Hi there to all. I want to analyze an extensive (FACS) single-cell RNASeq dataset, from Tabula Muris https://tabula-muris.ds.czbiohub.org/
The count matrix data are publicly available on line. I've processed the data as exposed in the method of the paper subsetting for cells with at least 500 quantified genes and Cells with more than 50,000 reads.
I'm not interested in catching differential expressions between cells. What I would like do to is correlate the expression of some genes accross cell lines. And also understand which genes are more expressed in each cell types. (Without comparing)
I use the annotated cell types given by the author, I don't need really to run again all the analysis. In order to understand which genes are more expressed in each cell type, I was thinking about trasforming for each cell the raw counts in TPM, and then average the TPM from each cell belonging to the same type in order to get an average expression for each gene in each cell lines.
For correlating gene expression, I was wandering of using CPM for each cell, average them according to their cell types and then correlate different genes in different cell lines.
It is this approach correct? Would you suggest some other method?I'm pretty new to scRNA-seq, I'm mining this data more for personal curiosity, but I really would like to hear your ides about.