Dear cBioPortal Team,
Last year, I identified recurrent missense mutations in five neuroblastoma datasets available on cBioPortal. Upon revisiting the data recently, I noticed that, for example, the TARGET dataset now appears to have an updated name and includes additional or altered
samples.
Some of the recurrent mutations I previously identified are no longer present in the updated versions. Could this be due to dataset updates? Is there a way to access the earlier versions of these datasets, specifically the versions prior to the update from
TARGET 2018 to TARGET GDC?
I’m trying to understand how or why some of these mutations may have been lost in the updated datasets, and I would greatly appreciate any information or access you can provide regarding previous dataset versions.
Thank you in advance for your help.
Best regards,
Katharina Schneider
Zeynep Karagöz
Product Owner & Data Engineer
--
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/791F6EEE-AB63-4827-9E5B-415A76679C43%40charite.de.
To view this discussion visit https://groups.google.com/d/msgid/cbioportal/CAJEmdvdaf5-NFZvnH9Km2si1GSJGQDVzoF5byMyZEUesfjb4MQ%40mail.gmail.com.
Dear JJ,
Yes, there are currently five neuroblastoma (NB) studies available in cBioPortal (see screenshot). At some point, one of the studies appears to have been exchanged or altered, specifically, the MSK study (Nat Genet 2023) seems to have been replaced by the TARGET study (GDC).
Now, when selecting all studies, I receive a warning regarding GRCh37/hg19 vs. GRCh38/hg38 genome builds. Many mutations are either slightly different, missing, or new ones have been added. For example, I can no longer find a single KRAS mutation, despite having identified multiple KRAS SNVs it in the data before, and such a mutation would be expected.
This makes me wonder whether the discrepancies could be due to alignment issues. Is there any way to access the versions of these datasets prior to the update?
I am also open to a meeting if this helps explain the issue I’m facing more!
Kind regards,
Katharina Schneider
From: JJ Gao <jianji...@gmail.com>
Date: Wednesday, May 21, 2025 at 14:22
To: Zeynep Karagöz <zey...@thehyve.nl>
Cc: "Schneider, Katharina" <katharina...@charite.de>, "cbiop...@googlegroups.com" <cbiop...@googlegroups.com>, "ritika...@gmail.com" <ritika...@gmail.com>
Subject: [ext] Re: [cbioportal] Neuroblastoma studies archived?
Hi Katharine,
Sorry for hte issue. Thanks for contacting us.
Would you please give us some more details, e.g. the specific studies and genes/mutations?
Thanks,
-JJ
On Tue, May 20, 2025 at 8:06 AM 'Zeynep Karagöz' via cBioPortal for Cancer Genomics Discussion Group <cbiop...@googlegroups.com> wrote:
Hi Katharina,
Thanks for contacting the group.
I copied Ritika in this email, she would have more information on this subject.
Best,
Apologies for the confusion.
The MSK neuroblastoma (NBL) study is currently on temporary hold as we work on some modifications. It will be re-released once those updates are complete.
The TARGET NBL study is a newly added cohort and was not intended to replace the MSK dataset. However, since the TARGET cohort is based on the hg38 genome build, while the other NBL studies use hg19, we display a warning to alert users. This is to discourage combining studies across different genome builds during cross-study analyses, as the results may be inaccurate due to alignment discrepancies.
Regarding KRAS mutations, I checked and found that only the MSK cohort had KRAS-mutated samples. That’s why you no longer see KRAS mutations in the current view.
To view this discussion visit https://groups.google.com/d/msgid/cbioportal/CABbxrZMxb5cQvz-drhUDe9H92bax70uURh-kkmV%2B_cj3c7KSPQ%40mail.gmail.com.