Re: Question about ABF

From: Jerome Henin (jhenin_at_cmm.chem.upenn.edu)
Date: Thu May 10 2007 - 10:26:30 CDT

Germán,
The question of the reaction coordinate really depends on the specifics of the
problem you are investigating. I am afraid no one can really tell you in
advance what coodinate will do the job. You may have to try different
solutions to get a feeling of where the problems are.
One thing I must tell you is that those calculations are likely to be
difficult and costly. Even if 27 atoms is a "small" molecule, it probably has
a great number of degrees of freedom, it can at least tumble and probably
adopt different conformations. All of that, and the interactions with the
protein, will have to be sampled for each position along the reaction
coordinate. You should try and see what the relaxation timescales are for
those degrees of freedom, that are not the reaction coordinate but are
coupled to it. They will be the limiting factor when you try to reach a
converged free energy profile.
Jerome

On Wednesday 09 May 2007 20:09, German Saavedra wrote:
> Hello
> I want to move a small molecule (27 atoms) along a protein cavity (similar
> to a channel) and get energy profiles using ABF. There is 6 types of
> reaction coordinate: distance, distance-com, absissa, zCoord, zCoord-1atom,
> xyDistance.
> zCoord-1atom is only for 1 atom and zCoord is useful for the estimation of
> transfer free energies between two distinct media. So, which is the best
> type of reaction coordinate for my system? Do I have to use zCoord?, could
> I use zCoord-1atom? other?.
> Thanks
> Germán
> .

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:39 CST