Point of merging?

[t1nnenissen]

Hi all,

I have a question about the my data analysis that I would be really like some input on.

I have an experiment with two time point and three treatment groups. All the cells were harvested the same day and the were then hashed together three wells on the same ship (the two different timepoints that got the same treatment being hashed together).

As far as I understand I so not need to integrate as all the cells were harvested the same day and run on the same chip. However, I am in doubt about when I should merge! Should I run scTransform on each of the three files data outputs and then do a merge before clustering, or should I merge them and then run scTransformed on the merged object?

Apologies if this matter has already been covered!

Best,
Tinne