Manually adding a macrostate? about cellrank HOT 4 CLOSED

Hi @mxposed,

Re 1: What's the biological question you're trying to solve with this? If you know the trajectory starts in monocytes, and you initialize your pseudotime in this cluster, then what additional information can you extract by identifying a macrostate corresponding to these cells? Usually, when the initial state is well known, terminal states, and fate probabilities towards terminal states, are more interesting.
Re 2: Yes, if you use the PseudotimeKernel, then the transition matrix depends on the Pseudotime you pass. Please see our CellRank 2 preprint for more information.

from cellrank.

Hi @Marius1311

Thank you for the quick reply. Re 1: you're right, I'm interested in the terminal states, their number and identity, the shape of the trajectory (branches, no branches), distribution of cell commitment towards them, if there is any difference between 2 experimental groups of samples, and the genes that are behind.

from cellrank.

great, in that case, it looks like you don't need to worry about your initial state. Closing this issue, feel free to reopen if your question persists.

from cellrank.

Thank you! I didn't realize I can proceed with these analyses without identifying initial states first!

from cellrank.