RNASeq data analysis without control

Question

[tothepoint]

Dear All, I am facing an issue with data analysis. I have RNASeq sample from three different tissues and total 15 samples but with no control. I am trying finding a lead to get DEGs from them but no success yet. I would be grateful to you all if given any possible way to proceed.

Thanks

[ATpoint]

It is always interesting that users expect the community to read minds. What is not working? Which software and commands did you use? Which comparisons did you make? What is the scientific question?

[tothepoint]

I apologise for the same and will try not to do this. We have 15 samples 5 each tissue. I already gone through QC-Trimming-Indexing-Mapping- and now at cufflinks cuffmerge cuffdiff stages. I have seen one post suggesting treat one as control and other as experimental but that for treating time series. I am confused what I can do as no such variation we have in our sample. Any lead in this regard or different pipeline/steps is what I am trying to explore but still not found any success yet.

Answer 1

[ATpoint]

So you have mutliple groups and each group has replicates? Then you can perform pairwise analysis between these groups and from this get differentially-expressed genes. I suggest you abandon the deprecated cufflinks pipeline and use a more recent DGE tool. Options can be among others DESeq2, limma, edgeR. Each has outstanding documentation. I suggest you read them first, this will also help getting a background what can be done with RNA-seq data.

[tothepoint]

Thankyou for suggesting DGE tool. I will definetly opt for the same. As I mentioned earlier we have three groups (L1(1-5), L2(1-5), L3(1-5)) how can we perform pairwise analysis I am confused. Please elaborate for the same I will be grateful to you.

[ATpoint]

You can compare L1 to L2, L1 to L3 and L2 to L3. Please read the documentation of the tools I linked, they explain how pairwise comparisons are performed.