Subset specific clusters from drop-seq data, merge them and produce geneXcount expression matrix (bulk-seq format).

Question

[cyd2gc]

I have been following the R script in the link below: https://github.com/jeremymsimon/MouseCortex/blob/master/E14_processing.R

My goal is to isolate only the radial glia clusters (which would be clusters 8, 10, 14 and 21), sum the expression in all those clusters, and create a gene X count expression matrix. Can one achieve this as an extension of the script above post acquiring the E14.5 clusters?

I see examples of how to do the above with Seurat objects but I am not sure if that applies to S4 class object in this script.