Clarification Regarding the "Mutation Count" Variable in the Breast Invasive Carcinoma (TCGA, PanCancer Atlas) Study

[ROSANA XAVIER SOUZA]

Dear cBioPortal Team,

I am currently working with data from the Breast Invasive Carcinoma (TCGA, PanCancer Atlas, 2018) study on your platform, and I would like to better understand how the "Mutation Count" variable, listed in the Clinical Data section, is generated.

I have read the available documentation, but I could not find a detailed explanation of the steps or criteria used to calculate this value. Could you please clarify:

What exactly is counted in the "Mutation Count" field, and how is this number derived from the sequencing data?

Understanding this process is important for my analysis, and I would greatly appreciate your help.

Thank you in advance for your attention and support.

Best regards,
Rosana Xavier

----

É obrigatória a utilização do e-mail @unifesp em todas as correspondências oficiais, institucionais e no acesso aos equipamentos e sistemas da Universidade Federal de São Paulo, conforme a portaria Reitoria n. 1182/2022 que define a Política de e-mail institucional da Unifesp.

[Prasanna Jagannathan]

Hi Rosana Xavier,

Mutation count is the number of mutations in the sample. 

Tumor Mutation Burden (TMB) is another way of quantifying the number of mutations. Tumor Mutation Burden takes the number of mutations and adjusts it by the amount of the genome that was profiled. 

One way to think about the difference is to consider two samples, one profiled with whole exome sequencing and the other profiled by a targeted gene panel. Both samples might have the same number of mutations, but the sample sequenced by whole exome will have a much lower TMB, as those mutations were identified over a larger portion of the genome compared to a targeted panel.

In our faq page for TCGA studies: https://docs.cbioportal.org/user-guide/faq/#tcga,  details of three different sources for TCGA data are given, which explains some of the finer differences about the mutation count in TCGA studies.

thanks

Jag

--
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/CANCh1QU2V-vH%3DG0GX9Ob0PZ1KmYOXVkNjx_x9Ffc-CdnT4eFcQ%40mail.gmail.com.

[ROSANA XAVIER SOUZA]

Dear Jag,

Thank you very much for your prompt and helpful response.

Your explanation regarding the difference between mutation count and tumor mutation burden (TMB) was very clear, and the link to the FAQ page provided useful context on the data sources.

As a follow-up question, I would like to know if it makes sense, from your perspective, to stratify patients based on their "Mutation Count" values for comparative analyses. Would this be a valid approach when aiming to explore associations with clinical or molecular characteristics?

Thank you again for your support.

Best regards,
Rosana Xavier