[Biopython-dev] First impressions of Pathway

[Tarjei Mikkelsen]

>Quibbles:
>    I'd prefer more neutral  nomenclature than parent-child because they 
>bias
>the reader toward a tree structure.

'parent' and 'child' is only used for the MultiGraph class, which is a 
generic directed graph rep intended for internal representation. It is the 
Network class that is exposed to the user. Network uses 'source' and 'sink' 
as the corresponding terms - although I'm not convinced that is the best 
naming either.

>    df_search seems to assume a connected graph ( at least it looks like it
>would konk out early with a disconnected graph )..  All assumptions should
>be documented.

I've updated the documentation to better reflect what it does

>   The following line needs a description of what each tuple contains?  
>Since
>python is typeless it requires more documentation at the interfaces.
>
>     catalysts   -- list of tuples of catalysts involved in the same 
>reaction
>                      step

This has been simplified to a list of catalysts . The type is arbitrary by 
design - it could be anything from a string descriptor to a Enzyme record 
object, depending on the needs of the user.

>   In MultiNetwork.remove_node the sense of the filter is reversed from 
>what
>I'd expect if you want to remove dangling edges.   My understanding is that
>filter returns items that make the condition true?

Yup, good catch. The whole remove_edges method was faulty. It's been 
corrected and a test has been added.


thanks,

Tarjei


_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp