From chemistry-request $#at#$ ccl.net Fri Jun 26 04:53:50 1992
Date: Fri, 26 Jun 92 08:28:04 +0200
From: martin #at# biokth.sunet.SE
Subject: Cut offs.
To: "chemistry #at# ccl.net" #at# kth.sunet.SE
Status: RO



Hi,

To my knowledge there are not many studies of how to treat cutoffs in 
the literature. However in Proteins: Struct. Funct. Genet. 6:32-45 (1989)
there is a quite extensive study. They have used the CHARMM force field,
however I think this may account for any similar force field.

My opinion is that the cutoffs that one has to use is very dependent on
how the electrostatics of the calculation is treated. 

I'm a Sybyl User that use the AMBER FF in Sybyl for dynamics and 
mechanics calculation of biomolecules. For vacuum calculations I use a
cutoff of 10-12 A (distance dependent dielectric constant) and 8-10 A
for solvent calculations.

The effect of the cutoffs to the vdW forces is less important I think.
These long range vdW forces does not have any large impact on the 
structure, but may of course change the total energy level of a 
calculation.

Finally, I think that what may become even more important is to include
polarizability to the electrostatics, especially for biomolecules. Not 
before that becomes real we will treat simulations of biomolecules in 
the best possible way. I have heard from a colleague about a new structure
of an ion transporting protein which had a narrow channel where in one the 
sides of this channel it was up to 5-7 lysines or arginines and the other
side contained about 4 aspartates or glutamates. These residues were 
clustered very near eachother. Would anybody think that a molecular 
dynamics simulation of this protein would be succesful using todays 
standard force fields ? I hesitate......

Any comments to the net regarding these questions would be very much 
welcomed by me, because I think it is very important for the improvement
of the MD techniques in biological chemistry (which I think and hope is 
and will be a very useful tool).

Greetings
Martin Norin
KTH
Stockholm             e-mail:martin /at\physchem.kth.se