[Bioperl-l] protein networks

[Boris Steipe]

Richard,

Thanks for this initiative !

I might suggest including "interaction" in the package name rather than 
"protein" since this is what you are graphing. Also there are 
Protein-DNA and DNA-DNA interactions (and there will be many more) 
which are being stored by some.

In terms of IO, I would like to suggest supporting the HUPO-PSI 
standard (which DIP supports) for consideration, and naming the module 
after the format, not the database.
<http://psidev.sourceforge.net/mi/xml/doc/user/>

In terms of methods: again, I am not sure you *want* to restrict 
yourself to protein only. MolecularInteraction.pm would suggest itself.


Best regards,


Boris

---
Boris Steipe
University of Toronto
Program in Proteomics & Bioinformatics
Departments of Biochemistry & Molecular and Medical Genetics
http://biochemistry.utoronto.ca/steipe/



> Richard,
>
> You may want to reconsider the name "ProteinGraphIO::xml.pm" since 
> other
> pathway databases (BIND certainly, KEGG perhaps) also export to XML.
>
> Brian O.
>
> -----Original Message-----
> From: bioperl-l-bounces at portal.open-bio.org
> [mailto:bioperl-l-bounces at portal.open-bio.org]On Behalf Of Richard 
> Adams
> Sent: Monday, March 08, 2004 4:31 AM
> To: bioperl-l at portal.open-bio.org
> Subject: [Bioperl-l] protein networks
>
> Hi,
> I've been writing some modules to apply Nathan Goodman's
> "SimpleGraph.pm" module to protein interaction networks. He has
> written this as a replacement for the the broken CPAN graph modules. 
> The
> aim of these modules is
>
> 1. To convert protein interaction data between differentet formats
> 2. To be able to link protein interaction data to protein sequence 
> data.
> 3. To be able to read in interaction data from separate sources, and
> make a merged graph with duplicate interactions flagged or removed.
>
>
>
> Could some core person give some idea for namespace to put these in 
> CVS?
>
> The modules are :
> 1. Nathan's modules
> SimpleGraph.pm                   - a generic graph  module with methods
> to build and represent a graph
> SimpleGraph::Traversal.pm - generic methods to navigate a graph.
>
> 2. My modules
>
> a. An I/O system analagous to SeqIO that lazyloads the correct modules
> for a particular format
>      ProteinGraphIO.pm      - with next_graph, write_graph methods
>        ProteinGraphIO::dip.pm    - parser for DIP format interactions
>          - to read in PSI_XML formatted records
> (still being worked on).
>
> b.   Modules to represent a protein network:
>     PrimaryProteinGraph.pm - subclass of SimpleGraph.pm. This will be
> made from a dip record, analagous to PrimarySeq
>     FullProteinGraph .pm       - subclas of SimpleGraph.pm,  contains
> PrimaryProteinGraph and metadata of graph, will be made from an XML
> file, analagous to RichSeq.
>
> c.  Classes to represent nodes/ edges.
>    NodeI.pm    - interface describing methods that all nodes should 
> have
>         ProteinNode.pm - implements NodeI, currently has methods
> primary_id(), seq(), secondary_id()
>          InteractionI.pm    - interface describing methods all edges
> should have
>         ProteinInteraction.pm - implements InteractionI, currently has
> methods weight(), interactors(), id().
>
> Cheers Richard
>
>
> --
> Dr Richard Adams
> Psychiatric Genetics Group,
> Medical Genetics,
> Molecular Medicine Centre,
> Western General Hospital,
> Crewe Rd West,
> Edinburgh UK
> EH4 2XU
>
> Tel: 44 131 651 1084
> richard.adams at ed.ac.uk
>