search

geneontology / go-site

A collection of metadata, tools, and files associated with the Gene Ontology public web presence.
http://geneontology.org
BSD 3-Clause "New" or "Revised" License
45 stars 89 forks source link

export-lego-to-legacy Jenkins job is failing #168

Closed cmungall closed 7 years ago

cmungall commented 8 years ago

https://build.berkeleybop.org/view/GO/job/export-lego-to-legacy/ all red balls.

Null pointer exception in Minerva

hdietze commented 8 years ago

The Jenkins job is now working again. The export is now available as GAF and GPAD. We now create specific files for taxon/mod groups.

ukemi commented 8 years ago

This is great. Where are the GAF files located?

ukemi commented 8 years ago

I found them in Jenkins, but will need a stable (PURL?) place to get them for import into our system. We will need to filter the models so that only ones that are not templates and are annotated as 'production' are used. Great progress!

hdietze commented 8 years ago

+1 Yes we need to create PURLs. Hosting files and hard coding Jenkins URLs is not stable nor future proof

The filter for production only models already exists. Only MGI has models marked as production and the preliminary URL: https://build.berkeleybop.org/job/export-lego-to-legacy/lastSuccessfulBuild/artifact/legacy/gaf/production/mgi.gaf

ukemi commented 8 years ago

Hi Heiko,

This is looking good. I Just added another model to production. There are still a couple of small issues.

Ceacam20 Mmus-- These should just be Ceacam20 etc. Any progress on making the annotations to folded terms (my columns might be off)? Example #1 with regulation: MGI MGI:96537 Il10 Mmus GO:0003674 PMID:26195794|gomodel:568b0f9600000294 IMP MGI:1857199 F gene taxon:10090 20160119 GO_Noctua negatively_regulates(GO:0006954),occurs_in(UBERON:0001155)

Should be: MGI MGI:96537 Il10 GO:0050728 PMID:26195794|gomodel:568b0f9600000294 IMP MGI:1857199 F gene taxon:10090 20160119 GO_Noctua occurs_in(UBERON:0001155)

Example #2 with an additional annotation: MGI MGI:99515 Syk Mmus GO:0004713 PMID:26195794|gomodel:568b0f9600000294 IDA F gene taxon:10090 20160119 GO_Noctua occurs_in(CL:0011108),positively_regulates(GO:0007249)

Should be: MGI MGI:99515 Syk GO:0004713 PMID:26195794|gomodel:568b0f9600000294 IDA F gene taxon:10090 20160119 GO_Noctua occurs_in(CL:0011108),positively_regulates(GO:0007249)

MGI MGI:99515 Syk GO:1903721 PMID:26195794|gomodel:568b0f9600000294 IGI MGI:MGI:97312|MGI:MGI:99515 F gene taxon:10090 20160119 GO_Noctua occurs_in(CL:0011108)

In the second case, we want to keep that original annotation because it is not to the root node and we want to add the annotation to the regulation term. -D

ukemi commented 8 years ago

Just realized an error in the above because I copied straight from Noctua: MGI MGI:99515 Syk GO:1903721 PMID:26195794|gomodel:568b0f9600000294 IGI MGI:97312|MGI:99515 F gene taxon:10090 20160119 GO_Noctua occurs_in(CL:0011108)

vanaukenk commented 8 years ago

@ukemi If we go the route of adding relations to the relations/qualifiers column of the GAF, would we still need the folded annotations?

ukemi commented 8 years ago

We would still need them for traditional enrichment.

ukemi commented 8 years ago

Actually I realize now that I read this wrong, I was thinking of the annotation extensions. But I still think that for enrichment, we want to have the annotations to the folded terms available for users.