[gmx-users] Stuck ...

[Erik Lindahl]

Hi,

On May 24, 2007, at 10:06 AM, steletch at jouy.inra.fr wrote:


> My amber experience is probably not good for gromacs since the  
> logic of using
> files is sometimes different (and it seems heating, equilibration and
> production is not used often in gromacs, probably due to the fact  
> you can
> isolate thermal bath for different parts of the system).

There's nothing wrong with heating, but in our opinion it's a bit  
difficult to control, particularly if you have e.g. lipids and  
proteins overlapping in membrane systems that can take multiple  
nanoseconds to relax.

Instead, we prefer to use position restraints (typically 1000kJ/mol/ 
nm, but that's variable) to the heavy atoms during equilibration. The  
advantage of position restraints is that even if water/ions/ 
sidechains initially make bad steric contacts and distort the local  
conformation they will gradually be brought back to the reference  
positions, and you can monitor the extent of the distortion by  
looking at the position restraint term in the energy file.

Cheers,

Erik