AMBER Archive (2006)

Subject: Re: AMBER: where to set bondi radii for halogens

From: David A. Case (case_at_scripps.edu)
Date: Fri Oct 27 2006 - 13:58:04 CDT


On Fri, Oct 27, 2006, Giulio Rastelli wrote:

> I am running a MM-PB/GBSA calculation using amber9
> on a protein-ligand complex whose topology was
> built with leap with default mbondi radii, using
> amber atom types for the protein and gaff types
> for the ligand. The ligand contains one chlorine
> and one fluorine atom. I noticed that in the RADII
> flag in the topology, the two halogens are assigned a value
> of 1.5, which seems to be the default when radii are
> not parameterized.

> Is there a way in unitio.c file in leap to tell
> that the radii should be Cl=1.75 and F=1.47, in
> each session pertaining to different igb options?

This sounds like a bug. The code in unitio.c looks like it should be
giving non-default radii for fluorine and chlorine. (Note that the mbondi
value of fluorine 1.5, the same as the default, but you should be getting
a radius of 1.7 for chlorine.)

One test you can make: go to $AMBERHOME/test/antechamber/sustiva, and use
the sustiva.mol2.save and frcmod.save files as input into LEaP. If you save
the prmtop file, you should see that the radius for the chlorine atom is 1.7
--anyway, that is what I get. So, see if that works for you.

You can certainly change the values that are assigned, either by editing
untio.c or by just editing the prmtop file once it is created.

[It could be that the mm-pbsa script is over-riding the prmtop values in some
fashion, but I would need to see more details.]

...regards....dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu