Re: ABF simulation with extendedLagrangian on keeps crashing

From: Jérôme Hénin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Thu Apr 21 2011 - 09:43:21 CDT

Hi Ajasja,

In general, I have found that stability can be a problem in extended
Lagrangian simulations when biasing forces become large. Just to
check, are you running the latest code? There used to be a bug in the
integrator that made it much less stable than it should have been. The
fix went into CVS on January 13.

Other than that, it's a bit hard to tell. It's all a bit experimental.
The integrator might need further improvements, such as a system to
damp out resonances that could lead to bad stuff.

Cheers,
Jerome

On 21 April 2011 13:30, Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com> wrote:
> Hi Giacomo,
>
>
>> You have dummy atoms in the dihedral definitions, which stay fixed in
>> space.  The rest of the system, however, can drift under the effect of the
>> PME truncation error, Langevin forces, etc as you know already.  Do the
>> other atoms fly very far away from the dummy ones?  Maybe that could be why
>> you get the dihedrals out of control?
>>
>> The backbone atoms of the helix are constrained, so the system can not
> drift. (The part of the backbone where the label is attached is not
> constrained, so that hopefully the orthogonality of colvars and constraints
> is OK.
>
> I tried changing the values of the coupling
>  extendedFluctuation 3
>  extendedTimeConstant 80
> With this end result<http://www.ijs.si/ijs/dept/epr/ajasja/mails/ABF/crash2.gif>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:23:51 CST