Compare fold change and p values for the same gene between different DESeq runs
Entering edit mode
15 months ago
kw486 ▴ 30

I have a question which I would like a second opinion on.

We have a single cell RNA-seq dataset containing cells of two different types (A and B, assigned by us based on characteristics such as tissue of origin). We have separated the dataset by type and generated 2 volcano plots of differentially expressed genes.

In short - we produce two volcano plots from two separate runs of DESeq2 on the two halves of the same dataset.

The DESeq pipeline is identical for both and the contrast we are interested in (cells from day 8 vs cells from day 18) is the same; both runs have the same design formula:

  • design= ~ 0 + Day
    • Cell type A --> Day 8 vs Day 18
    • Cell type B --> Day 8 vs Day 18

A member of my group asked me whether we could directly compare the x, y values (log fold change and log adj p value respectively) for a particular gene between the volcano plots. My initial answer was that we could not, because different runs of DESeq means different dispersion estimates and different normalised count values. Would appreciate if anyone could confirm or correct this?

RNA-Seq DESeq R volcano single cell • 422 views

Login before adding your answer.

Traffic: 1394 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6