GSE152014 (organoids): manual organoids and printed organoids

ami-day commented 3 years ago

As of 28/02/2022: needs update

Ingest: KidneyOrganoidsBioprinting

Needs update:

process IDs in the sequence file tab should be runs rather than experiment accessions.
Project cell count estimate

=====================================================================

Dataset/group this task is for:

Barbara Treutlin, Melissa Little

Wrangler responsible for this dataset/lab:

Ami

Google Sheet: sheet

submission: https://contribute.data.humancellatlas.org/submissions/detail?uuid=2552985f-707a-46dd-bf8a-3c88721d8109 old submission link

Publication:

Acceptance criteria for the task:

[x] curate metadata
[x] fill ontologies
[x] graph validator
[x] secondary review
[x] validate in ingest staging
[x] upload to ingest production

ami-day commented 3 years ago

Uploading fastq to hca-util upload area: 7518f6a7-1310-451d-9e19-a4dda2d766fa

rays22 commented 3 years ago

Secondary review notes

The metadata spreadsheet contains Project tabs. I thought the standard practice was to create and curate project level metadata using Ingest UI so that the project shows up in the project catalogue.

Dissociation protocol: it looks unfinished

missing ontology terms: dissociation_protocol.method.ontology
dissociation_protocol.protocol_core.protocol_description:

Organoids were dissociated after 12 days of growth. Emailed authors to ask about dissociation method.

Organoid: the biomaterial entity links look incorrect

A link from a biomaterial with id CRL2429-MAFBmTagBFP2/GATA3mCherry_iPSC_donor_sample to a protocol with id, "collection_protocol_ipsc", is not found in the spreadsheet.
A link from a biomaterial with id manual_organoids_replicate1 to a biomaterial with id, "CRL2429-MAFBmTagBFP2/GATA3mCherry", is not found in the spreadsheet.
A link from a biomaterial with id bioprinted_ratio0_organoids_replicate1 to a biomaterial with id, "CRL2429-MAFBmTagBFP2/GATA3mCherry", is not found in the spreadsheet.

Other issues:

organoid.age : I do not think an age range is allowed here.
Biomaterials: CRL2429-MAFBmTagBFP2/GATA3mCherry_iPSC_donor: is missing the required development_stage.
collection_protocol_iPSC is missing the collection_protocol.method.text
organoid_dissociation is missing the dissociation_protocol.method.text
gex_library_preparation: It looks like some fields are filled erroneously (copy-paste into wrong cells?). For example, library_preparation_protocol.nucleic_acid_conversion_kit.expiry_date : 16 .

Collection protocol: is it necessary?

There is no useful information here on how specimen was collected. It does not make sense for the iPSC cell line.

ami-day commented 3 years ago

Thanks @rays22 I have made the relevant updates and am uploading it to ingest prod. now for export.

Wkt8 commented 2 years ago

See issues here: https://embl-ebi-ait.slack.com/archives/C016FKV4JS1/p1645626628976429

Short version: This project has already been exported. Experiment accessions were added to the process id column as opposed to run accessions. Therefore there are multiple files (too many) linked to each process id.

The fix is to edit and change to the correct process ids - which will be a change to linking and create additional processes - and ensure that it can be properly exported. There are about ~60 fastq files.

ami-day commented 2 years ago

Exported project: https://contribute.data.humancellatlas.org/projects/detail?uuid=8787c238-89ef-4636-a57d-3167e8b54a80

Hca-util upload area: 7518f6a7-1310-451d-9e19-a4dda2d766fa/

ESapenaVentura commented 2 years ago

It says that it has been exported but I can only see the submission in graph-invalid status

ami-day commented 2 years ago

updating link between entities via spreadsheet download, edit and import is not yet supported in ingest.

ofanobilbao commented 1 year ago

Linked to https://app.zenhub.com/workspaces/operations-5fa2d8f2df78bb000f7fb2b5/issues/ebi-ait/hca-ebi-wrangler-central/775

ofanobilbao commented 1 year ago

@ESapenaVentura and @idazucchi to reconstruct dataset and re-export

ESapenaVentura commented 1 year ago

I am going to reconstruct the dataset - For that, first I will delete the submission entirely. There is a spreadsheet with UUIDs in the other ticket. The submission should be, worst case scenario happen, reconstructible from the db backups

ESapenaVentura commented 1 year ago

Can't delete the submission - will flag it tomorrow

idazucchi commented 1 year ago

Deleted the submission!

ebi-ait / hca-ebi-wrangler-central