search

kstreet13 / slingshot

Functions for identifying and characterizing continuous developmental trajectories in single-cell data.
259 stars 42 forks source link

One lineage but 2 slingPseudotime #174

Closed rmathieu25 closed 2 years ago

rmathieu25 commented 2 years ago

Hi,

Thank you for developping slingshot!

I want to get only one lineage in my data. I did a kmean with k=2 and I indicated the starting cluster in start.clus argument (start.clus=2 in that case). I obtained one lineage but I got 2 slingPseudotime values and the minimum pseudotime value did not begin at the starting cluster.

I do not manage to understand why I got such a result?

Thank you in advance for your answer

Capture d’écran de 2022-02-09 16-03-38 Capture d’écran de 2022-02-09 16-28-43 Capture d’écran de 2022-02-09 16-28-58

Capture d’écran de 2022-02-09 16-17-00

kstreet13 commented 2 years ago

Hi @rmathieu25,

Thanks for bringing this up, it is indeed a strange result. I think your strategy of only using two clusters to ensure that you get a single lineage while still being able to set a starting point makes perfect sense. So it is strange that there would be a slingPseudotime_2 variable. However, there is only one curve listed in slingCurves(sds), which indicates a single lineage. Is it possible that the slingPseudotime_2 variable is an artifact from an earlier run? And if not, can you provide the code you used to run Slingshot?

Thanks, Kelly

rmathieu25 commented 2 years ago

Hi,

My bad, you're right slingPseudotime_2 is an artifact from an earlier run. So everything is all right!

Sorry to have bothered you for that.

Thank you again

kstreet13 commented 2 years ago

No worries, glad it got sorted out!