Yes, I would like to have direct access to the element symbol data
that's in the file. Otherwise, anyone that needs the element type has to
create rules for interpreting it from the "atom name" field. It feels
wrong to attempt to deduce data when it is provided explicitly.


These PDB remediation project notes suggest using the element symbol
specified in 77-78 

http://nar.oxfordjournals.org/cgi/content/full/36/suppl_1/D426#SEC3 

"Atom types are provided for every atom (i.e. ATOM record columns
77-78), so prior atom name justification conventions should no longer be
assumed in reading atom names."

 

JMOL uses the PDB element symbol if present, else interprets from the
"atom name" field. 

http://wiki.jmol.org/index.php/AtomSets 

"On PDB format, Jmol will identify the element from columns 77-78
(element symbol, right-justified). If this is absent, then it will
interpret the "atom name" field (columns 13-14) to deduce the element
identity."

JMOL is LGPL. If it interpretation is desirable, could start with its
current approach. Personally, I would be happy just with access to the
data in the file.

 

________________________________

From: [email protected] [mailto:[email protected]] On Behalf
Of Andreas Prlic
Sent: Monday, April 26, 2010 8:08 PM
To: Andy Thomas-Cramer
Cc: [email protected]
Subject: Re: [Biojava-l] PDBFileParser and Atom element symbol

 

Hi Andy

Questions:

        * Is this pattern documented in the PDB specification? 


see here:
http://www.wwpdb.org/documentation/format23/sect9.html#ATOM
 

        * If this pattern can be relied on, why are columns 77-78 also
dedicated to the element symbol?

That is the atom's element symbol (as given in the periodic table), in
contrast to the first name, which contains numbering information.

        * Should reliance on the pattern be hidden behind a BioJava
method?


If you think that is important we could probably provide an enum for all
atom types. There are two categories though: the periodic table symbol
and the one that is related to the position in an amino acid....

Andreas 
 

         

         

        
________________________________


        From: [email protected] [mailto:[email protected]]
On Behalf Of Andreas Prlic
        Sent: Friday, April 23, 2010 6:52 PM
        To: Andy Thomas-Cramer
        Cc: [email protected]
        Subject: Re: [Biojava-l] PDBFileParser and Atom element symbol

         

        Hi Andy,
        
        you could check with  Atom.getFullname(), which contains the
space characters from the PDB file:
        e.g Calpha: " CA ", Calcium "CA  " 
        
        in addition the parent group of a Calpha atom is usually an
AminoAcid and for Calciums it is a Hetatom group...
        
        Andreas

        On Fri, Apr 23, 2010 at 3:58 PM, Andy Thomas-Cramer
<[email protected]> wrote:

        
        
        Is there an easy way to identify the type of atom referenced by
an Atom
        object?
        
        For example, if Atom.getName() is "CA", is the element calcium
or the
        atom carbon alpha?
        
        If not, would it be feasible to add a method providing this in
Atom,
        AtomImpl, and parsing it in PDBFileParser, using the columns
defined at
        http://www.wwpdb.org/documentation/format32/sect9.html#ATOM?
        
        
        
        _______________________________________________
        Biojava-l mailing list  -  [email protected]
        http://lists.open-bio.org/mailman/listinfo/biojava-l

        
        
        
        -- 
        
-----------------------------------------------------------------------
        Dr. Andreas Prlic
        Senior Scientist, RCSB PDB Protein Data Bank
        University of California, San Diego
        (+1) 858.246.0526
        
-----------------------------------------------------------------------




-- 
-----------------------------------------------------------------------
Dr. Andreas Prlic
Senior Scientist, RCSB PDB Protein Data Bank
University of California, San Diego
(+1) 858.246.0526
-----------------------------------------------------------------------


_______________________________________________
Biojava-l mailing list  -  [email protected]
http://lists.open-bio.org/mailman/listinfo/biojava-l

Reply via email to