spurious flags at five prime and three prime UTRs since update?

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

spurious flags at five prime and three prime UTRs since update?

Sheldon McKay
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon



Screen Shot 2013-09-11 at 10.11.30 AM.png (25K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: spurious flags at five prime and three prime UTRs since update?

Ed Lee
Hi Sheldon,

Which version of WebApollo are you running?  The 2013-09-14 release?  I can't replicate the issue you're having.  If you're using the latest release, then could you provide with a small GFF3 (just an annotation or two should suffice) so that I can see if I can recreate this problem?

Cheers,
Ed


On Wed, Sep 11, 2013 at 7:19 AM, Sheldon McKay <[hidden email]> wrote:
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon



Reply | Threaded
Open this post in threaded view
|

Re: spurious flags at five prime and three prime UTRs since update?

Chris Childers
I saw this issue today, and I am using the 2013-08-30 release. 

In my case I have two identical models and making one an annotation causes the flags and the other does not.  Zooming in to the base level confirmed that these are using the same start for the CDS. If I manually set the transition start site, the flags go away.

Thanks,
Chris


On Wed, Sep 11, 2013 at 3:50 PM, Ed Lee <[hidden email]> wrote:
Hi Sheldon,

Which version of WebApollo are you running?  The 2013-09-14 release?  I can't replicate the issue you're having.  If you're using the latest release, then could you provide with a small GFF3 (just an annotation or two should suffice) so that I can see if I can recreate this problem?

Cheers,
Ed


On Wed, Sep 11, 2013 at 7:19 AM, Sheldon McKay <[hidden email]> wrote:
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon






Reply | Threaded
Open this post in threaded view
|

Re: spurious flags at five prime and three prime UTRs since update?

Ed Lee
Hi Chris,

Could you send me a minimal GFF3 with both the working evidence model and the identical one that fails?  Please include the genomic sequence as well.  You might want to try the 2013-09-04 release (the 2013-08-30 is deprecated).

Cheers,
Ed


On Wed, Sep 11, 2013 at 3:50 PM, Chris Childers <[hidden email]> wrote:
I saw this issue today, and I am using the 2013-08-30 release. 

In my case I have two identical models and making one an annotation causes the flags and the other does not.  Zooming in to the base level confirmed that these are using the same start for the CDS. If I manually set the transition start site, the flags go away.

Thanks,
Chris



On Wed, Sep 11, 2013 at 3:50 PM, Ed Lee <[hidden email]> wrote:
Hi Sheldon,

Which version of WebApollo are you running?  The 2013-09-14 release?  I can't replicate the issue you're having.  If you're using the latest release, then could you provide with a small GFF3 (just an annotation or two should suffice) so that I can see if I can recreate this problem?

Cheers,
Ed


On Wed, Sep 11, 2013 at 7:19 AM, Sheldon McKay <[hidden email]> wrote:
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon







Reply | Threaded
Open this post in threaded view
|

Re: spurious flags at five prime and three prime UTRs since update?

Sheldon McKay
In reply to this post by Ed Lee
Hi Ed,

I am using the 9/4/13 release.  I have attached sample GFF and fasta files, plus a screen-capture of the region shown in webapollo.  Looking through the GFF exported from the webapollo annotation track, there are non-canonical three and five prime splice sites that are not in the original gene models.  It looks like webapollo is trying to find splice donor/acceptor sites at the start and stop codons.  I did not edit the canonical splice donor/acceptor sites in config.xml.

I also have an instance of the 5/16/13 release, where the error does not happen.

Sheldon


On Wed, Sep 11, 2013 at 4:50 PM, Ed Lee <[hidden email]> wrote:
Hi Sheldon,

Which version of WebApollo are you running?  The 2013-09-14 release?  I can't replicate the issue you're having.  If you're using the latest release, then could you provide with a small GFF3 (just an annotation or two should suffice) so that I can see if I can recreate this problem?

Cheers,
Ed


On Wed, Sep 11, 2013 at 7:19 AM, Sheldon McKay <[hidden email]> wrote:
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon





apollo_example.tar.gz (38K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: spurious flags at five prime and three prime UTRs since update?

Ed Lee
In reply to this post by Chris Childers
Hi Chris and Sheldon,

I've figured out and fixed the bug.  This arose in this version since we're no longer using our own modified version of the data generation pipeline (but rather the stock JBrowse) so some of the necessary massaging of the data isn't getting done.  The fix will be available in the next release.  In the mean time, you can download the patch and update your instance:


Uncompress the tarball and run utr_patch/apply_patch.sh WEBAPOLLO_TOMCAT_DIR where WEBAPOLLO_TOMCAT_DIR is the path to where WebApollo is installed (e.g., /var/lib/tomcat7/webapps/WebApollo).

Let me know if this resolves this issue.

Cheers,
Ed 


On Wed, Sep 11, 2013 at 3:50 PM, Chris Childers <[hidden email]> wrote:
I saw this issue today, and I am using the 2013-08-30 release. 

In my case I have two identical models and making one an annotation causes the flags and the other does not.  Zooming in to the base level confirmed that these are using the same start for the CDS. If I manually set the transition start site, the flags go away.

Thanks,
Chris



On Wed, Sep 11, 2013 at 3:50 PM, Ed Lee <[hidden email]> wrote:
Hi Sheldon,

Which version of WebApollo are you running?  The 2013-09-14 release?  I can't replicate the issue you're having.  If you're using the latest release, then could you provide with a small GFF3 (just an annotation or two should suffice) so that I can see if I can recreate this problem?

Cheers,
Ed


On Wed, Sep 11, 2013 at 7:19 AM, Sheldon McKay <[hidden email]> wrote:
Hi all,

I am getting errors flagged in gene models that were fine in the previous version.  Typically it is at the start and/or stop codons when there are UTRs (please see attached screen shot).  If I edit the 5' exon to begin at the start codon, not the 5' end of the UTR, the flag goes away.  I tried reloading the annotations with and without stop codons, CDSs, etc without success.  Basically, every single curated gene model I pull into the editing track is reporting errors where none were reported in the May 16 release.  

Sheldon