[Biojava-l] A software project suggestion

[Mark Schreiber]

Hi Martin -

Something that would be useful to have is a parser and object model
for Entrez Gene, mainly the XML format.  This format is available by
CGI-BIN, SOAP and FTP and contains a great deal of useful information.
The project is possibly a bit more challenging than it first may seem
as the format is arcane to say the least. You will not get anything
terribly useful by autogenerating something with JAXB.  Also the SOAP
WSDL doesn't work out of the box with JAX-WS, you can use AXIS but
again the autogenerated object binding to the XML is rubbish (due to
the very confusing XML structure).  If you made a parser for the SOAP
service you would be best to go to a lower level (such as SAAJ).

Anyhow the main idea is to get all of the tremendously useful
information extracted from the XML and into a friendly Java beans API.

- Mark

On Tue, Dec 8, 2009 at 3:54 AM, Martin Petr <martin.petr at matfyz.cz> wrote:
>
> Hi everybody,
>
> I'm a computer science student currently in the last year of bachelor
> studies and I'm looking for an interesting software project for my
> Java course. And since I also happen to study molecular biology (I'm
> just in the first semester now, so there is a long way ahead of me)
> and I'm very interested in bioinformatics, I decided to ask here for
> suggestions.
>
> Do you have any ideas for a possible BioJava related project? Do you
> miss any functionality in BioJava that I could add? I have to say that
> my knowledge of bioinformatics is very vague (although I have quite a
> solid background in general computer science, at least) but I guess
> that shouldn't be a big problem.
>
> I'm not talking here anything PhD level-like, not even BSc level-like,
> it may be just some "boring" technical stuff that needs to be done. I
> would just prefer to help and do something more useful, which really
> can't be said about a zilionth clone of IRC bot or something like
> that. :)
>
> I take it as a good opportunity to learn something about BioJava
> itself, since it very well may be my tool of choice when I finally get
> a chance to get my hands dirty in some research! In fact, that's why I
> got the idea to help BioJava in the first place.
>
> Thanks in advance for any replies and suggestions. Have a nice day.
>
> Martin Petr
> Charles University in Prague
> Czech Republic
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/biojava-l