Different results from pathway table

[ROCCA ANDREA]

Dear Sirs

At the beginning of 2024, I conducted some research on a subgroup of patients from the lung adenocarcinoma TCGA pancancer dataset, querying a set of genes (EGFR ERBB2 MET KRAS BRAF ALK NTRK1 NTRK2 NTRK3 RET ROS1) and then studying the percentages of gene alterations in certain pathways (RTK-RAS, NRF2, cell cycle, PI3K). I have now repeated the same research with the same criteria, and the percentages of gene alterations are the same as those detected a year ago for all genes except KEAP1 (NRF2 pathway), whose alteration percentages have increased from 11.5% a year ago to 22% currently. I believe I conducted the research in exactly the same way as last year. Is there a possible explanation for this variation? Or am I perhaps doing something wrong?

Thank you for your attention.

Best regards

Andrea Rocca, MD

Researcher in Oncology

Department of Medical, Surgical and Health Sciences

University of Trieste

34147 Trieste, Italy

[Ramya Madupuri]

Hi Andrea,

Thanks for reaching out! To help figure out why the KEAP1 alteration percentage looks different, could you let us know the exact query or filters you used in cBioPortal to get these numbers? This will help us check if the dataset has changed or if the query parameters may be affecting the results.

Thanks,

Ramya

--
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/AM8P250MB03420CC0D8FA377241EA8376F325A%40AM8P250MB0342.EURP250.PROD.OUTLOOK.COM.

[ROCCA ANDREA]

Dear Ramya,

I will try to show you the subgroup of patients I selected. This is the link to my virtual study, it should include 261 patients from the lung adenocarcinoma TCGA pancancer dataset:

https://www.cbioportal.org/study?id=688a4e0867c2b04a88c49b26

I queried these genes, altogether: EGFR ERBB2 MET KRAS BRAF ALK NTRK1 NTRK2 NTRK3 RET ROS1 (but selecting just one does not appear to affect the results).

From the Results View I then selected the Pathways tab and looked at four pathways: RTK-RAS, NRF2, cell cycle, and PI3K. 

The discordant value is in the NRF2 pathways, KEAP1 gene.

I did not use other filters such as those on driver or somatic mutations.

Thank you for your assistance.

Best regards

Andrea

---

Da: Ramya Madupuri <ramyama...@gmail.com>
Inviato: mercoledì 30 luglio 2025 16:30
A: ROCCA ANDREA <ANDREA...@units.it>
Cc: cbiop...@googlegroups.com <cbiop...@googlegroups.com>
Oggetto: Re: [cbioportal] Different results from pathway table

 

[Ramya Madupuri]

Hi Andrea,

Thanks for sharing the query details. We have looked into it from all angles but could not reproduce the 11.5%. There have not been any data changes that would explain a drop in the percentage. It’s possible that the difference may be due to some variation in the subset of samples or query parameters. 

Please feel free to reach out if you have any further questions.

Thanks,

Ramya

[ROCCA ANDREA]

Dear Ramya

thank you very much for your assistance. I will obviously keep the last result. Perhaps, as you said, there was some difference in the subset, although strangely all other figures were exactly the same. 

Just for documentation I send you attached the old results on NRF2 pathway.

Best

Andrea 

---

Da: Ramya Madupuri <ramyama...@gmail.com>
Inviato: giovedì 31 luglio 2025 16:46