[gmx-users] the force constant in constant speed umbrella pulling

[Yun Shi]

Hi everyone,

I am just wondering about the mathematical and physical meaning of this
force constant when pulling a ligand from its receptor.

So I can imagine a dummy atom is linked to the ligand via a spring, and it
is moving away from the receptor at 1 nm/ns with the spring force constant
1000 kJ/mol/nm^2. After it moved a very small distance, for example 0.0001
nm, which would be the change in the length of the spring, there will be a
corresponding force 0.1 kJ/mol/nm, which is 0.166 pN. So is this force only
directly exerted on the COM of the ligand? Or it's been distributed across
every atom within the ligand?

>From this imagined model, is seems the force constant does not have as much
effect on the quality (i.e. better quality means closer to a reversible
thermodynamic process) of the pulling process as the pulling rate. But I
wonder, would reducing force constant by 50% have the same effect on
decreasing computational load as reducing pulling rate by 50%? Supposing a
set pulled distance 5 nm is required, which one should save more
computational time?

Thanks for any suggestion.
Yun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20111114/9e3a14fa/attachment.html>