TCGA RNAseq normal and tumor samples problem

Question

[morovatunc]

Hi,

First of all, I want to clarify that I dont think TCGA can do RNA seq only for tumor sample? This is nonsense. This information must be somewhere. So, Can I reach this information with TCGA-Assembler.

1. referring to this post

TCGA: Does TCGA cancer studies have mRNA expression data for Control/Normal samples? I tried to use TCGA-Assembler however it does not give this information neither. Is TCGA-assembler let you download data that cannot be seen by the Data matrix? or It just eases the download procedure ? Because that I understood from the answers was not all the information was visible in data matrix so this R package allows to download all of them ?!?

1. When I checked the entries one by one, even though some of the entries have tumor and normal pairs, big majority of them are either only tumor or normal. I think this is wrong?

Any information is useful.

Thank you for your help,

Best,

Tunc.

Answer 1

[Amitm]

hi, AFAIK, not all TCGA cohorts have tumor-normal paired samples for RNA-seq. Some studies have e.g breast-cancer, and some dont like SKCM. If you look at this recent study, they analyzed all cohorts in TCGA RNA-seq which had paired normal available. You can restrict yourself to these cohorts (mentioned in the article) only if you want paired controls.

Answer 2

[Charles Warden]

For most cancer types, you will typically find tumors without paired normals (but you should almost never find normals without tumors).

Try comparing those resources to the TCGA FTP data:

https://tcga-data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/tumor/

For example, for breast cancer see nationwidechildrens.org_biospecimen_sample_brca.txt in https://tcga-data.nci.nih.gov/tcgafiles/ftp_auth/distro_ftpusers/anonymous/tumor/brca/bcr/biotab/clin to map patients to samples