Forum:bulk seq vs. sc-seq

Question

[SebKittelmann]

Hi all, I have so far only used bulk RNAseq experiments to study gene expression differences between conditions. Now I am planning experiments to determine DGE and differences in chromatin accessibility between wild type and mutant cells after FACS to isolate a more or less homogeneous cell population. I would have thought that bulk-seq (with enough depth) would allow me to identify differences, but was wondering if sc-seq has any clear advantages here? I guess it would be much more expensive, but would I even be able to find differential genes/accessibility if the differences are subtle? Or is it difficult to reach enough read depth in sc-seq experiments for this?

Answer 1

[dsull]

Based on your biological question, just go with bulk. Single-cell has its limitations including more technical noise.

Only do single-cell if you need it. Like, if there are subpopulations you're interested in identifying (which doesn't seem to apply to your situation unless you have a biological question about expression at different points in the cell cycle or are interested in modeling things like splicing rate, bursting, etc.).

[ATpoint]

Agree on that. Generally, single-cell (like 10x Chromium, which offers ATAC-seq, RNA-seq or a multiome combination) is good for heterogeneity and it allows to in silico deconvolute your sample, so your FACS / enrichment can be imperfect or absent. However, these assays on single-cell level are sparse and noisy, and differential analysis is a mess. You would need biological replicates to do pseudobulk analysis, for example via hashing individual donors, and this again is then not strikingly different from a bulk experiment, given that your population is, as you say, rather homogeneous. So unless you have a good reason to use single-cell (I hate it by now, we use it a lot) go with bulk.