Downloading annotation extensions should use the standard syntax

[ValWood]

Is the output meant to look like this?

PomBase:SPBC216.07c serine/threonine protein kinase Tor2 GO:0004672 "{""relationship"": {""relation"": [{""id"": ""GOREL:0000752"", ""label"": ""has_direct_input""}], ""id"": ""PomBase:SPAC1B9.02c"", ""label"": ""PomBase:SPAC1B9.02c""}}" PomBase NCBITaxon:4896 IDA PMID:22976295 20151027
PomBase:SPBC216.07c serine/threonine protein kinase Tor2 GO:0004672 "{""relationship"": {""relation"": [{""id"": ""GOREL:0000752"", ""label"": ""has_direct_input""}], ""id"": ""PomBase:SPAC22E12.14c"", ""label"": ""PomBase:SPAC22E12.14c""}}" PomBase NCBITaxon:4896 IDA PMID:22976295 20151027

[ValWood]

• add column headers to download file

2. I'd rather see extensions formatted in the same way that we construct them...there is a lot of duplication in theis output. And too many """"":[{]

[ValWood]

I now see this is jsn format? would people want json synax in a the default tab delimited file?

I tried a different option (annotation extesnion class label)

This gives me:

PomBase:SPBC14F5.08 mediator complex subunit Med7 GO:0030295 PomBase NCBITaxon:4896 IDA PMID:10625684 20160908 regulation of transcription from RNA polymerase II promoter
PomBase:SPAC2F3.15 P-TEFb-associated cyclin-dependent protein kinase Lsk1 GO:0004693 PomBase NCBITaxon:4896 IMP PMID:17502918 20151109

1. No extension if the extension is a gene product
2. No extension relation

[ValWood]

next option gives me some craziness:

PomBase:SPBC14F5.08 mediator complex subunit Med7 GO:0030295 PomBase NCBITaxon:4896 IDA biological regulation|biological_process|regulation of RNA biosynthetic process|regulation of nucleobase-containing compound metabolic process|regulation of macromolecule metabolic process|regulation of transcription, DNA-templated|regulation of nitrogen compound metabolic process|regulation of nucleic acid-templated transcription|regulation of biosynthetic process|regulation of transcription from RNA polymerase II promoter|regulation of cellular process|regulation of RNA metabolic process|regulation of gene expression|regulation of macromolecule biosynthetic process|regulation of metabolic process|regulation of cellular macromolecule biosynthetic process|regulation of biological process|regulation of cellular biosynthetic process|regulation of primary metabolic process|regulation of cellular metabolic process PMID:10625684 20160908

Is there an option which gives me the annotated extension?
i.e.
has_direct_input PomBase:SPBC19F8.07,part of regulation of transcription from RNA polymerase II promoter

[kltm]

I'm afraid this is also an extension of the issues around #201 .

[cmungall]

201 is about how to better query using information in the extension.

I think @ValWood just wants an extension column that is identical to the source GAF. This should be straightforward but plumbing at a few different places:

• Extend association schema to include a field annotation extension (source)
• Extend owltools-solr to populate this with the exact string, prior to parsing
• Remove json from list of defaults

Some additional side-issues

• Show column headers
• Remove unused fields (e.g. geo)
• General user-friendliness stuff

[ValWood]

Spot on! Thanks.

[kltm]

Dupe #94 then.
For the additional issues, one of the reason that we agreed to the "pseudo-gaf" is so we do not have to pipe the solr output through a proxy to add decorations.