Is total miRNA in EV constant? Implications for library size normalization.

Question

[Thomas.H.Hampton]

I often see studies using tools like DEseq2 or edgeR used to normalize reads in samples to quantify miRNA expression. Technical artifacts are a problem, and I understand why people studying miRNA in extracellular vesicles (EV) would be tempted to use tools that have been successfully used in other domains. However, I've always found the underlying assumption that mRNA production by cells is constant regardless of treatment a little surprising, and library size adjustment by edgeR and DESeq2 requires that assumption to hold. Maybe cells just churn out a given amount of mRNA no matter what is going on.

It feels less likely to me that cells produce miRNA at a constant rate no matter what, and quite unlikely that cells package a fixed amount of miRNA into EV not matter what. Do others share my suspicion that total miRNA in EV often varies as a function of treatment? If it does, then library size normalization is invalid for this type of data.

Answer 1

[Gordon Smyth]

Library size normalization in edgeR does not asume that mRNA production is constant. On the contrary, the edgeR normalization factors can be interpretted as estimating changes in overall mRNA production.

The actual assumption is that most genes (over 50%) are not DE. Consider a toy example with 10 genes. Suppose that nine of the genes have constant expression between conditions but gene 10 is enormously up-regulated in the second condition, so much so that total mRNA production over all genes in the second condition is doubled. If you (wrongly) assumed that mRNA production was constant, you would have to conclude that genes 1-9 are all downregulated in the second condition and only gene 10 is up. edgeR instead will infer that total mRNA production is up in the second condition, and will adjust the library sizes so that genes 1-9 are roughly constant and only gene 10 is changed.

I wonder where you have read that mRNA production is assumed constant. No such statement has ever been made by the edgeR authors. Similar comments would apply to DESeq2.