[phenixbb] clash guarde failure phenix coot

Georg Mlynek georg.mlynek at univie.ac.at
Tue Jan 19 09:57:25 PST 2010


Dear All, 

I´m refining a 2.1A with phenix 1.5-2 and get when I click on automatically
add hydrogens to model following error message:
refinement.pdb_interpretation.clash_guard.failure: Number of nonbonded
interaction distances < 0.5:391

I suppose that this comes, because coot mixes up the hydrogens. 

See 
Re: [COOT] hydrogen restraints

Paul Emsley
Wed, 09 Dec 2009 14:11:34 -0800

Sara Zueger wrote:

    Hi everybody,

Hi Sara,


    as I did not find a final solution to the hydrogen restraints problem
    (flying or exploding during real space refinement) in any of the mailing
    lists, I wanted to pick this topic up again.

    If one adds hydrogens by phenix.ready_set or reduce the labels for these
    hydrogens on let's say a Cbeta atom will be HB2 and HB3 (according to
the
    pdb v3.2 convention). However, the coot/ccp4 library names them HB1 and
HB2,
    so there are no restraints for HB3 atoms. I see that probably this
naming
    problem can be overcome by using phenix.reduce -OLD, or renaming them
    manually in the pdb file (but if one deposits the structure to PDB does
one
    have to name them back again?). Is there already some new solution?

There is not a solution yet.


    And how about hydrogen restraints for the most N-terminal amino acid,
where
    one has H1, H2 and H3? How are these handled? There seem to be no
restraints
    for those as well, because they also fly around during real space
refinement.

Same situation.


We need to do some drudge work to convert the hydrogen names in the
libraries. AFAICS it is straightforward, but just hasn't been done by anyone
in the CCP4/Coot community.

I wonder what dictionary the Phenix fellow use? If they have updated the
dictionary to 3.2 format, perhaps we could use their copy too.

(However, we deal with this, it should be done by the end of the 0.7.x
series.)

Paul.


What can - should I do now, to get my structure in the pdb, because just to
remove them is according to the manual not allowed 

See:

To remove hydrogens from a model:

% phenix.pdbtools model.pdb remove="element H or element D"

or Reduce programs can be used for this:

% phenix.reduce model_h.pdb -trim > model_noH.pdb

We strongly recommend to not remove hydrogen atoms after refinement since it
will make the refinement statistics (R-factors, etc...) unreproducible
without repeating exactly the same refinement protocol.

2. Have you ever seen that coot looses the restraints? I have a heme in the
structure which I could refine without any problems, but now he gives me
following error message: 

No restraints found
 Non-existent or minmal description of restrained residues. Are you sure
that you read a non-minimal mmCIF dictionary for this monomer? Are you sure
the PDB residue name matches the dictionary residue name?

If not, try File -Import CIF dictionary

Alternatively, did you check that the atom names of the PDB file matches
those of the restraints.


Thank you very much for your help.

Georg

-----Ursprüngliche Nachricht-----
Von: phenixbb-bounces at phenix-online.org
[mailto:phenixbb-bounces at phenix-online.org] Im Auftrag von jonathan
elegheert
Gesendet: Dienstag, 19. Jänner 2010 18:32
An: PHENIX user mailing list
Betreff: Re: [phenixbb] twin refinement in phenix-1.5-2

Hi Kelly,

I encountered the same with later nightlies than 1.5-2 also. The problem 
is the "fix_rotamers = True " command. You have to return it to its 
default value "None".

Apparently, some subroutines like that one don't work with twin_f_lsq 
algorithm. I believe because the fix_rotamers function depends on a map 
cc calculation which also in itself doesn't work with twin_f_lsq.

At least that's my understanding.

Jonathan Elegheert
PhD student
Ghent University

Kelly Daughtry wrote:
> Hello all,
> I am wondering if anyone has run across this problem and knows how to fix
it.
>
> I have a twinned dateset and I am using phenix.refine to refine the
dataset.
> I have the full 1.5-2 installation.
>
> When I include the twin law (-k, -h, -l) in the refinement, I get this
error:
>
>   File
"/usr/swr/phenix-1.5-2/cctbx_project/mmtbx/twinning/twin_f_model.py",
> line 1651, in map_coefficients
>     "m_gradient"
> AssertionError
>
> If I do not include the twin law, I do not get the error and the
> refinment continues as normal.
>
> When I downgraded to Phenix.1.4-3 (full install), And use the same
> twin law, I am able to run phenix.refine without an error.
>
> I have attached my definition file for phenix.1.5-2 amd the log file.
>
>
> Thanks in advance for any help you can provide,
> Kelly Daughtry
>
>
> *******************************************************
> Kelly Daughtry
> PhD Candidate
> Department of Physiology and Biophysics
> Boston University School of Medicine
> 590 Commonwealth Ave
> R 390
> Boston MA, 02215
> (P) 617-358-5548
> *******************************************************
> ------------------------------------------------------------------------
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb

_______________________________________________
phenixbb mailing list
phenixbb at phenix-online.org
http://phenix-online.org/mailman/listinfo/phenixbb

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20100119/eff9599b/attachment-0003.htm>


More information about the phenixbb mailing list