[Biojava-l] packages under construction

[Matthew Pocock]

Hi Gerald,

Gerald Loeffler wrote:

> Hi!
>
> I'd be very grateful for a little bit of information about the
> org.biojava packages and projects that would help me to get started with
> biojava.
>
> Specifically, I browsed through the org.biojava.bio.seq package and
> found it very expressive but I can't judge at all whether this or any
> other package is nearly finished or under heavy construction! So: Which
> biojava packages are considered approximately stable?
>

We should add some class or package javadoc tag - like @construction,
@stable to indicate this.

The seq package and sub packages are fairly solid. We need to get alignments
finnished (this means gapped sequences), and I guess the feature strategy
may change, but this package is ready for use today.

The dp package is a use-at-your-own-risk affair - it works for me, but is
under active construction.

The GFF package is there to be broken - again, it works for me, but I would
love to hear about any problems people have with it.

All of the other packages contain usefull stuff which if you are interested
in, you should look at, but which may change radicaly in the future.

>
> Additionally, is there any agreement that all contributions to biojava
> should adhere to the object model for sequences and annotations defined
> by org.biojava.bio.seq or any of the other "Biojava Core Packages"? (In
> other words, is biojava simply collecting any bio-related Java
> contributions or is it rather trying to build a consistent whole based
> upon the "Biojava Core Packages").

I hope that BioJava will grow to encompas many areas of informatics -
expression data, gene networks, ontologies, genomes - and also cover basic
tools - stats, threading, xml io, visualisation. The web site should contain
links to any and all related java bio resources. I think to make the whole
usefull, we have to define a consitent but flexible framework (like sequence
interfaces), and then expect other code in the repository to either use
these directly, or bridge to-from them (e.g. the Ace client uses the
sequence interface directly, but the GFF module uses its own concept of a
feature and bridges to-from the seq.Feature interface). Hopefuly if we get
the core interfaces close enough then this won't cramp anyone's style.

>
>
>         thanks for your help,
>         cheers,
>         gerald
> --
> Gerald.Loeffler@vienna.at
> Tel.: +43 676 3289588
>
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l@biojava.org
> http://biojava.org/mailman/listinfo/biojava-l