Bulk RNA-Seq Data That's Only Tumors (No Normal/Reference)?

Question

[James]

Hi,

I have a dataset of only tumor samples and their raw counts, and I'm not really sure where to go from here. Can I run DESeq2 with only Tumors and no reference group? Should I be grouping the samples using clustering in order to use those are the different groups to run the analysis. I'm a little stuck. I really just want to look at which genes are more/less expressed and get some general insights from the data.

TIA!

[GenoMax]

Answers/comments here encapsulate what you should be thinking about --> What Is The Proper Control Sample For A Tumor Rna-Seq Experiment?

[Rashid]

Start with differential gene expression and possibly run through a deconvolution process after you group by cell type or by gene. Prior to that you would need RNA seq data to do anything and ideally sorted.

[Ram]

Prior to that you would need RNA seq data to do anything and ideally sorted.

What are you talking about?

[gglim]

Maybe you could refer to some published datasets?

[ATpoint]

and I'm not really sure where to go from here

There is the problem: You do not tell what the scientific question is that you want to answer. Without knowing that one cannot advise what to do. It's the most critical thing to know!

[James]

I just want to look at which genes are being Differentially Expressed. Right now I am using K-means clustering to separate the groups and then run the analysis that way

[ATpoint]

Differential between what? "I just want to look..." is not a scientific statement. I can only help you if you clearly state what question you want answered.