Minimum number of UMI counts/cell

Question

[otieno43]

I have dataset in which majority of the cells have between 250-600 UMI counts (reads=nRNA). I know it is recommended to filter out cells with less than 500 UMIs. This number 500 I believe is based on data from model organisms like mice and rat that is little easy to obtain thousands of cell. My data is from non-model organism and obtaining individual cells is not very easy. I was wondering if it is possible to scale down to 250 UMIs and what is the implication of this on the quality of results which will be obtained in the downstream analysis.

Thanks

[ATpoint]

That cannot generally be answered as we do not see your data, nor know the characteristics of your organism. I would suggest to follow a standard tutorial (OSCA from Bioconductor or Seurat, or Python equivalents such as ScanPy tutorials) and then simply see how it goes (ignoring the recommended thresholds) and checking whether results make sense. Intuituvely 250-600 UMO 250-600 genes detected at all, but who knows. Just explore and see.