nih-cfde / helpdesk-collaboration

1 stars 0 forks source link

HELP-426 Problem using 'OBI:0001985' in assay type and analysis type #5

Closed jeremywalter closed 1 year ago

jeremywalter commented 1 year ago

During the last submission another problem surfaced when running the submission prep and frictionless tool:

Term 'OBI:0001985' has no name: if the term ID isn't a typo, please check integrity of 'assay_type' reference file(s).

Term 'OBI:0001985' has no name: if the term ID isn't a typo, please check integrity of 'analysis_type' reference file(s).

We submitted new data files based on glycan microarray assays. OBI does not have this term but a parent term (microarray assay: OBI:0001985). When adding this term to the “assay_type” column of files.tsv the first error above showed up (during the prep and submission). I also tried to use it for “analysis_type” but a similar error appeared. Overall, I am not sure about the difference between assay and analysis in this context anyway.

Can you please:

  1. Clarify the difference between assay and analysis in this context
  2. Let us know why 'OBI:0001985' was not accepted during the submission.
jeremywalter commented 1 year ago

September 29, 2022 at 4:58 PM

Hi,

I regenerated the files:

https://www.dropbox.com/s/1vjwga3keveav4v/2022.09.29.zip?dl=0

Best, René


Rayna Harris September 29, 2022 at 4:38 PM Hi Rene,

Thanks for the info and your patience while we look into this.

Can you send a copy of the DCC-prepared submission table files (including but not limited to file.tsv) that generated the error so that we can debug the issue?

Thanks,

Rayna

jonathancrabtree commented 1 year ago

Hi @ReneRanzinger, I've downloaded the tsv files you provided above and am looking into this. Thanks!

ReneRanzinger commented 1 year ago

This is related to https://github.com/glygener/glygen.cfde.generator/issues/22.

jeet-vora commented 1 year ago

This issue seems to have gone with the new update of files on osf and the prep script. The issue can be closed.

RLC-DCPPC commented 1 year ago

Based on Jeet's comment we are closing this issue.