Question: PAN - TCGA survival analysis
gravatar for rt.cgen
3.6 years ago by
rt.cgen40 wrote:

I am trying to run survival analysis (KM plot) for a set of genes on PAN-TCGA dataset, I am aware of tools like RTCGAToolBox that can help in downloading and running these analysis for a cancer type at a time. I have also tried cbioportals' web-API and R package ("cgdsR"). I finally resorted to using R package "survival" but that isn't helpful either. There are easy ways to run survival analysis for one cancer type but so far I haven't found anything that computes survival analysis for many cancer type for a set of genes.

TL;DR - Survival analysis for cumulative TCGA dataset (spanning a few TCGA dataset) for a set of genes.

Any help would be greatly appreciated. TIA

I am reposting this question since I did not get any response earlier.

genomics next-gen tcga • 1.9k views
ADD COMMENTlink modified 3.6 years ago by Jordan Anaya1.1k • written 3.6 years ago by rt.cgen40
gravatar for Jordan Anaya
3.6 years ago by
Jordan Anaya1.1k
Jordan Anaya1.1k wrote:

I'm not quite sure what type of plot you want to make. If you want to look at several genes in a single cancer I can understand that. For example, you can compare patients with high expression of both genes X and Y vs. low expression of both X and Y. To perform a pan-cancer Kaplan plot are you just planning on merging all of the patients together?

If that is what you are looking to do you could download the data from OncoLnc: for each cancer and gene and then make your plot with whatever software you want. One concern that I would have is the RSEM values are normalized by the TCGA per cancer so I don't know how appropriate it is to compare the RSEM values in one cancer to the RSEM values in another cancer.

ADD COMMENTlink written 3.6 years ago by Jordan Anaya1.1k

I am trying to make KM plots for a set of genes in multiple cancer types at the same time. I am planning on merging clinical data set for PAN TCGA to make pan-cancer KM plots but I am not sure if that's correct approach, because I would need to merge mutation data for all cancers as well.

ADD REPLYlink written 3.6 years ago by rt.cgen40
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1103 users visited in the last hour