Comparison of CyTOF and Flow Data

[emallen]

Hello all, does anyone have a tried-and-true way to compare multi-parameter CyTOF and Flow data? I would love to see gating strategies that have worked for other people, especially for the initial gating (cells vs debris, singlets, L/D, etc). Or, if these populations are incomparable, is it possible to use a widely expressed marker to obtain comparable populations (ie CD45 for PBMCs)?

[emallen]

If it makes a difference, I'm working primarily with human PBMC.

[mleipold]

We've generally had good luck using the same set of markers by flow and by CyTOF, and just looking at percent of parent.

I will say, though: if you're looking at flow, you will want to do marker-based gating rather than scatter-based, for monocytes vs lymphocytes: we occasionally found very small but reproducible percent-of-parent differences when using scatter rather than the same CD33 vs CD14 bivariate to separate them. (Probably also applies to lymphocytes vs granulocytes, though since we looked at just PBMC, I can't comment directly).

Generally, my PBMC gating strategy goes:

0) if using beads, Ir vs Ce140 to gate Cells (Ir+ Ce140-) from Beads (Ir-Ce140+) and Bead-cell doublets (Ir+ Ce140+). You'll want to do this, as several other bead channels are ones you will use for antibodies; if you don't remove these doublets, you'll have some unexplained bright populations in those channels.
1) Ir vs Ir: high Ir191+Ir193+ for Intact cells.
2) Ir vs Cell length: Cell length 20-50 would be Intact Singlet.
3) Cell length vs Live-dead: LD- as Live Intact Singlets.
4) CD33 vs CD14 for Monocytes (CD33+ CD14+ mostly, though single-positives as well) vs Lymphocytes (CD33- CD14-).


I've added a couple examples.


Mike

[emallen]

Thanks! This is very helpful... do you have a companion gating strategy for flow data? You caution against using scatter for analyzing flow data, but how do I gate out doublets and debris that is comparable to CyTOF gating?

[mleipold]

I don't have a companion one for flow. I'll see if I can get one from my labmate who helped me on that comparison.

However, what I meant was that you shouldn't use scatter to gate specifically on, say, monocytes vs lymphocytes like some people do. Using scatter to get rid of debris and doublets is usually fine.