Public Release Studies: February 2024

[Rima-Waleed]

Fix #2005

Cancer studies updated in this pull request:

Study_Id	Testing Instance Link	Sample Count
difg_msk_2023	https://triage.cbioportal.mskcc.org/study/summary?id=difg_msk_2023	73 Samples
cll_broad_2022	https://triage.cbioportal.mskcc.org/study/summary?id=cll_broad_2022	1154 Samples
brain_ccma_2023	https://triage.cbioportal.mskcc.org/study/summary?id=brain_ccma_2023	182 Samples
msk_spectrum_tme_2022	https://triage.cbioportal.mskcc.org/study/summary?id=msk_spectrum_tme_2022	82 Samples
sarcoma_msk_2022	https://triage.cbioportal.mskcc.org/study/summary?id=sarcoma_msk_2022	7494 Samples
mbn_sfu_2023	https://triage.cbioportal.mskcc.org/study/summary?id=mbn_sfu_2023	297 Samples
	Total	9282

NOTE:

sarcoma_msk_2022

• CNA pending from author

brain_ccma_2023

• RNA-seq & CNA pending from author

cll_broad_2022

• mutsig missing data pending from author.

difg_msk_2023 & msk_spectrum_tme_2022

• Study has been updated and merged to master branch.

*Missing data will be added as soon as authors respond.

[rmadupuri]

Thanks for the updates Rima! A few more issues that I noticed.

brain_ccma

• [x] The sequenced case list is not reflecting the WGS count(149) from paper.
• [x] There are 198k rows with just probe names and empty values in the file. Is the Methylation data formatted correctly?
• [x] RNA-seq data is available on the CCMA portal and can be converted to the matrix format. Pending from authod by EOW.
• [x] There's also drug screening data. We can add this in Generic Assay format. Pending from authod by EOW.
• [x] The paper states only 2 normal tissue primary cell lines. There's 38 in the portal. Can we double check? The paper mentions 2 normal tissue primary cells in the atlas, however there are 38 non-malignant cell lines .
• [x] Some mutations are missing in the portal? A few samples show up as unprofiled in the portal although the paper shows the variants. See the oncoprint in Figure 1B of the paper. Ex. the Osteosarcoma type. Also the mutation type in the paper and portal are a bit different. Reached out to authors as I found more discrepancies in mutation data.

We will have to hold off the CCMA study for a closer review.

cll_broad

• [x] We can remove the WGS samples (91) from the cohort since the genomic data is not available.
• [x] Although WGS data is missing, the other gene percentages are close to the Fig.1 but NOTCH1 is very low. 5% in portal whereas ~12% in paper. Can we double check?
• [x] We can add the focal events as generic assay. Data is in Supp tables. See the Oncoprint in Fig.1
• [x] Gene panel can be added based on the Sequencing Type attr.

sarcoma_msk_2022

• [x] Onoctree codes need to be updated based on the 'final diagnosis' from the supp table. The samples with NA code have a malignant cancer type.

[ritikakundra]

sarcoma_msk_2022:

• [x] Description does not include normals No matched normal tissue
• [x] What was used to define OncoTree codes? Orignal or corrected. Whichever was used should be defined in the description of the meta fields. Final diagnosis was used, added to meta description
• [x] Any idea why the counts in figure 1a don't match our cancer counts? Some final diagnoses don't have OncoTree codes and so were grouped with similar cancer types (ex. Osteosarcoma (OS) given to Extraskeletal Osteosarcoma, Sarcoma, NOS (SARCNOS) given to Sarcoma, Unclassified, and Round Cell Sarcoma, NOS (RCSNOS) given to Undifferentiated Round Cell Sarcoma/Ewing-Like)
• [x] For tumor purity, add computational tumor purity to the description.
• [x] Might be an ask, but is it correct to have Sarcoma as an ID or sarcoma or SARCOMA_ _Changed to sarcoma__
• [x] Can we confirm that no mutations were filtered out in the paper? Not mentioned, only variants of unknown significance were excluded from analyses.