15:32:32 From Pascal Sturmfels to Everyone:
	sorry folks some technical difficulties
15:33:54 From Lee Wohlen Organick to Everyone:
	Looks and sounds great!
15:43:50 From Sara Mostafavi to Everyone:
	question
15:46:00 From Sara Mostafavi to Everyone:
	Thanks
16:02:08 From Elisabeth Rosenthal UW to Everyone:
	these are volcano plots, correct?
16:24:13 From Elisabeth Rosenthal UW to Everyone:
	What I don't understand is what the UMAP plots are meant to tell us. Shouldn't we be able to see a pattern without the colors (like PCs of ancestry)?
16:25:34 From Elisabeth Rosenthal UW to Everyone:
	yes
16:32:04 From Pascal Sturmfels to Everyone:
	Some discussion in the room about notions of distance, e.g. nearest neighbors vs. l2
16:35:12 From Elisabeth Rosenthal UW to Everyone:
	Could label-aware be thought of as a Bayesian method, using prior information? Of course if there is overwhelming data, the prior won't matter
16:36:43 From Elisabeth Rosenthal UW to Everyone:
	It's not currently structured as Bayesian, but I think it could get there
16:37:24 From Elisabeth Rosenthal UW to Everyone:
	thanks