AMBER Archive (2009)

Subject: Re: [AMBER] Problem with "Atom does not have a type"

From: case (
Date: Mon Mar 30 2009 - 07:46:59 CDT

On Sun, Mar 29, 2009, Richard Tseng wrote:

> Since I have mixed RNA and proteins, I first
> load leaprc.ff03.r1 then I load leaprc.rna.ff02.

I don't recommend this at all: ff02 is a polarizable force field, and ff03 is
a fixed-charge force field. There is no reason to think these would work well

The ff99SB force field seems to give pretty good results for proteins, and
should be compatible with rna as well. The easiest thing to do is to modify
your pdb file to use RU,RA,RG,RC as residue names, so Amber will automatically
know that these are ribose sugars.
> The pdb file is directly extracted from protein data bank!!

OK: I had misunderstood the new PDB format. In versions 3.15 and above (and
mayber earlier?) residues like "G", "C" are RNA; for DNA, one uses "DG",
"DC", etc. So, Amber's leaprc files need to be updated to reflect this new
usage. [We probably also want to update the library files themselves, but this
is a bigger change and affects backwards compatibility in a bigger fashion.]

For now, you either have to modify the PdbResMap in leap to translate "G" to
"RG", etc, or use a text editor to make that change in the pdb file.


AMBER mailing list