I don't know how it is stored... Looking at the code (http://doc.bioperl.org/releases/bioperl-current/bioperl-live/Bio/DB/SeqFeature/Store/DBI/mysql.html , http://doc.bioperl.org/releases/bioperl-current/bioperl-live/Bio/DB/SeqFeature/Store.html#CODE82 , http://doc.bioperl.org/releases/bioperl-current/bioperl-live/Bio/DB/SeqFeature.html) doesn't help me too much as I don't really understand what is going on there...
Maybe I could write a web-service to provide the information I need...??
| [reply] |
This isn't a bad idea - you might consider delivering the data as JSON; I think that should be something that you could find a Java library to parse. It completely insulates the Java consumer from having to know anything about the data's on-disk format, which is the point here.
You may need to provide something to impedance-match the Java code and the returned JSON, and supply that to the Java folks, most likely as another .jar file.
I think this is really your best option, since it should be quite easy to do. You will need to write the BioPerl-to-JSON code to add to your server and the JSON-to-whatever-format library for the Java code, but those are both smallish and easily-testable pieces of code.
| [reply] |