[gmx-users] Error while running mdrun with rerun option for LIE calculation

[Justin Lemkul]

On 1/8/16 12:35 PM, shagun krishna wrote:
> Dear Justin,
>
> Thanks a lot. I have successfully completed mdrun with -rerun option. I
> have ran the mdrun command on the original mdrun.xtc using new .mdp
> settings without PME, and the result was stored in new .edr file. This new
> .edr file was then subjected to g_energy to obtain the values of LJ-17 and
> Coulomb-14. Hope we have to calculate this only not the LJ-(SR) or
> Coulomb-(SR) ?
>

As I said before, 1-4 interactions are intramolecular.  These are NOT what you 
want.  What you actually want are Coul(SR) and LJ(SR), i.e. the nonbonded 
interactions in the system.  It is the difference in these terms that matters most.

>>>>   mdrun -s md_sol_rerun_LLP.tpr (ORIGINAL tpr obtained from Ligand
> Simulation in water) -rerun md2_LLP.xtc ( ORIGINAL ,xtc obtained from
> Ligand Simulation in water) -x after_rerun_trial4.xtc (OUTPUT) -e
> after_rerun_trial4.edr (OUTPUT)
>
> After that I calculated value of Elj and Eqq from g_energy module by
> passing these output to g_energy:
>
>>>>> g_energy -f after_rerun_trial4.edr -o after_rerun_trial4_C-14.xvg
>
>>>>> g_energy -f after_rerun_trial4.edr -o after_rerun_trial4_LJ-14.xvg
>
> I got Elj= 271.04 and Eqq=-149.214. Then these values were passed to g_lie
> module to get the Delta G for unbound ligand.
>
>>>>> g_lie -f after_rerun_trial4.edr -b 7000 -e 10000 -Elj 271.04 -Eqq
> -149.214 -ligand M_R -o lie_ligand.xvg
>
>
> I am taking last 3000ps of my simulation for calculating energy. Is it good
> or should I calculate it from 0 to 10000ps.
>

You'll need to establish whether or not 10 ns is enough for convergence and at 
what point convergence was established.  Don't throw out good data.  But again, 
you need to be analyzing the right quantities.

> After passing these values to g_lie I am getting DGbind = -8.956 (214.311).
> In g_lie module we will give the original .edr file or that we have
> obtained after rerun. However, I have used the .edr obtained with rerun
> option. The values of Clj and Cqq were taken as default (0.181 and 0.5
> respectively). I want to know that do we also need to calculate the values
> of Clj and Cqq ???

I've already said this - you need to go into the literature and figure this out. 
  If you want free help, demonstrate that you're doing your own homework.

-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

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================