Single-cell RNA sequencing identifies celltype-specific cis-eQTLs and co-expression QTLs

[ami-day]

Short name SingleCelleQTLCoexpressionAnalysis

Primary wrangler Ami

Secondary wrangler Ida

Paper: Single-cell RNA sequencing identifies celltype-specific cis-eQTLs and co-expression QTLs https://www.nature.com/articles/s41588-018-0089-9#Sec2

Accession: EGAS00001002560

Google Sheet: https://docs.google.com/spreadsheets/d/1rxkVPdmzIQolyiQA4t08FV-P6o9i37tFpyNu5eDIXIE/edit#gid=1549964870

Ingest: https://contribute.data.humancellatlas.org/projects/detail?uuid=974df02f-fc85-4773-bbc5-a772baad95ef

current link: https://contribute.data.humancellatlas.org/projects/detail?uuid=40272c3b-4697-4bd4-ba3f-82fa96b9bf71

[ami-day]

I'm putting this dataset in the stalled column because I need to contact the corresponding authors with some questions about the donor metadata and data access

[ami-day]

I have emailed them and will wait to hear back. The raw data is managed access in EGA. I asked if we can get access to gene expression matrices.

[ami-day]

Their data should be here: https://eqtlgen.org/sc/datasets/vanderwijst2018.html Though the link is not working atm, their website is under maintenance.

[ami-day]

The authors gave me a new working link.

[idazucchi]

Hello! I've finished reviewing. Overall it looks good, I have some comments:

Donor

• there is no mention of donors with disease in the paper so I would fill out the known disease field for all donors as normal

Collection protocol

• publication doi: the doi points to a publication describing the cohort, maybe you can move it to the supplementary materials?
• you could add the Vacutainer in the retail kit field

Specimen

• is it worth it to put blood for organ part as well? I'm not sure if we want this field filled for blood/ if the browser team expects it to be filled
• like the donors, there is no known disease so I would mark all specimens as normal

Cell suspension

• the cell viability was assessed via trypan blue staining so you could fill in the cell viability method
• you can fill in selected cell type with peripheral blood mononuclear cell
• the authors mention that they pooled together samples from different donor with a cell count of 1,750 cells/donor. Since there's no sequence files it makes sense to model the cell suspension of each donor separately like you did. You could enter 1750 as estimated cell count for each donor

Library preparation protocol

• you could fill retail kit with Single Cell 3′ Library & Gel Bead kit version 2 (120237) and i7 Multiplex kit (120262)

Analysis protocol

• for analysis_protocol_gene_expression:
  • the computational method field is filled out but I think that information should be part of derivation process
• for analysis_protocol_cell_types
  • you could fill in the derivation process field

Analysis file

• for cell_barcodes.zip I would change the content description to Molecular property identifier because the content of the file is similar to the CellRanger barcode file

[ami-day]

Thanks @idazucchi , I have made most of these changes.

[ESapenaVentura]

Data wrangler is set up as contributor (Missing the data curator ontology), and it appears as if it's the first author. Needs update

[ami-day]

Updated with my role as data curator: https://contribute.data.humancellatlas.org/projects/detail?uuid=40272c3b-4697-4bd4-ba3f-82fa96b9bf71&tab=upload the status is still in "exporting" state though.

[ami-day]

Submitted the update.

[ami-day]

exported.

[ami-day]

Looks as expected in data portal.