search

sqjin / CellChat

R toolkit for inference, visualization and analysis of cell-cell communication from single-cell data
GNU General Public License v3.0
622 stars 142 forks source link

Confuse about the Ligand-Receptor interaction probability and signaling gene related expression #497

Open QianLLLL opened 1 year ago

QianLLLL commented 1 year ago

Hi! I felt really confused about the signaling-related gene expression and the ligand-receptor pair when I plot them.

Screenshot 2022-10-28 at 10 08 00 This figure showed that there had an interaction between T/NK cells and AST1 in the ApoeKO group but not in the ApoeWT group.

Then I checked PSAP signaling pathway-related gene expression.

Screenshot 2022-10-28 at 10 10 28 ApoeKO group Screenshot 2022-10-28 at 10 10 57ApoeWT group

I didn't see a significant difference in PSAP signaling pathway-related gene expression between these two groups.

What is written in the CellChat article is that the calculation of intercellular communication probability is based on the expression of genes. However, if the gene expression in the two groups of cells is not significantly different, how should I explain the obvious difference in ligand-receptor interaction between them?

Looking forward to your reply! Thanks in advance!

Sincerely, Qian Liu

sqjin commented 1 year ago

@QianLLLL Sorry for the late. You might need to check the calculated prob and pval, which give the interaction strength and pvalue information. you may can figure out the reason.

QianLLLL commented 1 year ago

@sqjin No worries! Thanks for your answer!

So from what I'm understanding, the Biologically significant CCI not only depends on gene expression but also on pval, right? If there is a significant difference in ligand-receptor gene expression, it might be explained more simply, otherwise, I need further to check pval and prob. I don't know if my understanding is reasonable.

image

And I got another question about the formula in the picture, I don't understand the AN. From the paper method part, I just saw the N was defined as the number of ligand-receptor pairs. I'm a little confused about the AN definition.

Looking forward to your reply! Thanks again!

Sincerely, Qian Liu

hyjforesight commented 11 months ago

Hello @sqjin and @QianLLLL , I met the same situation that I got significant communication of certain signaling when comparing 2 datasets, while the gene expressions were not quite different between them. Could you please explain more about this? Thanks! Best, Yuanjian