BioJava code freeze, modularization and action items for sub modules

[Andreas Prlic-3]

Hi,

In order to push the modularization and migration to Maven, I would like to
declare a code freeze on the current developmental trunk. Please commit all
new changes by

Thursday 27th of August 23:00 GMT.

In the week after I would like to refactor the code base and commit the
initial set of modules to a new developmental trunk.  All future development
will happen on that new trunk.

You will be able to follow the ongoing status of this at

http://biojava.org/wiki/BioJava:MavenMigration


Once the modules are in place it is a good moment to hand over the
leadership of the sub-modules to the new module-project leaders. It will be
up to the module-lead to take the modules into the direction that he/she
feels important. I would like to take this opportunity to suggest a couple
of people as module-leaders and propose some action items for the modules.
Feel free to comment or make additional suggestions...

Here a list of modules / action items and the people that I would propose to
become module leaders:

Module: biojava-core Lead: Andreas Prlic
 - break the new modules out of core
 - bring up to modern Java standards, use Generics
 - declare old/unused code obsolete
 - don;t break backwards compatibility

Module: biojava-sequence Lead: Richard Holland
 - Bring in Richard's new code that he started to develop on the biojava-3
branch.
 - provide a more scaleable and efficient basis for dealing with large
sequence files

Module: biojava-alignment Lead: Andreas Draeger
 - allow better access to underlying dynamic programming data structures
 - allow more customizable display of pairwise alignments (HTML/plain text,
etc)

Module : biojava-blast Lead: still looking for a leader
 - provide access to all details of the blast output
 - add support for RPS blast

Module: biojava-phylo Lead: Scooter Willis
 - provide improved NJtree /Jalview

Module: biojava-biosql Lead: Richard Holland
 - merge the new biojava-sequence module with the current biojava-biosql
code


Module: biojava-structure Lead: Andreas Prlic
 - add support for SCOP file parsing
 - add support for easy access of domains (in terms of coordinates)
 - add secondary structure assignment
 - improve structure alignments
 - better integration with 3D viewers (Jmol, RCSB viewers)

Module: biojava-web services:
The details seem still to be under discussion and perhaps we need multiple
modules here?
also what about REST vs. SOAP? To be discussed. People who expressed
interest are:
Niall Haslam,Scooter Willis, Sylvain Foisy

Module?: biojava-ws-blast
Module?: biojava-ws-biolit

Module: biojava-sequencing Lead: ???
  - support FastQ files
 - support parsing of output for various new sequencing machines

This is only an initial set of modules and I think it is safe to say that
more modules will be added after more discussions (and people volunteering
to contribute).

Andreas
_______________________________________________
Biojava-l mailing list  -  Biojava-l@...
http://lists.open-bio.org/mailman/listinfo/biojava-l

[Michael Heuer]

Andreas Prlic wrote:

> Here a list of modules / action items and the people that I would propose to
> become module leaders:
> ...
>
> Module: biojava-sequencing Lead:  Michael Heuer
>   - support FastQ files
>   - support parsing of output for various new sequencing machines

I have volunteered on the open-bio mailing list to implement FASTQ
support.  A nice collection of test data is being created in collaboration
with the other open-bio projects.  If anyone has interest in a particular
data set, please let me know, as I will also need data for performance
tuning.

   michael

_______________________________________________
Biojava-l mailing list  -  Biojava-l@...
http://lists.open-bio.org/mailman/listinfo/biojava-l

[Andreas Prlic-3]

Great, thanks for "volunteering", Michael.

To add another Module:

biojava-das : Lead: Jonathan Warren
probably deprecate the old DAS code in BJ and replace it with
the up to date Dasobert library

Thanks to Jonathan for volunteering as well.

Andreas




On Thu, Aug 27, 2009 at 10:01 AM, Michael Heuer<heuermh@...> wrote:

> Andreas Prlic wrote:
>
>> Here a list of modules / action items and the people that I would propose to
>> become module leaders:
>> ...
>>
>> Module: biojava-sequencing Lead:  Michael Heuer
>>   - support FastQ files
>>   - support parsing of output for various new sequencing machines
>
> I have volunteered on the open-bio mailing list to implement FASTQ
> support.  A nice collection of test data is being created in collaboration
> with the other open-bio projects.  If anyone has interest in a particular
> data set, please let me know, as I will also need data for performance
> tuning.
>
>   michael
>
> _______________________________________________
> biojava-dev mailing list
> biojava-dev@...
> http://lists.open-bio.org/mailman/listinfo/biojava-dev
>

_______________________________________________
Biojava-l mailing list  -  Biojava-l@...
http://lists.open-bio.org/mailman/listinfo/biojava-l

[Mark Schreiber-2]

I'm happy to volunteer code for:


   1. BLASTXML parser as long as I can change the ssbind APIs (other parsers
   could go into a legacy module??). Actually I would prefer to completely
   decouple from the sequence/ feature module as many people would like a blast
   parser without the rest of biojava thrown in.
   2. BioSQL/ JPA bindings. I have already generated JPA compliant entity
   beans for mapping to BioSQL as well as JPA handler code that makes sure
   modifications presist properly. Currently the object model very closely
   follows the BioSQL table structure.  Also the current beans are what people
   call Anaemic beans in that they hold data and provide getters and setters
   but no biological behaivour. I can easily provide bio-smarts to the beans
   but it might be better to hold off until there is a module that contains
   sequence/feature interfaces which the beans could implement.
   3. Happy to provide code for an enterprise module if there is sufficient
   interest. This would probably take the form of SessionBeans and WebServices
   that can be deployed to Glassfish/ JBoss etc to provide biological services
   for people who want to make client server or SOA apps.

- Mark


On Fri, Aug 28, 2009 at 1:30 AM, Andreas Prlic <andreas@...> wrote:

> Great, thanks for "volunteering", Michael.
>
> To add another Module:
>
> biojava-das : Lead: Jonathan Warren
> probably deprecate the old DAS code in BJ and replace it with
> the up to date Dasobert library
>
> Thanks to Jonathan for volunteering as well.
>
> Andreas
>
>
>
>
> On Thu, Aug 27, 2009 at 10:01 AM, Michael Heuer<heuermh@...> wrote:
> > Andreas Prlic wrote:
> >
> >> Here a list of modules / action items and the people that I would
> propose to
> >> become module leaders:
> >> ...
> >>
> >> Module: biojava-sequencing Lead:  Michael Heuer
> >>   - support FastQ files
> >>   - support parsing of output for various new sequencing machines
> >
> > I have volunteered on the open-bio mailing list to implement FASTQ
> > support.  A nice collection of test data is being created in
> collaboration
> > with the other open-bio projects.  If anyone has interest in a particular
> > data set, please let me know, as I will also need data for performance
> > tuning.
> >
> >   michael
> >
> > _______________________________________________
> > biojava-dev mailing list
> > biojava-dev@...
> > http://lists.open-bio.org/mailman/listinfo/biojava-dev
> >
>
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l@...
> http://lists.open-bio.org/mailman/listinfo/biojava-l
>

_______________________________________________
Biojava-l mailing list  -  Biojava-l@...
http://lists.open-bio.org/mailman/listinfo/biojava-l

[Andreas Prlic-3]

Thanks, Mark.

Guess we should start collecting all this info on a wiki page. I started to edit
http://biojava.org/wiki/BioJava:Modules

module leaders: feel free to edit the plans for your module...

Andreas


On Thu, Aug 27, 2009 at 10:37 PM, Mark
Schreiber<markjschreiber@...> wrote:
- Show quoted text -

On Thu, Aug 27, 2009 at 10:37 PM, Mark
Schreiber<markjschreiber@...> wrote:

> I'm happy to volunteer code for:
>
> BLASTXML parser as long as I can change the ssbind APIs (other parsers could
> go into a legacy module??). Actually I would prefer to completely decouple
> from the sequence/ feature module as many people would like a blast parser
> without the rest of biojava thrown in.
> BioSQL/ JPA bindings. I have already generated JPA compliant entity beans
> for mapping to BioSQL as well as JPA handler code that makes sure
> modifications presist properly. Currently the object model very closely
> follows the BioSQL table structure.  Also the current beans are what people
> call Anaemic beans in that they hold data and provide getters and setters
> but no biological behaivour. I can easily provide bio-smarts to the beans
> but it might be better to hold off until there is a module that contains
> sequence/feature interfaces which the beans could implement.
> Happy to provide code for an enterprise module if there is sufficient
> interest. This would probably take the form of SessionBeans and WebServices
> that can be deployed to Glassfish/ JBoss etc to provide biological services
> for people who want to make client server or SOA apps.
>
> - Mark
>
>
> On Fri, Aug 28, 2009 at 1:30 AM, Andreas Prlic <andreas@...> wrote:
>>
>> Great, thanks for "volunteering", Michael.
>>
>> To add another Module:
>>
>> biojava-das : Lead: Jonathan Warren
>> probably deprecate the old DAS code in BJ and replace it with
>> the up to date Dasobert library
>>
>> Thanks to Jonathan for volunteering as well.
>>
>> Andreas
>>
>>
>>
>>
>> On Thu, Aug 27, 2009 at 10:01 AM, Michael Heuer<heuermh@...> wrote:
>> > Andreas Prlic wrote:
>> >
>> >> Here a list of modules / action items and the people that I would
>> >> propose to
>> >> become module leaders:
>> >> ...
>> >>
>> >> Module: biojava-sequencing Lead:  Michael Heuer
>> >>   - support FastQ files
>> >>   - support parsing of output for various new sequencing machines
>> >
>> > I have volunteered on the open-bio mailing list to implement FASTQ
>> > support.  A nice collection of test data is being created in
>> > collaboration
>> > with the other open-bio projects.  If anyone has interest in a
>> > particular
>> > data set, please let me know, as I will also need data for performance
>> > tuning.
>> >
>> >   michael
>> >
>> > _______________________________________________
>> > biojava-dev mailing list
>> > biojava-dev@...
>> > http://lists.open-bio.org/mailman/listinfo/biojava-dev
>> >
>>
>> _______________________________________________
>> Biojava-l mailing list  -  Biojava-l@...
>> http://lists.open-bio.org/mailman/listinfo/biojava-l
>
>

_______________________________________________
Biojava-l mailing list  -  Biojava-l@...
http://lists.open-bio.org/mailman/listinfo/biojava-l