RE: Rigid body simulations in charmm

What about fixing all internal coordinates in the to-be-rigid-molecule:
 using the CONS ICcommand ?
 all bonds must have the same restraint constant (same for angles, dihedrals
 etc..) , but since you want to define a rigid body, you can use one unique
 very high constant. You would have to define explicitly all ICs in the CONS
 IC command, but doable since you're dealing with a dipeptide..
 ----------------------------------------------------------
 Jerome Baudry, Ph.D.       Senior Research Scientist, VizLab Manager.
 School of Chemical Sciences    University of Illinois at Urbana-Champaign
 505 S. Mathews Ave.; Box 2.1; Urbana, IL, 61801
  Phone: (1) (217) 244 3210; Fax: (1) (217) 333 3120;    e-mail:
 jerome~at~scs.uiuc.edu
 > -----Original Message-----
 > From: Computational Chemistry List
 > [mailto:chemistry-request~at~ccl.net]On
 > Behalf Of Konrad Hinsen
 > Sent: Tuesday, May 04, 2004 5:02 AM
 > To: Ravindra VenKatramani
 > Cc: chemistry~at~ccl.net
 > Subject: CCL:Rigid body simulations in charmm
 >
 >
 > On May 3, 2004, at 7:02, Ravindra VenKatramani wrote:
 >
 > >       My first thought is to define the dipeptide structure only
 > > thorugh
 > > bonds (no angles or dihederals) and use SHAKE to keep the peptide
 > > fixed.
 > > Doing so i`m running into SHAKE convergence problems during
 > the heating
 > > and equilibration stage. Am i missing something in this
 > simple minded
 > > approach? Is there an alternate way to perform rigid body
 > simulations?
 > >>