ARGOT2: a new tool for function annotation

[enrico]

Hi all,

here in Padova we have developed a new tool for protein/gene annotation called ARGOT2 (the first version has been published on PlosOne, Fontana et al 2009).
We would like to make it available to everybody interested, in order to get some feedbacks useful for an incoming publication. The web server is to be considered in its beta-release so some bugs may still be present
You can find the web server, with a tutorial explaining all sections, at:

http://www.medcomp.medicina.unipd.it/Argot2/index.php

Feel free to contact me if you have questions, suggestions or encountered problems

Best,

Enrico

[jmw86069]

Enrico, thank you for posting the link to the tool and reference, I'm interested in testing it out and providing feedback (if I have anything helpful to say!)

It looks to me that Argot2 is functionally similar to "annot8r" but with a layer of GO graph trimming which I haven't seen in annot8r. (But I could be missing something in my use of annot8r.) I'm intrigued to see what the trimmed and grouped results look like. Have you compared with annot8r, and if so what were your thoughts?

In the publication (which refers to argot not Argot2) it suggests the code is downloadable, but I didn't see a link to download Argot2. I may want to run a whole eukaryotic genome's worth of predicted genes through Argot2, but I'm not sure I'm ready to post it all into the tool at this point. Is it possible yet to download the source and run on my server? I understand if it's not feasible yet, but I wanted to ask just in case.

Either way, I'm looking forward to testing it out! Thanks again for posting about it.

[enrico]

ARGOT2 is not downloadable, because it requires a heavy management of mysql database in order to populate the tables.

But we have implemented a batch section instead (see "Batch processing" on the left menu). In this section you can upload your own BLAST and HMMer results (follow the tutorial for generating them in the suitable format) and you get annotations in a few minutes. Remember to zip the files before uploading.

In our tests we get annotations for 3000 proteins in about 5 minutes, so we think you could upload an entire set of eukariota genes and get the results within an hour. The results will be in tabular format (tsv). You just need to run BLAST and HMMer by yourself, because we have not enough computing resources.

As regards annot8r, we have not compared it with our tool yet, but we will do it soon. ARGOT2 relies on BLAST and HMMer searches, not on KEGG, EC-numbers or others at the moment, but we plan to add some of them in the future.

Thanks for replying!

[enrico]

Argot2 is now stable!

Hi everybody,
I just want to announce that Argot2 is now stable and ready for annotating small sets of sequences up to entire genomes, you can find it at:

http://www.medcomp.medicina.unipd.it/Argot2/index.php

Let me know if you encounter any problem or have some questions.
Regards

Enrico

[enrico]

The paper on Argot2 is now published on BMC bioinformatics:

http://www.biomedcentral.com/1471-2105/13/S4/S14

The tool was among the top performers at CAFA challenge (Critical Assessment of Function Annotations)