> On Feb 16, 2015, at 2:11 PM, George Tzotzos <gtzotzos.me.com> wrote:
>
> I’ve run a 100ns MD trajectory of a receptor liganded to two identical ligands.
>
> My objective is to determine the relative binding energies of the ligands.
>
> I’ve stripped the trajectory of water molecules, ions and ligand_1, prepared new topology files and performed MMGBSA.
>
> I repeated the above process leaving in the complex ligand_1 and stripping ligand_2.
>
> Is this approach legitimate?
I’m not sure if it would be “better” to do what you did, or to define the “receptor” as the receptor plus one of the ligands (i.e., strip just the water and ions rather than also one of the ligands).
You could also do both and compare them. I would say the difference there would be related to allostery (and that argument could be strengthened if you used separate trajectory files), but the use of a single-trajectory approach eliminates the most important part of allostery IMO -- the conformational rearrangement.
I would *probably* leave one of the ligands in, though, since that is a smaller perturbation to the complex simulation than taking *both* out in the unbound state...
It might be easiest to actually draw out the thermodynamic cycle -- that should tell you more precisely what to compute.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Feb 16 2015 - 14:00:03 PST
