Nowadays, with single-cell ATAC-seq we can find subpopulations within a cell-type that has a different chromatin accessibility pattern. I was wondering if there is a way of having some glance of this kind of heterogeneity from bulk ChIP-seq or ATAC-seq. Maybe some way of clustering that, for example, lets you know that in one cell type the ChIP or ATAC signal has different components.
Imagine: I have tissue samples and I FACS sort for two cell types, cell A and cell B. I suspect cell A is actually more complex and has subtypes within. For this, it would be great to do a follow-up single cell experiment. Imagine that I cannot do that - is there a way of digging into that from the bulk ChIP or ATAC? Thank you !
•
written
by
msimmer92
This is a pretty tough question for the data in question. There are several methods to do so from DNA methylation data, but that's because DNA methylation is binary - it's methylated or not, whereas ATAC and ChIP signal is much more messy. There are several papers that deconvolute bulk samples after performing single-cell ATAC or RNA sequencing on them (like this one), but I haven't seen any ChIP deconvolution in my brief search.
This paper might help:
Mars, J. C., Sabourin-Felix, M., Tremblay, M. G., & Moss, T. (2018). A Deconvolution protocol for ChIP-Seq reveals analogous enhancer structures on the mouse and human ribosomal RNA Genes. G3: Genes, Genomes, Genetics, 8(1), 303-314.
Log in to answer this question.
