Group Filtering Issues

[Megan Harper]

Hello,

I am trying to work with a very large dataset focusing on pancreatic
ductal adenocarcinoma (PDAC) by combining 33 studies. Since I am
combining so many studies and many include many types of pancreatic
adenocarcinoma, I am attempting to filter out any cancer types or
histologies that are definitely NOT PDAC. I am trying to include all
samples labeled as "pancreatic adenocarcinoma" even if they do not
have additional corresponding histology. However, when I try to
include samples like this in the histology description, ones that are
grouped as "NA" are still being filtered out instead of being included
in the cohort, which is over 10,000 samples. Can this be resolved so
that I can include these samples in my analyses?

Please see the following data set:
https://www.cbioportal.org/study/summary?id=6914c8967783fa02c7f35550%2Cmsk_chord_2024%2Cmsk_impact_2017%2Cmetastatic_solid_tumors_mich_2017%2Cmixed_allen_2018%2Csummit_2018%2Ctmb_mskcc_2018%2Cntrk_msk_2019%2Cmsk_ch_2020%2Cpan_origimed_2020%2Cpancan_pcawg_2020%2Cmsk_met_2021%2Cmsk_ch_2023%2Cmsk_ch_ped_2021%2Cbraf_msk_archer_2024%2Cbraf_msk_impact_2024%2Cmakeanimpact_ccr_2023%2Cmsk_ctdna_vte_2024%2Cmsk_access_2021%2Cpancan_mimsi_msk_2024%2Cmixed_selpercatinib_2020%2Cmixed_cfdna_msk_2020%2Ccfdna_msk_2019%2Cpancan_pdmr_2025%2Cpog570_bcgsc_2020%2Cmixed_impact_subset_2022%2Cmixed_msk_tcga_2021

Thanks,
Meg Harper

[Aaron L]

Thanks Megan, this does look like a bug.  Will look into it.

--
You received this message because you are subscribed to the Google Groups "cBioPortal for Cancer Genomics Discussion Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cbioportal+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cbioportal/CADQrjQ%2BLcHKxtPuSspiGaPakYt3f_Z8G0aChXXa5tgjRGmvq-A%40mail.gmail.com.

[Aaron L]

Ok, I think we have an idea about the problem, though it may be difficult to fix unfortunately.  There is probably a way to obtain the list of samples that you want and create a virtual study.  This would be a workaround.   If your need is urgent, I'm happy to try to help you with it.

--Aaron