[gmx-users] problem with energy groups and mdrun -rerun option

[Mark Abraham]

On 22/10/2010 8:21 AM, jagannath mondal wrote:
> Hi,
>   I have  used gromacs 4.0.7 to do  MD simulation of two solutes A & B 
> in water ( solvent) .
> Initially, I had set "energy groups = system " and used mdrun to do 
> the simulation.
>
> Now,I wanted to get the potential energy contribution from due to 
> interaction of  A-A, B-B,A-B A-solvent, B-solvent, solvent-solvent.
> For that I followed some discussions in mailing list regarding using 
> mdrun -rerun option:
> This is what I did:
> I used make_ndx and created 3 groups: A, B and solvent.
> Now, I modified the  grompp.mdp file so that I have now
>   energy groups = A B solvent
>
> Then I used grompp -c conf -f grompp -n index -o topol
>
> Then I used mdrun -s -rerun traj_old.xtc
>
> ( here traj_old.xtc is the old xtc file I obtained during the original 
> mdrun)
>
> But, now, after using this -rerun option , if I try to use the 
> g_energy on the  resulting ener.edr file to obtain the individual 
> potential energy of interaction for A-A, B-B , A-B, A-solvent
> I get following output
>
>
> Here is the output from g_energy -f ener.edr :
>
> Select the terms you want from the following list by
> selecting either (part of) the name or the number or a combination.
> End your selection with an empty line or a zero.
> -------------------------------------------------------------------
>   1  Bond             2  Angle            3  U-B              4 
>  Ryckaert-Bell.
>   5  Improper-Dih.    6  LJ-14            7  Coulomb-14       8  LJ-(SR)
>   9  Coulomb-(SR)    10  Coul.-recip.    11  Potential       12 
>  Kinetic-En.
>  13  Total-Energy    14  Conserved-En.   15  Temperature     16 
>  Pressure-(bar)
>  17  Cons.-rmsd-()   18  Vir-XX          19  Vir-XY          20  Vir-XZ
>  21  Vir-YX          22  Vir-YY          23  Vir-YZ          24  Vir-ZX
>  25  Vir-ZY          26  Vir-ZZ          27  Pres-XX-(bar)   28 
>  Pres-XY-(bar)
>  29  Pres-XZ-(bar)   30  Pres-YX-(bar)   31  Pres-YY-(bar)   32 
>  Pres-YZ-(bar)
>  33  Pres-ZX-(bar)   34  Pres-ZY-(bar)   35  Pres-ZZ-(bar)   36 
>  #Surf*SurfTen
>  37  Mu-X            38  Mu-Y            39  Mu-Z            40 
>  Coul-SR:A-A
>  41  LJ-SR:A-A       42  Coul-14:A-A     43  LJ-14:A-A       44 
>  Coul-SR:A-B
>  45  LJ-SR:A-B                           46  Coul-14:A-B
>  47  LJ-14:A-B                           48  Coul-SR:A-Solvent
>  49  LJ-SR:A-Solvent                     50  Coul-14:A-Solvent
>  51  LJ-14:A-Solvent                     52  Coul-SR:B-B
>  53  LJ-SR:B-B                           54  Coul-14:B-B
>  55  LJ-14:B-B                           56  Coul-SR:B-Solvent
>  57  LJ-SR:B-Solvent                     58  Coul-14:B-Solvent
>  59  LJ-14:B-Solvent                     60  Coul-SR:Solvent-Solvent
>  61  LJ-SR:Solvent-Solvent               62  Coul-14:Solvent-Solvent
>  63  LJ-14:Solvent-Solvent               64  T-System
>  65  Xi-System
>
> It provides me contribution from each of the energy groups on 
> nonbonding terms: LJ(SR),Coulomb(SR),Coulomb(14),LJ(14) .  But, it 
> does NOT provide me their contribution to Bonding term( i.e 
> bond,angle, U-B,RB,improper-Dih) !!
> As a result, I am a not sure how to get the net potential energy from 
> each of the 3 energy groups. Am I doing something wrong ? Do I need to 
> use any other utilities ?
>

Decomposing the bonded potential requires you to hack about in your .top 
for some more mdrun -rerun runs. For speed, in your .mdp file set 
rlist/rvdw/rcoulomb to some tiny value - this only affects non-bonded 
potential, which you don't care about this time. Then remove the bonded 
interactions from the .top for 2 of your groups, and a rerun will 
compute the bonded interactions of the third. Rinse, repeat.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101022/dfae828a/attachment.html>