2009/9/9 Arnaud Basle <[log in to unmask]>:
> Hi,
>
> I have a peptide structure at 0.92A and for the latest stage of refinement I
> would like to real space refine a Met residue including hydrogens. The H are
> flying away from their original position and density.
>
> I am using shelxl for refinement (.res and .ins format not pdb).
>
> I read from the coot mailing list archive that coot 0.4.1 was not able to do
> so ("coot currently does not handle PDB V3 hydrogens. You need to tell
> reduce to write out old style hydrogens. This should take of this
> problem.").
>
> Is it still the case with the latest version of coot?
>
Well, yes and no.
Coot's been pdb v3 compliant for a while. This helps very little with
SHELX hydrogens though, as SHELX has a different convention for H
names. AFAICS, Met H atoms behave just fine (that's 0.6-pre (r2276) vs
SHELX-97-2). Other H atoms, e.g. those of Leu and Ile and the like are
somewhat more volatile. (The easiest is to ignore this and get them
HFIXed again in the next round of shelxing).
But hey, there are zillions of other reasons for upgrading to the
latest Coot (and a bleeding-edge OS... hmm).
JED.
