non matched control for PDAC samples

Question

[Omics data mining]

Dear all, Dear all,

I have RNA-Seq samples of pancreatic patients. but I do not have matched control samples. Sequencing of matched control samples was not done, due to some technical issues. But, I want to use control samples in this study. What are possible non matched control samples that could you used for these type of samples ??? By looking into TCGA https://portal.gdc.cancer.gov/projects/TCGA-PAAD , I found 185 cases under project name Pancreatic Adenocarcinoma. I was wondering , If it a right or wrong , If I download some of the non matched control samples from TCGA and perform the differential study against samples that I have with me (also belong pdac) ??

All suggestions will be appreciated

Best regards Archana

Answer 1

[Kevin Blighe]

It would be preferable to not do that, i.e., use the TCGA normal RNA-seq data, as it would introduce an extra confounding factor into your study design. What were the technical issues that prevented your normal samples from being sequenced?

If you must use something (like the TCGA normal data), then you will have to introduce it into the design formula as a BATCH variable for the purposes of normalisation.

If you just use your tumour samples, then you can check expression of certain genes in the Genotype-Tissue Expression (GTEx) project. You could also conduct differential expression analyses between your tumour samples based on histopathotype.

Kevin