I can obtain different results when I process my data through ti_slingshot or R package "Slingshot"

dynverse / dyno

Inferring, interpreting and visualising trajectories using a streamlined set of packages 🦕

https://dynverse.github.io/dyno

Other

163 stars 32 forks source link

I can obtain different results when I process my data through ti_slingshot or R package "Slingshot" #103

Open chuiqin opened 3 years ago

chuiqin commented 3 years ago

Thank's for your great work! It's easy for us to infer different trajectorys by Dyno. But I obtain different results when I run my data through ti_slingshot of dyno and R package "Slingshot" with the same dimensionality reduction. Can you help me please?

dkrozelle commented 3 years ago

Wanted to piggyback on this question since I'm experiencing a is similar issue. Didn't know if any insight from @kstreet13 ?

I'd also love if someone can describe the method (or point me to documentation) on how the individual slingshot lineages are collapsed into the branched tree displayed with the dyne plot_dimred method. When using the dyne methods I can't seem to access any of the individual lineage curves?

kstreet13 commented 3 years ago

It looks like ti_slingshot is directly calling slingshot, so this is a strange issue. If anyone can provide a reproducible example, I'd be happy to play around with it.

Also, I should warn everybody that Slingshot is getting an update with the new version of Bioconductor (coming in May). I think ti_slingshot will be affected, but only very slightly (we are switching from SlingshotDataSet objects to PseudotimeOrdering objects. All the accessor functions still work except for reducedDim, which should be changed to slingReducedDim).