Hi Dale,
On Tue, Dec 11, 2007 at 01:11:51AM -0800, Dale Tronrud wrote:
> If the PDB folks have removed this specification from their
> documentation they probably thought hard about doing it and things
> might have changed.
It's possible that this original specification is still in there
... somewhere. I get confused these days, since the latest format
specification available as HTML (3.0.1) at
http://www.wwpdb.org/documentation/format30/index.html
only lists differences to the 2.3 specification at
http://www.wwpdb.org/documentation/format23/v2.3.html
And the current PDB format specification for 3.1 is only available as
PDF at
http://www.wwpdb.org/documentation/format3.1-20070719.pdf
So I tend to jump back and forth between specifications ... and get
utterly confused. Maybe the PDB/wwPDB could get someone to create a
full 3.1 specification document - as PDF and HTML?
I basically don't care what the order is: any comparison between two
PDB files should be done ideally by a program ... which should be
clever enough to do the proper matching. Yes, diff (or tkdiff) are
nice tools - but each program/library seems to change something in a
PDB file (TER records, REMARK headers, spacegroup name, atom order,
sequential numbering, number of spaces to the right of a ATOM/HETAM
record).
> P.S. Clemens, you are thinking like a graph theorist
Gosh ... and I wasn't even aware of that ;-)
> when you want to place CB right after CA. I think this ordering
> stuff goes way back and the idea was that the main chain would
> always be the first four atoms. This allows one to select the
> backbone of the peptide w/o anything complicated like testing the
> equality of strings.
It might even go back to the times when the sequence of the protein
wasn't known and just getting a structure with the main-chain (or even
only Calpha) was already a great achievement?
Cheers
Clemens
--
***************************************************************
* Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com
*
* Global Phasing Ltd.
* Sheraton House, Castle Park
* Cambridge CB3 0AX, UK
*--------------------------------------------------------------
* BUSTER Development Group (http://www.globalphasing.com)
***************************************************************
|