Searching for RNA-seq melanoma samples in TCGA or cBioPortal

Question

[JACKY]

I'm a junior bioinformatician, and I know TCGA and cBioPortal are among the most important databases.

I'm trying to find as much data as possible for melanoma (SKCM), preferably RNA-seq, and with some important data like LDH levels, tumour purity, and more.

I've tried searching in TCGA but there is too much info there, I got a bit lost. Can anybody guide me or could perhaps tell me where to look?

Thanks!

[Friederike]

Can you specify a bit more what the end goal is? I.e. what type of data (processed, raw, ...) would you like to download and what would you like to do with it?

[JACKY]

I want processed SKCM data (preferably counts, but normalized data would also work.. like RPKM or TPM). And for the same samples, I want the metadata which is the most important part.

I need LDH, tumour purity, ECOG.. but the most important feature I need is LDH.

What I'm doing is studying the relationship between LDH levels and tumor purity in melanoma. In case there is no tumour purity I can just calculate it myself, which is why I also need the gene expression RNA-seq matrix.

I believe there is tons of data out there in TCGA, and cBioPortal which is based on TCGA data, but mining there and finding what I need is pretty complicated, I got lost very quickly...

Answer 1

[Vitor1]

So there are two main ways of getting your data there.

1. Go directly to the webpage and search manually for the subtypes and query the files that you need (this tutorial is pretty decent in navigating there, in your case the you would need the check the Data Category - Clinical)

2. Go for the TCGABiolinks R package, I would recommend to see the vignette and some tutorials like [this one] and some scripts4. Just remember that the new files for expression (counts, FPKM, etc) are now all in one file called "STAR - Counts" I guess

There are also some very good posts here that can help you: Guide to TCGA data , Download data from TCGA

I just checked here the clinical data and there is no mention to LDH levels directly measured in the patients from skcm coohort.