Re: AMBER: <dV/dL> fluctuations in TI

From: Thomas Steinbrecher <steinbrt.scripps.edu>
Date: Wed, 17 Nov 2004 05:06:40 -0800

Hello Nelson,

> intermediate molecule that is transformed into R or S by
> mutation of the apropriate group into dummy atoms.

This sounds like you have atoms appearing and disappearing simultaniously.
The mixing rule with klambda = 6 is suited for disappearing atoms only.

Very large dV/dl fluctuations would be the result of applying it to
appearing ones.

If this is the cause of your troubles, try setting up the simulations as
two half steps toward an intermediate state and run both simulations in a
direction so that dummies occur only in the l=1 state.

Regards,

Thomas

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Nov 17 2004 - 13:53:00 PST
Custom Search