Re: Protein-ligand simulation

From: Kenno Vanommeslaeghe (kvanomme_at_rx.umaryland.edu)
Date: Sat Jun 29 2013 - 12:36:32 CDT

On 06/29/2013 01:02 AM, James Starlight wrote:
> The HB name is present in the top_all36_prot.rtf
> -snip-
> ATOM HA HB1 0.09
> -snip-
> ATOM HB HA1 0.09 ! | CG1--CD--HD2
> -snip-

Just to get our terminology straight, what you're showing us here is a
*topology* file. One of its functions is to map atom *names* (as found in
the pdb file) to atom *types* as found in the *parameter* file
par_all36_prot.prm . So if your pdb file contains the symbol HB (an atom
*name*), the above *topology* file instructs psfgen to assign the atom
*type* HA1, so if you're looking at the *parameter* file, you have to look
for parameters involving HA1. Similarly, the atom HA in the pdb file
receives the atom type HB1, so it will be referred to as HB1 in the
parameter file. As JC said, in CHARMM22, the atom *type* HB1 used to be
HB. So if you have a psf file generated with the CHARMM36 topology file,
this will not work if you give NAMD the CHARMM22 parameter file (or vice
versa). The solution is to regenerate your psf file using
top_all36_prot.rtf AND specify par_all36_prot.prm in your NAMD input.

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:24 CST