Log odds ratio in Mutual Exclusivity analysis

[tvil...@gmail.com]

We installed a local cBioportal instance (v. 1.14.1) for our data and it seems that the odds ratio calculated by cBioportal is calculated differently from the equation described in Gao et al 2013: OR = (# altered in both genes * # altered in neither gene) / (# altered in gene 1 * # altered in gene 2). How is the odds ratio calculated in the current version of cBioportal?

[JJ Gao]

Hi,

We are using the same equation when calcuating OR. Please note that we are report ln(OR) on the web site.

Would you please give us an example that the odds ration is different from expected?

Thanks,

-JJ

On Thu, Oct 4, 2018 at 4:47 PM tomas....@nih.gov <tvil...@gmail.com> wrote:

We installed a local cBioportal instance (v. 1.14.1) for our data and it seems that the odds ratio calculated by cBioportal is calculated differently from the equation described in Gao et al 2013: OR = (# altered in both genes * # altered in neither gene) / (# altered in gene 1 * # altered in gene 2). How is the odds ratio calculated in the current version of cBioportal?

--
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 post to this group, send email to cbiop...@googlegroups.com.
Visit this group at https://groups.google.com/group/cbioportal.
To view this discussion on the web visit https://groups.google.com/d/msgid/cbioportal/eb0776b3-2755-4edc-8ed6-ccce5ded6e07%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.