On 31/05/16 12:59, Georg Mlynek wrote:
> Dear All, after changing from coot 0.8.2 to 0.8.3 I encounter
> following problem.
>
> If I refine in coot the model with hydrogens, the structure does not
> fit nicely to the electron density (green model in attached picture).
>
> If I refine the same model where I deleted the hydrogens, everything
> is fine (yellow model). (It does not matter, on which computer and
> which protein structure I am using, however my neutron structure seems
> to be ok).
It is not clear to me from your picture that anything is going wrong.
Heavy atoms can be pushed around by hydrogen atoms. (This is not to say
that there is not a bug in the hydrogen atom interactions, of course.)
>
> I guess it has something to do with this change in the current release.
>
> CHANGE: Hydrogen atoms are not longer fit to density
>
>
> Is this what it should be?
Because the new method results in better hydrogen geometry.
>
> [I am using ubuntu 15.10 and phenix which writes out hydrogens by
> default compared to refmac which one has to tell explicitly (as far as
> recall).]
>
>
>
>
>
You are not doing yourself any favours by burying a question about Coot
in a discussion about meso-quasi-para-isomorphism.
New Topic -> New Thread (you are not the only one).
Paul.
|