In scRNA-seq -how does one tell the difference between cells which are in a different temporal phase than other cells hence have a different gene-expression signature, versus actually different cell-type? Is there any way of ensuring a cell population is synchronous before it is subjected to single-cell RNA-seq?
Or, is it that different cell-types are simply defined by those having a different transcriptional profile at that time point, compared to other cells?
You can certainly synchronize cell cycle across your cells via treatment with a mitotic inhibitor or serum deprivation prior to sequencing, but doing so will likely impact expression of other genes in a fairly significant way as well. This paper provides a fairly good review of the options.
You can also regress out differences in cell cycle phase from your dimensionality reduction, which will help alleviate clusters dominated by cells in a given cell cycle state. This can be done with scran or Seurat (among others) quite easily.
Some cell types are indeed characterized as actively proliferating (activated cytotoxic T cells come to mind), so removing these differences may diminish real, interesting cell populations, but it's up to you to determine whether that may be the case for your cells or not.
