kstreet13
commented
2 years ago Hi @Rahul1711arora,
That's an interesting question! I think there are some things you could try, though I don't know if any of them are exactly what you described.
First, I would definitely recommend checking out our workflow for comparing trajectories across conditions. This seems like a case for treating the different datasets as different conditions. There are a number of ways you can produce a joint dimensionality reduction by integrating the datasets (we use Seurat integration in the workflow, but there is also mnnCorrect, Harmony, etc.). If they are sufficiently similar, this could lead to a joint trajectory analysis.
Alternatively, you could try comparing sets of dynamically expressed genes. So for each dataset independently, you run slingshot and tradeSeq, resulting in a list of significant genes for each dataset. Comparing these lists could be informative. Similarly, if you rescale the pseudotimes to [0,1], you could even try combining them and running tradeSeq as if they were different conditions.
Hope this helps! Kelly
Rahul1711arora
Hi,
I am currently working with single cell RNA data and is new to slingshot. I have two different datasets from different labs. Now, I intend to use slingshot for both independently giving two separate principal curves. My question is, is there a way to compare the two separate principal curves and see if there is any link between them? As in if the two diseases are linked at the cellular level? (It is given that they come from different diseases but from same cell type). Or perhaps a way to plot these two trajectories together on a same plot? Thanks in advance for your help.