On Thu, Nov 29, 2018, Braden Kelly wrote:
>I do not see why I wouldnt use linear coupling for the electrostatic
>part. I do this for regular classical MD free energies. The softcore LJ
>is the same calculation for both MM and QM/MM is it not?
To make a brief addition to an already long thread:
Although the math for softcore LJ is the same for QM/MM and MM, the
practical effects are quite different. A softcore LJ allows two atoms
to get quite close together. This is fine for MM, since the atomic
partial charge has either been removed (in an earlier, de-charging,
step) or is itself scaled with soft-core electrostatics.
Neither option is available for a QM atom: you can't remove or scale QM
charges, since they are determined by the SCF procedure. Hence, bad
things are expected to happen if an MM atom gets very close to a QM
atom. In short, in the current codes you can't "deouple" a QM atom from
the rest of the system: you can make the LJ interactions softcore, but
you can't modify the electrostatic interactions in the way you can with
MM.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Nov 30 2018 - 05:30:04 PST
