Re: TER suggestion
Peter Keller (bsspak@bath.ac.uk)
Fri, 29 Sep 1995 10:40:57 +0100 (BST)
Hi all,
The thing which worries me about this, is that _atom_site.group_PDB is not
meant to contain fundemental information about the atom site. Herb's
suggestion, as I understand it, would mean that the interpretation of a
row then depends on the contents of _atom_site.group_PDB, which changes
its role in the ATOM_SITE category completely.
Personally, I agree with Phil, that the terminal residue of a chain is
determinable as the one with the highest _atom_site.entity_seq_num for a
particlular _atom_site.entity_id and _atom_site.label_asym_id.
Moreover, CIF says nothing about the order in which records within a
category appear in a file, so a TER record would not necessarily appear
after all the other ATOM_SITE records of the particular chain. (Even if
you construct a CIF file so that it does, initially, an application is
perfectly free to re-order a category for its own purposes. This might be
necessary, for example, to match up corresponding records between category
and sub-category tables.)
The fact that a residue is terminal, is not a property of an atomic site
in the structure, but a property of the sequence (i.e. of the chain as a
whole). If we want to indicate the end of a chain explicitly (and please
don't get me wrong - I'm not saying that this is necessarily a bad idea as
such), I think that the proper place to do it is in the ENTITY or
ENTITY_POLY_SEQ categories. Perhaps an item such as _entity.term_seq_num
(which corresponds to _atom_site.entity_seq_num and _entity_poly_seq.num)
would do what Herb wants?
Alternatively, you could define _chem_comp.id for a dummy residue (TER
say), in the CHEM_COMP category, and add it to the end of the chain in the
ENTITY_POLY_SEQ category. You would then know that any atom with
_atom_site.entity_seq_num which was one less than than the value of
_entity_poly_seq.num for which _entity_poly_seq.mon_id is TER, belonged to
a terminal residue. You would not actually need to put a row in the
ATOM_STIE category, with _atom_site.label_comp_id as TER. This is getting
a bit more convoluted, though.
Cheers,
Peter.
========================================================================
Peter Keller. \ "Having beguiled with fiction until I had
Dept. of Biology and \ none left I resorted to facts, which
Biochemistry, \ also ran out."
University of Bath, \ - Alisdair Gray
Bath, BA2 7AY, UK. \
------------------------------\-----------------------------------------
Tel. (+44/0)1225 826826 x 4302 | Email: P.A.Keller@bath.ac.uk (Internet)
Fax. (+44/0)1225 826449 | P.A.Keller%bath.ac.uk@UKACRL (BITNET)
========================================================================