parm94 and glycam93 in AMBER

[Bernd Kallies <kallies - at - serv.chem.uni-potsdam.de>]

 We are currently dealing with molecular mechanics calculations on
 proteins coupled with oligosaccharides. We decided to use an AMBER force
 field together with GLYCAM. The original GLYCAM paper (Woods et al, JPC
 1995) said that this parameter set (called glycam93 in the AMBER
 program) was developed in agreement with the old AMBER force field
 (Weiner et al, JACS 1984, called parm91), but it might also be used with
 the new developed AMBER parameters (Cornell et al, JACS 1995, called
 parm94) because of the similar philosophy of assignment of atomic
 charges (both parm94 and glycam93 use RESP charges from HF/6-31G*,
 whereas parm91 used HF/STO-3G).
 Regarding to the coupling of glycam93 and parm94 we have several
 questions:
 1. Does anybody know if these parameter sets should be coupled in
 principle?
 2. If one can do so, which scaling factor of 1-4 nonbonding interactions
 should be used (0.5 as for parm91 and as used in Woods' paper for
 parameterization, or 0.83 as for parm94)?
 3. The Woods paper refers to some parameters in the GLYCAM force field
 which were adopted from the parm91 parameter set. Should these
 parameters be used from parm91 also for the coupling with the parm94 set
 or should they be replaced by the new parm94 parameters, which are a bit
 different?
 Thank you for your assistance.
 Bernd Kallies
 --
 Dr. Bernd Kallies
 University Potsdam
 Institute for Physical and Theoretical Chemistry
 Am Neuen Palais 10
 14469 Potsdam
 GERMANY
 e-mail: kallies - at - serv.chem.uni-potsdam.de
 WWW   : http://www.chem.uni-potsdam.de/~kallies/kallies.htm
 Tel   : [+49] (0)331 / 977-1313
 Fax   : [+49] (0)331 / 977-1315