[gmx-users] gromacs charges

Erik Lindahl lindahl at stanford.edu
Mon Sep 30 22:42:06 CEST 2002

> Q1:
> If charges for all OPLS atom types are specified in ffoplsaanb.itp, does 
> the reading format for  ffoplsaa.rtp require to specify the charges in 
> the topology of the molecule. I mean do I have still to fill in the 
> charges in the rtp file? I saw that in ffoplsaa.rtp the charges for all 
> amino acids are specified even they are given in ffoplsaanb.itp.

We first look in the rtp file (or rather the topology, which is created
from the rtp file), and if the charge isn't specified there we take the
default value from the ffXXXnb.itp file.

The only reason they are duplicated in OPLS is that I used a script to 
do it...

> Q2:
> If I need to introduce a new OPLS atom type (e.g. sulfonium ion [SR3]+), 
> what code should I give to it? Codes run up to opls_965 in foplsaa.atp 
> file.

You can name it any way you want, but I would recommend something that 
won't be used later by some other opls type, like opls_xxxB  or 
something (if there already is an opls_xxx type that is quite similar).

Since OPLS uses combination rules for everything, you won't have to add
any parameters if you just give the new atomtype an existing 
bonded_atomtype (the second column in ffoplsnb.itp).



More information about the gromacs.org_gmx-users mailing list