[gmx-users] Re: gmx-users digest, Vol 1 #1276 - 12 msgs

[Michael Shirts]

> ps: out of couriosity, is this supposed to renstrain a ligand
> in a binding pocket in a free energy calc as suggested by
> by, e.g., Rebecca Wade ? If so, how do you plan to correct dA for the
> restraint?

Basically, we won't be correcting for the restraint, because them it would be
equivalent to putting the ligand ANYWHERE in the box of solvent + protein.
Obviously, if this was a harmonic restraint, it would be a problem, as many of
the configurations we were interested in would be weighted slightly
incorrectly.  But with a sufficiently wide flat bottomed potential, the bound
states should all be located in the well.  For a tight binder, the it
shouldn't actually matter that much.

Talking to Erik Lindhal, it looks like that implementing such a restraint will
be relatively easy to implement using the distance restraint functionality
usually used for NMR analysis.


Cheers,
Michael