[BioC] how to store multiple relationships between nodes in graphNEL?

[Seth Falcon]

Paul Shannon <pshannon at systemsbiology.org> writes:

> Hi Seth,
>
>> It seems to me that what you really want is a nice way to represent
>> multigraphs (one node set with a set of edge sets).
>
> You're right:  a multigraph seems like the best solution.  How hard
> would it be
> to build such a class so that it is compatible with the graph class
> and RBGL?

Yes, the goal would be a class that integrates with methods in the
graph and RBGL packages.  As for how hard it will be, I'm not sure.

> If the timing works out, I might be able to help.

That would be great.  I don't expect to be able to take a closer look
at this before the BioC2007 conference, but things could change.

> One small suggestion:  it might be nice if 'directed' could (at least
> optionally)
> be an attribute on each edge, rather than on the graph as a whole.

There are algorithms that work on undirected graphs and algorithms
that work on directed graphs.  Introducing mixed graphs would
complicate matters.  What would be nice about such mixed graphs?

+ seth

-- 
Seth Falcon | Computational Biology | Fred Hutchinson Cancer Research Center
http://bioconductor.org