[Bioperl-l] Interest in Bio::Community modules

[Florent Angly]

Hi Chris,

On 29/11/11 05:47, Fields, Christopher J wrote:
> I think the idea is sound, it would be nice to have.  Jason is working a bit in this area, maybe he has some additional thoughts?  Would there be some redundancy with any current code (Bio::Tree, Bio::Taxon, etc)?
None of these features would be duplicated. Rather, they would be used 
attributes of the Bio::Community::* objects. For example, a member of a 
community could have a Bio::SeqI attached to it as well as a Bio::Taxon, 
etc...

> I do think it should be developed on it's own, per our recent discussions re: slimming down core.
Yes, the features are so different that it makes sense to have the 
Bio::Community::* modules as a separate BioPerl distribution (like the 
Bio-FeatureIO BioPerl distribution).

> Re: using Moose and BioPerl, from experience it's a little tricky integrating the two as BioPerl uses perl's simple OO system while Moose allows much more (though it has it's own issues, memory and speed being the top two).  Specifying inheritance from a BioPerl module is a bit trickier, you'll have to wrap the module via MooseX::NonMoose.  Beyond that, I think if you stick with using is-a/has-a based relations for the time being when defining interactions with other BioPerl modules you'll be fine.
Beyond Bio::RootI, I cannot imagine that any of the Bio::Community::* 
modules would need to inherit from any other BioPerl modules. 
Considering this and the performance aspects of Moose, do you think that 
using Moose is a wise design decision?

Best,

Florent


> chris
>
> On Nov 27, 2011, at 10:35 PM, Florent Angly wrote:
>
>> Hi all,
>>
>> I have been thinking about starting a set of Perl modules that would useful for (microbial) ecologists to represent communities of organisms. At the moment, there does not seem to be anything like this in Bioperl. I am happy to make these modules available under the Bioperl umbrella using the Bio::Community::* namespace.
>>
>> I envision the following modules:
>> * Bio::Community::Member module representing members of a community.
>> * Bio::Community::IO modules to read/write files that describe community composition (a.k.a. OTU table, or site by species table) as used programs like QIIME, Pyrotagger, GAAS, ...
>> * Bio::Community::Tools modules to help manipulate communities, e.g. to take some members at random, normalize the community to a given number of individuals, or do rarefaction curves.
>>
>> The idea is to implement these modules in Moose to teach myself Moose. The members of a community could be a sequence (Bio::SeqI), a species (Bio::S), an arbitrary string or even other things. I am not quite sure if Bioperl provide facilities to attach some arbitrary information to an object.
>>
>> Any interest? Ideas? Comments?
>>
>> Thanks,
>>
>> Florent
>>
>> _______________________________________________
>> Bioperl-l mailing list
>> Bioperl-l at lists.open-bio.org
>> http://lists.open-bio.org/mailman/listinfo/bioperl-l