get lots of NA value for slingPseudotime_1 about slingshot HOT 3 CLOSED

Hi @lixin4306ren

This is likely because Slingshot is detecting more than one lineage (the "_1" in the variable name indicates that there are more lineages). Cells that were identified as being specific to Lineage 2 will have NA values for slingPseudotime_1, but they will have real values for slingPseudotime_2. These assignments are based on weights (accessible via slingCurveWeights), where a weight of zero will cause the corresponding pseudotime value to be NA.

from slingshot.

That was the case, thank you. Another related question, I tried to plot Pseudotime of all cells from all lineages, but the shared cell of multiple lineages showed different Pseudotime values. Eg, slingPseudotime_1 and slingPseudotime_2 of shared cells are slightly different (some showed big difference), which will result in different color codes. Which slingPseudotime value should I use (average them or just pick random one)? And how should I fix this problem? Thanks in advance.

from slingshot.

I think this depends on your downstream analysis plan. For plotting, you can either pick one value or plot both values with some transparency based on the weight of assignment.

If your downstream methods require a strict assignment (or only a single pseudotime value), then you could either select the value with the larger weight or use the weights as multinomial sampling probabilities and assign them using a random draw.

from slingshot.