map grin taxa to ncbi taxa

[jhpoelen]

As discussed in #11.

[jhpoelen]

@AustinMeyer @cmungall just created a first pass at using NCBI LinkOut to link GRIN taxa to their NCBI counter parts. I ran a 16k version of samara grin scrape (see https://github.com/jhpoelen/samara/releases/tag/v0.1.10). Any chance you can kick-off a full run using the jenkins job?

[cmungall]

I kicked it off https://build.berkeleybop.org/job/extract-grin-traits/20/

but out jenkins may be down over TG

@elserj - any plans for a planteome jenkins instance?

[jhpoelen]

While #31 is still pending a successful run, preliminary results suggest that ncbi taxa are now linked to grin taxa in the scrape using recently implemented method using ncbi linkout. Interesting to see that while ARS-Grin has provided linkout data to ncbi taxonomy, they do not link back to ncbi taxonomy from their own webpages.

@austinmeier Closing issue, please feel free to comment/re-open when functionality is not as desired.

First couple of lines from https://build.berkeleybop.org/view/Planteome/job/extract-grin-traits/20/ artifact show that (verbatim_taxon_id) GRINTaxon:300359 is linked to (resolved_taxon_id) NCBITaxon:3879 :

verbatim_taxon_id	verbatim_taxon_name	resolved_taxon_id	descriptor_id	descriptor_name	descriptor_definition	method_id	method_name	observed_value	accession_id	accession_number	accession_name	collected_from	citations
GRINTaxon:300359	Medicago sativa L. subsp. sativa	NCBITaxon:3879	GRINDesc:68104	By pass protein	In-vitro dry matter disappearance (ivdmd) expressed as a percent of the cultivar venal.  Higher than 100% suggests low digestibility & higher by-pass protein.	GRINMethod:391002	ALFALFA.PROTBYPASS.93.VOLENEC	140	GRINAccess:1140225	PI 162787	'PAMPA'	Argentina	D.Z. Skinner. 1999. Non random chloroplast DNA hypervariability in Medicago sativa. Theor Appl Genet Theoretical and applied genetics; international journal of b.|D.H. Basigalup, D.K. Barnes, and R.E. S
[...]

[austinmeier]

This looks great. I will give it a test run, and see what it looks like. I'll report back.