Re: Using ABF to explore the conformational space of a spin label attached to a membrane protein

From: Jérôme Hénin (
Date: Thu Dec 16 2010 - 07:47:42 CST

Dear Ajasja,

>>Actually 3700000, including the last stage, which repeats the first
>>one (useful to check for hysteresis). 10000 steps might not be needed
>>for convergence, but the restraints seem to make it slower than a
>>standard minimization.
> Just one more question: What happens when the transition from one stage to
> the next occurs? I would imagine that only the reaction coordinate increases
> for one step, but the rest of the system stays minimized (since the last
> frame of the previous stage is minimized).

Indeed. Then the atoms next to the rotated bond move quickly in
response to the restraint, but then this has to propagate to the rest
of the molecule. That can be quite slow. A more efficient (but more
involved) method would be to prepare initial coordinates for each
stage by performing rigid-body rotations around the bond (e.g. in
VMD), saving those coordinates as a sequence in a DCD file, and
writing a NAMD script that loads new starting coordinates between each
stage, with alternating "coorfile read" and "run" commands.


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:56:28 CST