"identifyOverExpressedGenes()" Same data but different results

[wahahaaaa123]

I got result A using a dataset of 20 cell lines.

I extracted two cell lines from the same data set and calculated again, and got result B.

A and B have different results.

A and B have a large overlap, but the number of A is more than that of B, and there are some parts in B that are not in A.

I would like to ask what is the reason for introducing such a problem?

If I focus on the relationship between 20 cell lines, what if I calculate the pairs separately? Or calculate together? What's the difference?

Here is my calculation process, just the input data is different： cellchat <- identifyOverExpressedGenes(cellchat) cellchat <- identifyOverExpressedInteractions(cellchat) cellchat <- computeCommunProb(cellchat)

[sqjin]

@lijiyang This is possible due to the variability across different cell lines, which is similar to the case with different patients.