How do decide if a processed scRNA-seq data is read counts or UMI format ?

Question

[JACKY]

I found this scRNA-seq atlas, click here to take a look. But, I'm unsure if it uses read counts or UMI format. The paper mentions UMI, but many datasets I've encountered offer read counts even when they used UMI methods. So, I'm not entirely sure about the format used in this atlas. How can I check this to be 100% sure? Thanks!

[ATpoint]

It's 10x Chromium so UMI, though I wonder why for the end user it would matter...

[JACKY]

Cause I have other read counts datasets that Intend to combine it with. Can't combine UMI with read counts, if I'm not mistaken.

[ATpoint]

I think the count type is the least of your problems given the inevitable batch effect.

[JACKY]

Quick question. So you're saying whenever I see 10x Chromium it automatically means UMI counts, and not read counts ?

For example - This data, this, or this. All are UMI counts although the word UMI is not mentioned anywhere in the description.. ? Cause in all, "10x Genomics Chromium" was used. Sorry if I'm asking trivial questions I'm new to single-cell data.

And regarding the batch effect, I'm aware of that.

[ATpoint]

10x protocols use UMIs, and unless your preprocessing pipeline does not deduplicate UMIs then its UMIs. Again, I would not even bother.

[dsull]

Can you give an example of an analysis with read counts even though it’s UMI data? That would be highly unusual so I’m just curious.

[JACKY]

Never mind I was wrong. It was UMI all along..