[gmx-users] Protein-MG-GDP (Protein-Ion-ligand) system simulation

[Justin A. Lemkul]

neeru sharma wrote:
> Dear Gromacs users,
> 
> I have to simulate a system consisting of Protein+MG+GDP where Protein 
> is in complex with MG and GDP. I generated the GDP parameters using 
> antechamber and then converted the topology,coordinate and parameter 
> files to the corresponding gromacs format.
> Then, I tried to start with the simulation using AMBER ff03 forcefield. 
> The system ran well till energy minimization step, but at equilibration 
> steps it is showing lincs error at NVT equilibration step.
> 

What was the output of energy minimization?  What I'm looking for is the actual 
screen output with Epot and Fmax.  Just because EM "completed" does not mean it 
was necessarily sufficient.

> I used 2 coupling groups for my system with lincs constraint algorithm 
> as follows:
> *tc-grps     = Protein_MG_GDP Water_and_ions *
> 
> I tried turning off lincs warning too, den NVT equilibration was done 
> finely but the system was blowing up during NPT equilibration. I tried 

No, NVT was not fine - you ignored a major error that indicates instability in 
your system.

> with 3-4 steps of minimization also, but it didnt work (I have tried 
> with lincs warning off in NPT too). I have visualized the NVT output 
> trajectory and compared it with the initial structure too, and the 
> trajectory was fine.
> 

You'll have to define what you think "fine" means, because if the simulation can 
only run when disabling LINCS warnings, something has to be severely wrong.

> 
> I am unable to figure out the reason for why the system is blowing up. I 
> am following the drug-ligand simulation tutorial but still can't figure 
> out the problem. Can anyone suggest any possible means to overcome this 
> situation?
> 

Yes, a general workflow is here:

http://www.gromacs.org/Documentation/Terminology/Blowing_Up#Diagnosing_an_Unstable_System

The source of the error could be any number of things, including incorrect 
topology, unsuitable .mdp settings, inappropriate system 
construction/minimization, or any combination of those factors.  Posting a 
complete .mdp file would be helpful.

-Justin

-- 
========================================

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

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