On Thu, May 20, 2004, riccardo nifosi wrote:
>
> I'm trying a free energy calculation of a threonine to serine substitution
> with the thermodynamic integration module of sander (I'm using amber7).
> What is the behavior of shake when the equilibrium distance between the dummy
> hydrogens (those of the disappearing methyl) and the carbon is different in
> the perturbed and unperturbed systems? I noticed that though shake is turned
> on these distances vary during the MD window run, so I suppose that they are
> not constrained. Using shake in my case would not be correct... Am I right?
Atoms that are being perturbed are never shaken. This is because the
equilibrium bond length might change during the perturbation, and then SHAKE
would be doing the wrong thing. It would probably be possible to run some
check, and allow SHAKE to operate if the bond length in the lambda=1 state
were the same as in the lambda=0 state, but this sort of code has never been
written.
....regards...dac
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Thu May 20 2004 - 19:53:01 PDT
