AMBER: gaff: ca-ca-c -o torsion parameter

From: Oliver Hucke <>
Date: Tue, 17 May 2005 13:54:24 -0700

Dear AMBER users/developers,

I am simulating a ligand with a carbonyl group connected to a phenyl
ring. I noticed that in my simulations the CO is always coplanar with
the phenyl, even at the expense of deviations from planarity of the
amide group, that the CO is part of.

I am using the gaff force field for the ligand (AMBER8).
gaff applies a torsional barrier of 29kcal/mol to the bond between the
phenyl carbon and the carbonyl carbon, if I am not mistaken:

X -c -ca-X 4 14.5 180.0 2.0 (line 4341 of gaff.dat)

I found experimental values of 3.15 and 4.35 kcal/mol in one of the
papera of Thomas Halgren about his MMFF force field (JCC, v17, 1996, pp.

My conclusion is that the following line should be added to the gaff
force field to get a barrier of 4kcal/mol:

ca-ca-c -o 1 .5 180.0 2.0

Please correct me if I am wrong!

Thanks and regards,


Oliver Hucke, Dr.
Biomolecular Structure Center Health Sciences Building - K418C
Dept. of Biochemistry 1959 NE Pacific St.
University of Washington phone: (206) 685 7046
Box 357742 fax : (206) 685 7002
Seattle, WA 98195-7742 email:

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to
Received on Tue May 17 2005 - 22:53:01 PDT
Custom Search