viSNE analysis and clustering

[hamidullah]

Hello,

I am not sure if this post will ethically fit on this platform, if not I deeply apologies.
I am working as a postdoc in University of Wisconsin-Madison. I have single cell fluorescence intensity for around 10K cells(5-different proteins) out of IHC images. I am trying to map 5-diffirent proteins as 5- different clusters in one single map using viSNE. Is it possible to do the same? (like image 6)
I am using FCS express 6 viSNE analysis but unable to make 5-clusters in a single map instead software generates 5 different maps for each single proteins (please see images protein-1-5). I am not sure how to do the same.
I am very naïve to viSNE analysis
I would really appreciate if you could help me to figure this out.
Thank you very much
Hamidullah

https://imgur.com/a/dYM69Oo

[mleipold]

Hi Hamidullah,

I'm not sure I understand your concerns.

In your case, you *do* have 5 different plots. However, those plots are of the *same* tSNE analysis......they are just colored for *each* of your proteins (5 proteins, therefore 5 plots). That's the default for tSNE dimensional reduction.

However, tSNE dimensional reduction is *not* the same as clustering, like you might have seen in tSNE plots from something like Cytofkit (Phenograph/DensVM/FlowSOM/CusterX). In those cases, you can still color the plot by a particular marker, but there's *also* the option to color the plot by Cluster (akin to your plot 6).


Does that answer your question?
Mike