QC before single cell RNA seq?

Question

[jevanveen]

Background: We have done single cell RNA seq, it went well, and we are preparing to do more projects.

When we did it the first time, we troubleshot our cell dissociation protocol and made sure that we could get good RINs from bulk RNA before proceeding. Now we are dealing with projects that have fewer cells available for isolation making it harder to verify RNA integrity before proceeding. What do you guys do to make sure your cells are healthy and intact before proceeding with a costly experiment like scRNA seq? Just live/dead and then gun it, and sort it out informatically on the back end?

Thanks in advance, sorry if this question is a little wet for biostars.

[Kevin Blighe]

I am sure that the service provider who will do the sequencing can offer tips. I know hat 10x and Illumina have tips online, too. There's a lot of single cell RNA-seq (scRNA-seq) going on here at my institute. Getting the best cells can save a lot of headaches further down the line for the analysts. Indeed, from the analyst's perspective, a big issue with scRNA-seq is low counts, to the extent that many genes come back with nil / zero counts. Algorithms can attempt to 'impute' these missing values, but it is obviously preferable to avoid having to do this. Equally, the 'market' is now swamped with scRNA-seq tools, as documented by this recent publication: Exploring the single-cell RNA-seq analysis landscape with the scRNA-tools database