international online community for dna barcoding professionals
After you aided with our previous problems so nice and quickly (thanks a lot!!) - here's a new one ;-):
We are implementing the Biocode plugin (I use 2.3.3 with Geneious 5.6.3) for our Sponge Barcoding Project. However we do not quite understand some results we get from the "verify taxonomy" function. We have a FIMS with taxonomic information (see screenshot) but verifying the taxonomy with the biocode plugin provides:
- an empty field for the "Hit taxon" (it should be the NCBI-result taxonomy here I guess)
- a query-taxon field with the NCBI-Taxonomy instead of our FIMS-taxonomy (we do not use the taxon "Ceractinomorpha" in our Database).
Would you have a suggestion what is wrong here or is there a bug in the plugin?
BTW, why do we get a high "Assembly bin" value for a single sequence? I see the same in the example picture on your biocode page, where single .abi traces get "Assembly bin" values. see http://software.mooreabiocode.org/index.php?title=File:Verify_taxon...
Thanks a lot in advance for your help - as usually highly appreciated!
Thanks for the nice feedback - we always like to solve peoples' problems as quickly as possible.
Both of your problems are labelling issues
1) the empty field for hit taxon happens when you have not checked the "fully annotate hit summaries (slow)" checkbox in the options.
We use our sequence search service to do the BLAST part of the verify taxonomy operation, so the options you are seeing come directly from there - we really should make that checkbox more descriptive or get rid of it altogether since you would always want to fully annotate the hits for verify taxonomy. I'll see what I can do for future versions.
2) The column always says "Assembly Bin" since we don't keep a record of whether we were BLASTing a conensus sequence or a standalone document. It's a little clear as it is, but in your case it just means the bin of the individual sequence you're checking.
Please let me know if you have any further questions
Hi Steven - you did it again!
Thanks for the quick reply. Yes, it works nicely now. However, we were also a bit puzzled by the fact that Biocode fills up the higher taxonomic fields from NCBI in case they are not in the FIMS. In case that this not computationally necessary I would suggest to somehow avoid that as the NCBI taxonomy is pretty slow in changes and invalid taxa persist for ages somehow.
Anyway - here's another question: ;-)
Each time we start geneious we receive an error report on the connection to the FIMS (screenshot) and when we log in to biocode we have to specify all plate information, Tissue and Sample ID, and taxonomy fields again (set to "none"), and I guess our settings should be saved somehow. Any suggestion what we do wrong?
Thanks again and best regards,
I've got to the bottom of your problems - they were both cause by the same issue where the plugin couldn't load the MySQL driver file unless a connection had already been made to a database.
I'll get a fix out (hopefully in a couple of days), but until then a workaround is to check the "connect on startup" checkbox in the bottom left hand corner of the connection dialog. That will make sure that the MySQL driver is loaded correctly.
Biocode fills up the higher taxonomic fields from NCBI in order to do the binning comparison for verify taxonomy, but there is no real need to write this back out to the sequences - we were just doing it as a convenience. If it's counter-productive as you suggest then we'll remove it from future versions.
thanks, "connect at start up" solves the driver problem, but the issue with the error report on the connection to the FIMS (see Dirk on May 31, 2012 at 9:39am above) still persists. Can't wait for the fix ;-)
Keep up the good work -- best
Hi Dirk and Gert,
Sorry for the late reply - I've been travelling all this week so it was hard to check my messages.
In answer to your question, we fill in the taxonomy values from NCBI in order to help with the comparison for binning, but we only write the updated values to the document as a convenience. As you've pointed out, it's often not helpful, so I think the best thing for us to do is stop it writing the updated taxonomy out (but still using it for the binning calculations). This will be updated in the next plugin version.
The problem with your FIMS sounds like a bug. Would you be able to send me the following, to help me diagnose the problem?
* The contents of the folder named "com.biomatters.plugins.biocode.BiocodePlugin" inside your Geneious user directory
* The name of the connection that is giving you trouble
* A screenshot of the whole connection manager window, when it is configured correctly (so I can see what the connection options are supposed to look like)
The plugin not recognising the driver file location is indeed interesting - I will look into the issue and report back in a couple of days once I know more.