[Xplor-nih] Standard formats of PDB ??

[John Kuszewski]

Hi,

Jut to expand on Charles's answer a bit,



On Oct 3, 2007, at 4:22 AM, R.M. van der Werf wrote:

> Hi,
>
> Perhaps this is a stupid question, but I do wonder why
> - Xplor cannot read PDBs as provided by the protein databank?
> - Xplor output is not PDB standard as required for submission to the
> protein databank.

The two major differences between xplor-standard PDBs and PDB-standard
PDBs are the atom names and the absence of the segid field in the PDB  
standard.

Xplor has used nonstandard atom names for a long time because they  
greatly facilitate
atom selections.  The PDB's standard atom names for, for example, leu  
delta protons
are 1HD1, 1HD2, etc.  Xplor's are HD11, HD12, etc.  With xplor's atom  
names, you can
do selections like (resn leu and name h*) that cover all the atoms in  
the residue.

As an aside, lots of our scripts (in Python, TCL, and in the classic  
xplor language)
are dependent upon the xplor atom naming convention.  So changing  
naming conventions
would require changes to a lot of stuff.

The second major difference between xplor and PDB-standard PDB files  
is xplor's addition
of the segid field.  As far as I can tell, Axel added the segid  
because the PDB standard's
chainID is only one character long.

As Charles said, we have some pretty good tools for reading and  
writing PDB-standard
PDB files, but we're always looking for cases where xplor's tools  
don't yet work.
So please email us if you find such a case.

--JK