[gmx-users] binding free energy calculations on charged ligands

[Diane Fournier]

Hello all,
 
I'm still trying to get the binding free energy for my ligands, and I now think I will have to do another simulation, since my first one used PME and I didn't record velocities in my trajectory.
 
But first I need some advice, since I can't find a paper which describes unambiguously what I want to do. 
 
First, my ligand is charged (carboxylate) ; what is the best (and easiest) method to calculate the binding free energy for charged ligands ? Is MM-PBSA preferable to LIE in that case ?
 
I tried LIE with PME, and I know that you can get the PME contribution by doing a mdrun -rerun with zero charge on all charge groups alternatively. I also understand that this was described in the mail archive, but I can't find the mails in question. Can somebody tell me exactly when those mails were written ?
 
And do the 'standard' settings for PME (as described in the manual) need to be changed ? I understand that the Coul-LR and LJ-LR terms are not produced when rcoulomb = rlist and rvdw = rlist respectively ? Or those are the terms that are calculated in the zero-charged reruns ?
 
Also, what is to be done with the counterions ? Since it does not seem possible to have an equilibrium distribution in a reasonable time (on a ns scale), must I run the simulations without them ? 
 
All advice will be very appreciated.
 
Diane
 
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20060915/701dd40e/attachment.html>