Hi Peter,

There are still no plans to add this into the main data model, but it should be possible to add in a sub-class to your own instance. If you need help details - please let us know,

Thanks,

Jen Galaxy team

On 10/5/10 7:34 AM, Peter wrote:

Hi all,

For several tools it is useful to restrict input to just protein FASTA files, or just nucleotide FASTA files (or even just DNA or just RNA).

Galaxy has the concept of file types, e.g. "tabular" or "fasta". Is there any way to add additional meta-data on top of this, or file sub-types?

I remember asking James Taylor about this particular example (types of FASTA file) at a conference a few years ago (probably BOSC/ISMB), and I recall at the time he said there were no plans in this direction... is that still true?

Thanks,

Peter _______________________________________________ galaxy-dev mailing list galaxy-dev@lists.bx.psu.edu http://lists.bx.psu.edu/listinfo/galaxy-dev

Hi Peter,

I don't understand how a sub-class would work here - an example would be great.

See GFF and GTF for an example of how subclassing datatypes works; also, the BED classes in lib/galaxy/datatypes/interval.py use subclasses as well. Typically, additional metadata and functionality is provided in the subclass.

In addition to the type of sequence in FASTA (DNA, RNA, or protein) it struck me that the variant of FASTQ file (Sanger, Solexa, Illumina 1.3+) would also be a situation which might benefit from file sub-types in Galaxy.

Perhaps. Because there are so many types of FASTA files, careful thought is needed about how to organize the many FASTA types so that they are useful both to users and to Galaxy. It might be good to come up with an idea of how things would work and post back to galaxy-dev to get some feedback before implementation.

J.