Majlinda Lako - MRC scCornea.

[ESapenaVentura]

Continued from https://github.com/HumanCellAtlas/hca-data-wrangling/issues/232

Primary Wrangler: Ami Secondary Wrangler: Enrique

Associated Markdown Found here

Google sheet: https://docs.google.com/spreadsheets/d/11CLPU-rZrK7cH5YgepKJasm5UMNO32y8Zx2UuHehJc8/edit#gid=253173527

Latest Google Sheet Version: https://docs.google.com/spreadsheets/d/1Gi-HT8_D-TYIHN4C4kRUW7_HtMkcMRHLNBrsA4LQrGM/edit#gid=65395503

Publication: https://www.sciencedirect.com/science/article/pii/S1542012421000215?via%3Dihub

Key Events

• [x] Contact them about their project for the MRC grant.
• [x] Have a call/meeting with the team to assess project whereabouts.
• [x] Send initial instructions about contributing, including questionnaire and T&Cs
• [x] Receive project questionnaire and move project from potential to in_progress projects
• [x] Assign the Primary Wrangler
• [x] Receive T&C form filled out by at least 1 representative of group/project
• [x] Send custom metadata template spreadsheet, how-to guide and directions for submitting data files
• [ ] Receive spreadsheet (to a sufficient standard to submit dataset)
• [ ] Receive data (to a sufficient standard to submit dataset)
• [ ] Curate metadata spreadsheet with ontologies
• [ ] Upload sheet to validate metadata (most likely in staging)
• [ ] Check linking
• [ ] Validate data files
• [ ] Ask the Secondary Wrangler for an end-to-end review of the project. Ask the Expertise Wrangler to review specific tabs if needed
• [ ] Get final approval of submission including metadata spreadsheet and data files
• [ ] Submit dataset to Production and hand off tracking to data dashboard

Please track the below as well as the key events:

1. Track date first spreadsheet received and final spreadsheet sent by editing ticket to include date next to event.
2. Track spreadsheet iterations by placing asterisks next to receive spreadsheet event.
3. Track any metadata issues/tickets made for dataset with a bulleted list of links under received spreadsheet event. Links should be to the ticket in the metadata repo.

24/06/2019: Sent e-mail to plan catch-up call 25/06/2019: Set up call for July 4th, 2019

24/06/2020: Sent spreadsheet and instructions for filling the spreadsheet and start data upload

[ESapenaVentura]

Received an email on July 22, due to time constraints they are going to submit to SRA but will come back to us before the end of the year to submit to DCP.

[ESapenaVentura]

BioRxiv manuscript here: https://www.biorxiv.org/content/10.1101/2020.07.09.195438v1

[ESapenaVentura]

There is not enough metadata to begin wrangling this dataset. Once the manuscript is ready or they come back to us, I will begin working on this. For now I'll put it in the icebox

[rays22]

There is a recently published HCA paper that looks related to the pre-print above: Collin J, Queen R, Zerti D, et al. A single cell atlas of human cornea that defines its development, limbal progenitor cells and their interactions with the immune cells. The Ocular Surface. 2021 Apr. DOI: 10.1016/j.jtos.2021.03.010 PMID: 33865984

[rays22]

Data availability: GSE155683

[ESapenaVentura]

This dataset contains ATAC-Seq, which we can't push to the DCP until we get the Read3 enum value

[ami-day]

Wrangling this dataset.

[ESapenaVentura]

I have reviewed the spreadsheet! here are my comments:

Project

• Please add the cell count estimate (21343)
• More than a review, remember to add the UCSC cell browser link (http://retinalstemcellresearch.co.uk/CorneaCellAtlas/) into the external portals spreadsheets so that it can be properly linked in the data portal

Project - Contributors

• Linda lako’s email is wrong. Should be: majlinda.lako@newcastle.ac.uk

Donor organism

• Missing HDBR accessions for the developmental donors (It’s the ID that you used)
• I’d change the “fetal stage” for more specific ontologies (e.g. “10th week post fertilisation human stage”)
• All non-diseased donors should have “normal” in the disease status

Specimen from organism

• the amniotic membrane specimen does not look like a specimen used in the experiment for me. It looks like they obtained the amniotic membrane from other sources, and used it as a growth medium for the cell lines. I’d take it out of specimens
• In the specimen with limbal dysplasia, I am not entirely sure about the term you have chosen. Doing some searching I have found “limbal stem cell deficiency”, which seems to be more accurate, although I have no idea if this is the correct term either. There’s a bit more info here https://eyewiki.aao.org/Limbal_Stem_Cell_Deficiency

Cell line

• Continuing with the amniotic sample, I’d just remove the linked specimen

Sequence file

• It’s missing all the sequence file filenames and some metadata flagged in yellow
• Missing process ID for grouping

Analysis protocol

• Should have “Alignment”, “cell filtering”, “gene filtering”, “doublet removal" in the single cell RNA seq derivation process. I would also remove from the description the second part (referring to the creation of the uMap/tSNE)

Analysis files

• I’d add the files that you can download from the UCSC cell browser (http://retinalstemcellresearch.co.uk/CorneaCellAtlas/), at least the ones related to gene expression and cell type annotation

Please address these comments and let me know when the spreadsheet is ready for another review :)

[ami-day]

To do: add new ontology term for "dysplasia". Keep "cornea neoplasm" for now and make update when the ontology term is available.

[ami-day]

Requested the fastq files from NCBI cloud delivery service. Only 1 fastq file is available via ENA. Hopefully NCBI will provide the paired-end reads.

[ami-day]

Thanks @ESapenaVentura , I have made all your suggested changes except:

• The limbal dysplasia ontology term as we discussed, and I have commented on what we agreed for now above.

• Sequence file names and scATAC-seq reads metadata: I will fill these once I receive the fastq from NCBI cloud delivery

• I added the cell annotations files from the linked portal (http://retinalstemcellresearch.co.uk/CorneaCellAtlas/). However the gene expression files were large so I didn't download and add those. The counts files are already available, downloaded and names added to the google sheet via GEO.

Here is the updated version if you could re-review to double-check all ok, thanks! https://docs.google.com/spreadsheets/d/1Gi-HT8_D-TYIHN4C4kRUW7_HtMkcMRHLNBrsA4LQrGM/edit#gid=65395503

[ami-day]

Next ontology term request: https://github.com/HumanCellAtlas/ontology/issues/100

[ami-day]

upload area: 75b63f7f-1eb8-473c-b751-8b9376563c22

[ami-day]

Submitted.

[ami-day]

Assigned E-HCAD id: E-HCAD-46 Pre-converted the files which are stored here: https://drive.google.com/drive/folders/1MfRn7erOuzDpiASZfSYUVZvgw3uqWyfd

[ESapenaVentura]

Same comment as in here https://github.com/ebi-ait/hca-ebi-wrangler-central/issues/3#issuecomment-1058135936

[ami-day]

Re-assigning this dataset to E-HCAD-50.

[ami-day]

Manually curated the idf and sdrf files and uploaded them to Gitlab: https://gitlab.ebi.ac.uk/ebi-gene-expression/scxa-metadata/-/merge_requests/294

E-HCAD-50

[ami-day]

This dataset needs updating. The library prep. method should be 10X 3' v3 not Drop-seq. I edited it in ingest and now it is raising an error because no spatial barcode info is added. However this is not a 10X visium dataset: https://contribute.data.humancellatlas.org/submissions/detail?uuid=50d64edf-6fbe-48ea-89aa-80f3543dd077&project=6ac8e777-f9a0-4288-b5b0-446e8eba3078

[ami-day]

Submitted update.

[ami-day]

Submitted updates.

[ami-day]

exported.

[ami-day]

looks as expected in the hca data portal.