Survival analysis based on gene signature

Question

[Sy80]

Hi all

Does anyone have any advice on how to create a KM plot (in R) based on the expression of a set of genes (hi vs low)?

I am able to test the genes individually (and generate survival curves for each gene) but would like to assess whether the signature has a prognostic effect in the TCGA datasets.

Any help is much appreciated - thank you!

[Hamid Ghaedi]

In such cases, I would suggest first doing a single-sample gene set enrichment analysis (like GSVA, ssGSEA) and then doing a survival analysis based on scores from interested pathways.

[Sy80]

Great idea. I will give this a try. Thank you!

[jv]

The Cox regression is an option for multivariate survival analysis

[Sy80]

I will try Cox regression with single sample enrichment scores as a continuous variable as per suggestions. Thank you!