Is RNAseq data from a monoclonal cell line technically considered scRNAseq?

Question

[sithlordashley]

Hello. My team is planning bulk RNA sequencing of a clonal human epithelial cell line we isolated. I understand that even though this is a clonal cell line, different cell populations may exist, and phenotypic indicators of this cell line seem to suggest there may be different populations.

I think it is important to understand what we are doing, and we are all fairly inexperienced with this type of work. This question may be semantic but is data generated from bulk RNA sequencing of a monoclonal cell line considered to be scRNA sequencing data, or is further processing needed for that claim?

If the latter is true, does anyone recommend packages or tools to estimate cell composition such as Bisque (R)?

Thanks in advance!

Answer 1

[Soheil]

Bulk RNAseq from any sample is still considered a bulk sample. Running bulk RNA sequencing on a sample with any level of cell type/state composition will result in a single expression profile (1xn where n is the number of detected genes) representing the average expression of genes across all cells. Bulk RNAseq and single-cell RNAseq data are usually used to answer different questions. The statistical assumptions and thus the methods used to analyze bulk and single-cell RNAseq data are also different.

If you are just interested in calculating cell type composition in bulk RNAseq data, you might be able to get away with a bulk deconvolution approach. Methods such as CIBERSORT allow you to deconvolute bulk RNAseq data according to a reference single-cell dataset. See this paper (Figure 2b) as an example. However, note that, even with deconvolution methods your data will never be considered as scRNAseq data and can not be used for a comprehensive heterogeneity analysis. Hope this helps.