From: Victor Ovchinnikov (ovchinnv_at_MIT.EDU)
Date: Thu Nov 15 2007 - 21:00:51 CST
attached mail follows:
The IC table has default parameters (internal coordinates) that CHARMM
uses for building molecular coordinates when they are unknown (I don't
know whether psfgen works in a similar way). Each entry has two angles,
one dihedral and two distances. Note that there are no parameters
necessary for computing forces.
If all atomic coordinates are known, then the IC table is not used. In
particular, it is not used in any force evaluations. If you are
developing/porting a CHARMM potential, I recommend the original CHARMM
paper -- Brooks et al, JCP 1983 for some more info. You can also go to
the CHARMM website www.charmm.org and browse the documentation
(rtop.doc, struct.doc, intcor.doc), which might clarify things.
Sorry, I'm not sure what you mean by `.par' files. Typically, files
that begin with `par' contain bond/angle/dihedral parameters, and files
that begin with `top' define atoms, their masses, provide connectivity
info for residues, and declare partial charges on the atoms in a
`.str' files usually have a mixture of topology/parameter info that
extend the force field to other molecules, e.g. lipids, NA.
On Thu, 2007-11-15 at 17:02 -0800, Ilya Chorny wrote:
> Hi All,
> I am working on porting a third party lipid forcefield to NAMD format.
> When looking at the POPE forcefield in CHARMM27 (all_27***) there is a
> something called an IC table which looks like a definition of dihedral
> potentials. Is that correct? Also what is the relationship between the
> .par and .inp files. Does NAMD default to the definition in the .inp
> file prior to looking up up the values in the .par file. For example
> if the IC table is the definition of the dihedral potential would namd
> use those definitions and ignore what is in the .par file?
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:45:33 CST