Hi guys.
I am planning to do single cell RNA sequencing on the neural progenitor cells of embryonic mice (E11.5) using the 10x Chromium approach. The plan is to perform DE analysis on a mosiac population of the progenitors to understand how their transcriptional profile differs from each other. I'm just wondering what would be the appropriate sequencing depth for an experiment like this? I've read that 20,000 reads/cell is enough for accurate clustering, however as I said the main part of this experiment is the DE analysis of the moasic cells so I believe we would need greater depth. Is there a cap for this sort of experiment using 10x? Has anyone using 10x to sequence mouse neural progenitors before? Thank you in advance!
•
written
by
foxiw •
20k is the minimum recommended by 10X, most people shoot for 50k for increased library complexity and a bit of a buffer.
Hi Jared,
Thanks for your reply. As I understand it, sequencing saturation occurs at a depth of around 50,000 reads/cell, so anything higher than this is a waste of money. However, I have also seen some people sequence at 100,000 reads/cell. Are there any advantages to sequencing at this depth, or is it just overkill? Thanks!
Every experiment/library is going to be different. You may need to experiment with yours to see if 50K is enough or 100K would be better. Bottom line is if your cells are expected to have less RNA content then you will need less sequencing to achieve saturation.
https://kb.10xgenomics.com/hc/en-us/articles/115005062366-What-is-sequencing-saturation-
Agreed. I'd check what some other papers that looked at NSCs/NPCs did. And consider a pilot experiment with 1 or 2 samples. For most tissues I've looked at, you start to get diminishing returns much over ~70k reads, but as Genomax said, it really depends on the experiment.
Log in to answer this question.
