[gmx-users] Crashes during protein-ligand simulation

[Justin A. Lemkul]

On 7/7/12 11:08 AM, James Starlight wrote:
> justin,
>
> It seems that problem was in the big charge groups in the ligand.itp
> file. In particularly I've devided largest group into several smaller
> and there haven't any crashes been occured yet.
>
> With my last system with the the default COM group I've obtained crash
> on the 25ns with the error about ligand's big cngr exactly.
>
>
> By the way could you tell me how I could analyse stability of the
> protein-ligand system as well as contributions of the individual
> non-covalent contacts with the .EDR file?
>
> In my mdp file I've defined energygrps= Protein Ligand   as the
> separate energy terms.
>
> During analysis of the edr file I've observed some options like
> LJ-SR:Protein-ADN     ( this value was slightly negative (-100) during my trr )
> Coul-14:Protein-ADN ( this value was constantly zero )
> Coul-SR:ADN-rest ( this value was equal to the LJ-SR:Protein-ADN  -100 )
>
> What conclusions in general could I do based on that values ? How I
> could measure stability of the ligand ( beside dirrect RMSD
> measurement) from such energy terms?
>

In isolation, these values are not indicative of anything.  Their absolute 
values are completely dependent upon the parameters set in the topology.  Force 
field parameterization methodology does not require that any topology reproduce 
protein-ligand energetics; only true free energy calculations might tell you this.

In conjunction with other metrics (hydrogen bonding, contacts, RMSD, SASA, etc) 
you may be able to make some meaningful observations about ligand stability in 
the protein binding site.

-Justin

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

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

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