sqjin
commented
2 years ago @LucaTucciarone Sorry for the late reply. I would suggest to pull together 10 CTR samples and run CellChat. At the same time, you can run cellchat on the aggregated 8 disease samples. Lastly, you can perform comparison analysis using the tutorial.
Since CellChat infers the communication probabilities based on the average expression of signaling genes in each cell group. Thus the cell proportion will not affect the results. Please set 'population.size = FALSE' when using computeCommunProb(cellchat)
To have a list of inferred L-Rs, please check the tutorial on the function 'subsetCommunication(cellchat)'
Hello!
I have been using cell chat quite a lot this past weeks for my project and I think is great. I have some doubts on how I am supposed to be using biological replicates of the same experimental point though: Let's say I have 10 scRNAseq biological replicates for CTR and 8 scRNAseq biological replicates for Disease and I wanna run cell chat: should I need to "pull" the scRNAseq together and run the "individual" sample pipeline? (https://htmlpreview.github.io/?https://github.com/sqjin/CellChat/blob/master/tutorial/CellChat-vignette.html)
I did that and I don't know if it's the right way to go. Then, other than that, how cell chat handles cell pops with different cell concentration? So let's say I have this cell pop "A" that seems like to be the main "sender" but is also the most abundant one. Does cellChat take into consideration the relative "Abundance" of cells into the assay? Does it normalise in any way?
last question: Let's say that I'd like to keep my 10 bio samples separated to then analyse them together with the "comparative" pipeline. (https://htmlpreview.github.io/?https://github.com/sqjin/CellChat/blob/master/tutorial/Comparison_analysis_of_multiple_datasets.html) Would that be possible?
Also, what's the best way to take the raw data out of the program? I would need to have lists of Pairwise cell-cell interactions (signalling and ligands).
Lemme know! Thanks