Re: [Treebase-devel] application-to-TreeBASE direct submission?

[William P. <wil...@ya...>]

On Nov 4, 2011, at 6:53 AM, Arlin Stoltzfus wrote:

> I mentioned that TreeBASE has a top-secret interface that Dryad uses  
> to submit NEXUS files, and that this could be the basis for a  
> submission interface for other applications.  My understanding is that  
> this is done via web-services, and that the user gets a link to a  
> temporary submission that must be completed interactively.  I hope I  
> didn't give the wrong impression.

Although note that is is really basic and primitive: it just saves the user the pain of retyping the citation; saves the user re-uploading the datasets; and it creates mutual linking IDs between the Dryad submission and the TreeBASE submission. But nothing fancier than that.. 

What would surely be useful is if we had an XML/OWL/RDF standard for describing any kind of analysis parameters in a generic fashion. That was one of the goals of CIPRES: a common communication format than any software program could be taught to understand and socket into, and one that TreeBASE could store, or that Kepler could incorporate in a pipeline. Rutger's NeXML had its origins in that, but only so far as replicating what the universal blocks in NEXUS could store. The ability to describe substitution parameters or tree search strategies in generic fashion did not materialize out of CIPRES.

Short of this, TreeBASE has a primitive and generic way of storing a list of commands for a program -- i.e. as a text blob -- the downside being that it's only interpretable by the program that it's designed for, and it may or may not be valid (usually not valid -- e.g. when taxa are referenced with numbers, yet the order of taxa in the taxa block are easily shuffled). So a low-hanging-fruit solution is to have programs like MEGA express NeXML along with a text-blob of whatever MEGA-commands were used between the input data and the output data. TreeBASE would be taught to ingest this, validate the NeXML, and then populate the analysis records with the MEGA-command blob. 

(In the meantime, I've been hoping to find time to write up instructions for submitting data to TreeBASE from MEGA -- it's not difficult, but many people are still naive as to how to do this). 

At any rate, thanks for pursuing these valuable conversations with Sudhir, etc. 

bp