ssGSEA normalization method selection

[Maite E]

Hi,

I am performing a ssGSEA in GenePattern. My input data is TPMs. I am unsure about what normalization method to select among "none", "rank", "log", "log.rank". I read the documentation, and it says the following:

• Sample normalization method*
  • Normalization method applied to expression data (if the option “none” is selected, data will not be normalized). Supported methods are none, rank, log.rank, and log. (Default: none)

It is not clear to me whether I should select the normalization method I already applied, or the one GenePattern will apply when calculating my ssGSEA values. If the second option, what normalization method do you recommend to apply and why?

Thank you!

Maite

[Anthony Castanza]

Hi Maite,

My apologies for the delay in getting back to you here. This parameter represents an additional normalization step that would be performed if it was selected. If you are using TPM data there is generally not a reason to select an additional normalization here and it can be left on "none".

-Anthony

Anthony S. Castanza, PhD
Curator, Molecular Signatures Database
Mesirov Lab, Department of Medicine
University of California, San Diego