Hi,
if your TI calculation changes the system total charge, you have no choice
but to have one side of your transformation contain a net-charged box. As
far as I know, there is no specific problem involved with doing so, i.e.
sander will give you the 'right' answer using the neutralizing plasma.
Please note that you would not only add the effect of the neutralizing
plasma in one step, you would also remove it on the second leg of your
thermodynamic cycle (changing the ligand peptide back in water), so one
can hope that any errors cancel. The alternative of adding/removing an ion
to keep the change neutral is undesirable for the sampling reasons you
mention. My opinion on this is, if (!) you trust a normal MD simulation
with a charged box, you can also trust a TI calculation with one.
Please note that Amber11, which will be released soon (April...) will
contain a modification to the soft core TI codes that would allow you to
do such transformations easier in single steps. You might want to think
about upgrading if you plan to do many TI runs.
Kind Regards,
Thomas
Dr. Thomas Steinbrecher
BioMaps Institute
Rutgers University
610 Taylor Rd.
Piscataway, NJ 08854
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jan 27 2010 - 01:00:03 PST
