed? One can
> only find the vector between both pull groups. I looked into older
> Gromacs manuals and in versions<  4 it seems as if this data is printed
> to the pullx.xvg file or something similar.
>
> With kind regards,
> Christian
>