I am having a single cell RNA seq data from ~500 Cells. I do understand this number is very low for present day analysis tools like Seurat. So, I want to know what if one have a such a small number of cells which tool works best?


  • $\begingroup$ you can perfectly use seurat as part of your analysis pipeline. It is hard to suggest tools or statistics without knowing the type of data you are looking at and what questions you want to address. $\endgroup$
    – Mack123456
    Jul 26 '19 at 21:59
  • $\begingroup$ @Mack123456 Thanks for your comment. I am trying to use single cell RNA seq data on MiNION. On a trial basis I just started with ~500 Cells. $\endgroup$ Jul 26 '19 at 22:09
  • $\begingroup$ Hi, you will need to forward as much information as possible, or simply accept @Mark123456 answer. The problem is that we don't know the objective of the study. $\endgroup$
    – M__
    Jul 27 '19 at 12:16
  • $\begingroup$ @MichaelG. Thanks. I am ready to accept the comment as answer. But I cant find the accept answer tag in comment section. $\endgroup$ Jul 27 '19 at 16:30
  • $\begingroup$ Hi @user3377241 no need these are just comments not a full answer. If you look at the other questions it's detailed technical advice rather than broad answers. If you have a specific technical issue you can return and ask about it $\endgroup$
    – M__
    Jul 27 '19 at 16:40

You can perfectly use any of the single cell processing packages out there. Seurat, Monocle3, Scanpy, Kallisto, etc. Following some of the tutorials and vignettes will most likely help you figuring out how to proceed.

These packages provide useful functions to explore single cell data, process the data and each of them has a series of advanced features to address specific questions. You don't need 10k cells to look at number of gene/cell, transcript/cells, % mitochondrial genes, etc. You can perfectly normalize and standardize 500 cells and run PCA, and analyze the complexity that describes your data.

To address if 500 cells is enough for you pilot we need to know what question you want to address. The following tool for example gives some intuition on how many cells to sample if you wish to detect n-cells per cluster.


Hopes this helps


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.