[gmx-users] Add new atom type

[Jian Zou]

Anton wrote:
> And, *PLEASE*, unless you have a very *good* and very *specific* reason,
> you should not use the *VERY* *OLD* ffgmx forcefield. It is *old* and
> has been replaced by the *newer* and much *BETTER* ffG43a1 (of which
> even *newer* and *better* versions are available as well).

Hi all,

I'd like to continue this discussion on running pdb2gmx with respect to 
ffG43a1.

here is a simple example
for a pdb file ADE-ADE, try pdb2gmx -f na.pdb, and choose ffG43a1 to 
generate the topology,

the connection between 2 ADE is absolutely wrong. (bond, angle, dihedral... 
lost all)

check ffG43a1.rtp-[ADE] section, modify "+N", "-C", "-O" to "+P", "-C3*", 
"-O3*, respectively, run pdb2gmx again, the new topology seems to be right, 
except that you should modify 1 dihedrals of two termini.

I don't know if this is common, or I miss sth. in the manual.
I don't know if my modification is the right way to treat these kind of 
problems of ffG43a1.rtp either.


Hope that some one could give me some advice or comment.

Thank you so much for you help.

Regards,

Jian Zou