[Bioperl-l] Parsing PDB entries in BioPerl

[Kris Boulez]

Quoting Ewan Birney (birney@ebi.ac.uk):
> 
> 
The following basic object layout was proposed by Ewan. It looks
reasonable to me (see small remarks).
> 
>   Bio::Structure::Chain.pm (a particular PDB chain)

you also need 
		  ::Model.pm (one Entry can consists of multiple models,
		  NMR data is like this)

>                 ::Entry.pm (a structure entry, containing one or more
> chains and an annotation object)

- An Entry then consists of one or more models. Default would be the first/only
model.
- I think the Entry also should contain a Bio::PrimarySeq which would give
people access to the protein sequence (to do similarity searches, ...)
- All the non-coordinate stuff from PDB can be hidden in this annotation
object.

>                 ::Residue.pm (perhaps - chain made of residues?)

- Most definitely a Residue object is needed. People in this field tend to
think in terms of residues.
- Do we store the 'non-standard' residues (e.g. haem group) ('HET'
  records in PDB) in here as well. Or is the name 'residue' in peoples
  heads linked to "amino acid residue". I could envision a ::HeteroGroup
  object.

>                 ::Atom.pm (Residue made from Atoms)

- one small question here: what do people expect to happen with atoms
  that have alternate locations (i.e. position of the atom could not be
  uniquely be defined and n possibilities are given) ? Should the first
  alternate be choosen, or the one with the highest occupancy, or ?


We need a place to store the (non standard) links between
residues/hetero groups. Also things like disulfide bridges, cis peptide
bonds (CONECT, SSBOND, LINK, ... from PDB).
		
		::Connect.pm (two (or more) Atoms are connected)

> 
>                 ::IO.pm   (Bio::Structure::IO top level file, ala SeqIO)
>                 ::IO::pdb (pdb IO module)
>                 ::IO::cif (cif IO module)
>                 ::IO::mmdb (mmdb IO module)
> 

Kris,
-- 
Kris Boulez 				Tel: +32-9-241.11.00
AlgoNomics NV 				Fax: +32-9-241.11.02
Technologiepark 4 			email: kris.boulez@algonomics.com
B 9052 Zwijnaarde 			http://www.algonomics.com/