Hi Elena,

Yes indeed we have been using 10X Genomics for our scRNAseq in the paper A vd Heuvel et al Hum Mol Genet 2018.

So far we have been very happy with the results. I guess the outcome will also depend on the question you would like to answer, but so far the 10X system has been very suitable for us.

Are you interested in using it yourself?

Kind regards,

Anita

Our Core Facility (Northwestern University NUSeq Core) has adopted the 10x scRNA-seq platform for a year and a half now. This platform has been working pretty well for us.

I have used 10X Genomics 3' scRNA-seq V2 and V3 as well as snATAC-seq kit for multiple iPSC derived cell types. Always works like gold. I think the success of this company is mostly due to robustness of their kits.

Do you have specific concerns?

Hi Anita,

thank you for sharing your experience, this is good to know! I am new to the scRNAseq and checking different options and protocol. We would like to do it on zebrafish embryos.

Have you worked with freshly isolated/prepared cells or did you also used cells that were stored before?

Kind Regards, Elena

Hi Agnieszka,

Thanks for your answer. I was wondering how noisy it is. What was the minimum number of cells you could detect the expression profiles in? For example, if we are interested in a small population of cells (5-10%) of the total sample, will we be able to detect it? Let's say total cell number is 1000 cells. Or would you push it higher to 5000-10000 cells?

I would say this depends a lot on the transcriptome signature of your low population. We have for example been able to detect the presence of sometimes only 1-2 disease-affected cells out of 700-800 cells, but this was only possible because we had quite a clear disease signature geneset...

You can perform a reliable genomic analysis from as little as total of 500 cells but you probably should stay on the higher end. Everything depends on how heterogeneous is your population of cells and what is the purpose of the analysis. You wrote that your cells of interest contribute to 5-10% of the population but is remaining 90-95% homogeneous or also very heterogeneous? And, most importantly, how specific to your cells of interest is the expression profile of your cells - is it a continuous change of a certain gene set (i.e. differentiation of maturation of a cells) or is it a very specific cell type that has a very specific expression profile. If you are only planning your experiments I can recommend the V3 version of the scRNAseq 3' kit which allows for even more comprehensive expression profile. If you can, you could capture 10,000 cells and in this way your population of interest will be quite sizable and you will be able to detect even subpopulation within your cell type of interest.

Thank you so much Anita and Agnieszka for your detailed answers. It is very encouraging to know that the system has the potential to detect low number of cells. We expect the 90-95% of the population be rather homogeneous during the time frame we are interested in. For the cells of interest, we expect a change because there is a massive actin polymerization that we observe in these cells. We do have a few marker genes for the cells of interest, which would hopefully, help us to track them down.

That's sounds like your analysis should be very straightforward. To go back to your original question: I would always recommend going for the high number of cells. Good luck!