Re: strange result using implicit solvent

From: Branko (
Date: Fri Mar 29 2013 - 09:41:09 CDT


If you use ligand docked to active site, it is high probability that
ligand leave active site, so retry with the same settings and
co-crystalized ligand on the same or on the similar protein. Using of
explicit solvent is always better choice. Try to find force field with
which you can treat BOTH protein and ligand (variance of CHARMM exist),
and use the same type of charges for the protein and for the ligand
(Gastieiger for example).


On 3/29/2013 2:39 PM, Niklaus Johner wrote:
> This isn't strange, it just means your ligand is not stable in the
> binding-site. The most likely problems in my opinion are either your
> force-field, your docking pose or your equilibration procedure.
> What force-field are you using? From what I know using gaussian for
> parameterization gives partial charges that aren't necessarily
> compatible with the CHARMM force-field. Maybe run a simulation of just
> the solvated ligand, to make sure your dihedral terms are ok and that
> it samples the conformations you expect.
> If you are confident in your binding pose, parameters and
> equilibration procedure, then maybe the implicit solvent
> representation isn't good enough for that particular case, so try to
> run an explicit solvent simulation.
> N
> Niklaus Johner
> Weill Cornell Medical College
> Harel Weinstein Lab
> Department of Physiology and Biophysics
> 1300 York Avenue, Room D-501
> New York, NY 10065
> On Mar 29, 2013, at 9:00 AM, Fugui wrote:
>> Dear all,
>> I used implicit solvent to do a MD of a protein-ligand complex,
>> however, i got a very strange result. The ligand came out of the
>> binding pocket. I used gaussian03 to produce the force filed of the
>> ligand, i have no idear about the reason , could anyone give me some
>> advices?
>> By the way, i got strange results using 999 and 16 for cutoff.
>> BW
>> Fugui

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:21:05 CST