AMBER Archive (2003)

Subject: Re: bad atom type f(fluorine) in MM-PBSA?

From: Holger Gohlke (gohlke_at_scripps.edu)
Date: Wed May 21 2003 - 10:49:20 CDT

Lishan Yao wrote:
>
> Hi:
> There is one fluorine atom in my organic molecule. I use MM-PBSA to do
> some calculation. But I get an error message: bad atom type f. What
> should I do?

The error occurs because F has not been parameterized for the LCPO model
(s. Weiser et al., JCC 1999, 20,217), which is used to calculate the
solvent-accessible surface area for GBSA calculations within sander. If
you "come up" with some reasonable parameters for this, you can add
these parameters to the list in mdread.f (in $AMBERHOME/src/sander; see
the code after
"c --- construct parameters for SA calculation; note that the
 c radii stored in L165 are augmented by 1.4 Ang.
"
).

Alternatively (and probably the easier way), you might want to determine
the SAS by using molsurf (set MS=1 in the mm_pbsa.in file). Have a look
at the "atmtypenumbers" file, to make sure that the atomic radius for F
is included there.

Best regards

Holger

>
> Sincerely
> Lishan 

-- 
+++++++++++++++++++++++++++++++++++++++++++++
Dr. Holger Gohlke
Dept. of Molecular Biology, TPC15
The Scripps Research Institute
10550 N. Torrey Pines Rd.
La Jolla CA 92037  USA
phone: +1-858-784-9788
fax:   +1-858-784-8896
email: gohlke_at_scripps.edu
+++++++++++++++++++++++++++++++++++++++++++++