Thanks for the answer, Ed.
An update on that: I tried this on my computer at home running Java
1.5.0 (instead of 1.6.0), an there the coordinates did get filled with
the correct values. So I tried the same in the lab, ans again the
coordinates were right this time. On the other hand, a Mac Mini here in
the lab (running Java 1.5.0) also got the coordinates wrong. So I guess
this is a finicky interaction and I'll just have to warn my
collaborators to check the coords before submitting any BLAST from
Apollo web-start...
Anyway, in all cases (home computer or lab Mac Mini or Linux), the error
about the NCBI_BlastOutput.dtd file still occurs after the analysis is
(presumably) done.
Thanks
J
Ed Lee wrote:
Hi J,
The NCBI site seems currently down at the moment, so I can't test out
the remote analysis. But the analysis window from Web Start is
displaying the correct coordinates, so I'm not sure why you're
getting the wrong coordinates...
I'll give more feedback once I can access the qBLAST server.
Cheers,
Ed
On Wed, 6 Jan 2010, J.M.P. Alves wrote:
Hi again,
The problem with the analysis results happens when using the web-start
version, but not the standalone local version -- see my email from a
few minutes ago for the details on my setup.
When I select one annotation, right click and select "analyze region"
on the standalone version, it works perfectly (at least for the
NCBI-BLAST analysis, which is all I've tried so far).
When I do the same using the web-start one, some problems occur:
- first, the range is wrong in the genomic region start an end fields,
it seems to get only the stop codon coordinates (e.g. 33347 to 33349,
instead of the correct 33347 and 36331).
Obnoxious, but not show stopping -- the users can always correct that
manually if I warn them beforehand of the problem. But unfortunately
it does not stop there.
After correcting the coordinates, I click on run, and Apollo tells me
it will take about X seconds to finish.
A while later, it give the following error message:
"Error running analysis: Relative URI "NCBI_BlastOutput.dtd"; can not
be resolved without a base URI"
The results do not get added to the analysis panel, I'm assuming they
don't even get parsed...
It appears to me that the file in question is sent by NCBI to Apollo,
is that right? I have used grep and find on all the Apollo directories
I could think of, and found nothing that seemed obvious to me.
Any idea how to solve this? Did I do anything wrong when I created my
web-start instance? To compile Apollo for the web-start, all I did was
copy my edited chado-adapter.xml, .style, and .tiers files to the conf
directory in the source Apollo distribution, and then run "ant jar"
from src/java. Did I miss anything?
Thanks
J
--
-------------------------------
João Marcelo Pereira Alves (J)
Post-doctoral fellow
MCV / VCU - Richmond, VA
http://bioinfo.lpb.mic.vcu.edu
f. 1-804-828-3897
_______________________________________________
apollo mailing list
[email protected]
http://mail.fruitfly.org/mailman/listinfo/apollo
--
-------------------------------
João Marcelo Pereira Alves (J)
Post-doctoral fellow
MCV / VCU - Richmond, VA
http://bioinfo.lpb.mic.vcu.edu
f. 1-804-828-3897
_______________________________________________
apollo mailing list
[email protected]
http://mail.fruitfly.org/mailman/listinfo/apollo
