GSE123904 - RegenerativeLineagesLungMetastasis

[idazucchi]

Project short name:

RegenerativeLineagesLungMetastasis

Primary Wrangler:

Ida

Secondary Wrangler:

Arsenios

Associated files

• Google Drive: folder

Published study links

• Paper: Regenerative lineages and immune-mediated pruning in lung cancer metastasis

• Accessioned data: superseries GSE123904

  • GSE123903 mouse host, humans scRNA-seq
  • GSE123902 human scRNA-seq

• Ingest

Key Events

• [x] Convert published metadata to HCA spreadsheet
• [x] Manually curate dataset to meet HCA metadata standard
• [x] Collect any matrix and cell-type annotation files
• [x] Are the analysis files suitable for CellxGene? If something is missing get in touch with the authors to request it
• [x] Upload sheet to validate metadata
• [x] Transfer raw files to ingest to validate data files
• [x] Check linking using ingest graph validator
• [x] Ask the Secondary Wrangler for an end-to-end review of the project. Ask the Expertise Wrangler to review specific tabs if needed
• [x] Submit dataset to Production
• [x] Complete the Export SOP
• [ ] Convert project data to SCEA format following the SCEA conversion SOP if appropriate

[idazucchi]

Human donor metadata All metadata is in a picture, I converted it to xlxs fromat --> here

Mouse host data Mice are employed as xenograft metastasis model. Mice are injected with H2087-LCCs (lung cell cancer) then:

1. metastasis are collected from 6 mice, cultured with antibiotic to select human cells over host cells

   We isolated lung cancer cells from two BLI-negative kidneys that showed no signs of metastasis, yet contained DTCs. We also isolated cancer cells from one case of incipient lung metastasis and from three individual lung macrometastases that evolved spontaneously. Cells were uniformly isolated through antibiotic selection for one passage in culture and then subjected to scRNA-seq. This selection step, which was essential because DTCs are extremely rare, likely has some effect on gene expression and is a limitation of our model.

2. 6 mice are treated to depleate NK cells, to increase metastasis potential. Metastasis are collected

   H2087-LCCs were intracardially injected into the arterial circulation of athymic mice. Thirty days later, we administrated anti-GM1 antibody to trigger NK cell depletion which facilitates the robust outbreak of macrometastases as evidenced by BLI signal.

[idazucchi]

when reviewing please pay attention to the ontology terms and the inDrop library preparation (it would be good to review the entry in the cheat sheet) for the xenograft modelling I relied on the best practice document

[arschat]

Hello, Ida! Great job! Here are my comments.

Secondary review

Project

• [x] Lung atlas is now available in hca_bionetwork.hca_tissue_atlas module in the latest update here.
• [x] Emails can be added for the two corresponding authors. peerd@mskcc.org & j-massague@ski.mskcc.org

Dissociation protocol

• In the best practice document, it is mentioned: No dissociation protocol for blood and other fluids, since the cells are already separated. Here, since we don't have separate RBC suspension (would be considered an enrichment protocol), my suggestion would be to skip the red_blood_cells_lysis protocol. I understand however that this is the most suitable place for this.

Library preparation

• [x] Concerning the cell barcode length, in the paper with the modified inDrops protocol, they mention two barcodes: one with 11 bp and one with 8 bp with the probe sequence in between.

  /5Acryd/PC/CGATGACGTAATACGACTCACTATAGGGATACCACCATGGCTCTTTCCCTACACGACGCTCTTCCGATCT[12345678901]GAGTGATTGCTTGTGACGCCTT[12345678]NNNNNNNNTTTTTTTTTTTTTTTTTTTV, where 5Acryd is an acrydite moiety, PC is a photo-cleavable spacer, the letters in bold indicate T7 RNA promoter sequence, and underlined letters indicate the site for Illumina PE Read 1 Sequencing primer. The numbers indicate cell barcodes, which were specifically designed for this experiment to have 50% GC content and Hamming distance of > = 3 between each pair of barcodes. For that reason, I think the length should be 19 instead of 16.

Sequence protocol

• [x] Missing parenthesis in the sequencing_protocol.10x.fastq_method

Analysis file

• [x] missed formula at tab Analysis file, cell H37

[idazucchi]

thank you for reviewing! from my prerspective the point of red_blood_cells_lysis as a dissociation protocol is to highlight that red blood cells have been excluded from the lung cell suspension. Since the methods say "Red blood cells were lysed in red blood cell lysis solution (once or twice depending on red blood cell content)" it felt like a meaningful thing to record

Library preparation --> I'm going to put 19 as cell barcode length for now, but I'm not sure what's the best way to represent split barcodes like this one. Is it more important to get the total barcode length or the nucleotide lenght in the read, also accounting for the gap?

Thanks for spotting the formula! I've checked the submission and it looks like the formula didn't make it to ingest, which is surprising If we get a weird validation issue it might be related this!

[idazucchi]

After discussing the dataset with the wrangler I've changed red_blood_cells_lysis into an enrichment protocol and updated the cell barcode length for inDrop to 41, to include the spacer length

[idazucchi]

additional cellxgene info in this thread