RNAseq requirements: Biological and Technical replicates

Question

[Rim]

Hello,

I am planning to perform RNA-seq on blood samples with a limited number of biological samples (from patients vs controls). Are technical replicates mandatory? If so, at which step of data analysis should they be considered, and with which tools, given that DESeq2 and limma are designed to handel biological replicates?

Thanks!

[jkim]

I don't think technical replicates are necessary.

[Ram]

Why not just say "technical replicates are necessary"?

[jkim]

Ouch, my mistake. Updated my comment.

Answer 1

[ATpoint]

What is a technical replicate to you? A repeated library prep on the same pool of RNA? If so then no, I don't see the point as downstream analysis, at least the ones I know or usually do, cannot make good use of these, so it would be a waste of money and time. What is critical is to design the study well. Meaning, RNA extraction and library prep should be done at the same time if possible, or in group-balanced batches. Usually specimen from clinical samples come in in irregular intervals, so I would take the cells/tissue and snap-freeze it in a trizol-like preservant so extraction can be done without batch effects.

[i.sudbery]

However, one might argue that multiple biological samples (i.e. multiple blood draws) from each patient might be useful. Still, probably not as useful as more patients.