Re: [ccp4bb]: cannot calculate structure factors

[Richard Baxter <rbaxter@midway.uchicago.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Dear All,

I have received a veritable flood of replies suggesting that my PDB file
contains misaligned atom names. At first I could not understand this,
because I thought I had already adjusted the position of my atom name to
the left. However, on closer inspection I realise that my atom names are
all justified to column 14 rather than column 13. Correcting this for the
Magnesium ion allows the script to run successfully.

For the record, I am using Charlie Bond's PDB mode for Xemacs to
manipulate my PDB files:

http://stein.bioch.dundee.ac.uk/~charlie/scripts/pdb-mode.html

I think it is very good - I see that the majority of the atom names (that
are three letters long) are right justified to column 16 and two or
one-letter atom names left justified wrt to those, hence my C-alpha
carbons have the 'C' in column 14 of the PDB file, I presume this avoids
incorrect assignment to calcium?

So my final question is about the other atom names in my PDB file that
might give me trouble, the iron ions FE, and the four porphyrin nitrogens
NA, NB, NC and ND. Currently these are all left justified to column 14 of
the PDB record. Will this be correct to identify them as iron rather than
flourine, and nitrogens rather than sodium, niobium and neodynium?

regards,

Richard Baxter

On Fri, 14 Feb 2003, P.J.Briggs wrote:

> 
> Dear Richard
> 
> I didn't see any other replies to your question. The warning you see
> isn't from atomsf.lib but from the PDB reading library "rwbrook.f", and
> the problem is most likely in your input PDB file - I suspect that the
> atom name for "MG" is mis-aligned.
> 
> The atom name should be left justified in columns 13-16 of the PDB file,
> since the contents of columns 13-14 are what the library uses for
> comparision. Please check the format and let me know if this doesn't
> solve the problem.
> 
> (Btw a description of the PDB format used by CCP4 can be found in the
> CCP4 documentation, for example see
> http://www.ccp4.ac.uk/dist/html/pdbformat.html#part1ato
> for a description of the ATOM lines.)
> 
> Best wishes
> 
> Peter.
> 
> On Thu, 13 Feb 2003, Richard Baxter wrote:
> 
> > ***  For details on how to be removed from this list visit the  ***
> > ***          CCP4 home page http://www.ccp4.ac.uk         ***
> > 
> > Dear All,
> > 
> > I am tryaing to calculate structure factor amplitudes and phases for a
> > model of the photosynthetic reaction center (B. viridis). This is a
> > four-subunit polypeptide with many cofactors, and includes iron and
> > magnesium ions.
> > 
> > SFALL fails with the following warning and error:
> > 
> > WARNING:  *UNKNOWN ATOMIC FORMFACTOR  MG  IN BCB   301*
> > .....
> > 
> > The program run with command: sfall HKLOUT
> > .....
> > has failed with error message
> >  SFALL:  No match for atom ID
> > 
> > It seems strange to me since the library
> >   /opt/pkgs/ccp4-4.2.1/lib/data/atomsf.lib
> > exists and is referenced by the program, and the log file shows no
> > complaint for the iron atoms present in the haem cofactors that appear
> > before the bacteriochlorophyll cofactors in my PDB file. Any advice is
> > greatly appreciated,
> > 
> > regards,
> > 
> > Richard Baxter
> > University of Chicago
> > 
> > 
> > 
> 
> --
> _____________________________________________________
> Peter J Briggs, pjx@ccp4.ac.uk   Tel: +44 1925 603826
> CCP4,           ccp4@ccp4.ac.uk  Fax: +44 1925 603825
>                 http://www.ccp4.ac.uk/
> Daresbury Laboratory, Daresbury, Warrington   WA4 4AD
>