Hello everyone,
I am running ligand-receptor analysis using LIANA on a single-cell RNA-seq dataset. Immune cells only are sequenced (CD4, CD8 and NK). When running ligand-receptor analysis, I am not getting highly specific interactions. Are there any biological reasons that could explain such results?
Thank you very much!
Immune cells frequently use similar signaling pathways and receptors for a variety of activities. Certain cytokines and chemokines, for example, bind to receptors found on several types of immune cells, resulting in common signaling responses. This can lead to several identified interactions that are not highly selective to a single ligand-receptor pair or cell type. That's one reason I thought about. Isn't that related to the code you are using?
one potential reason for this could be the inherent sparsity of single-cell RNA-seq data. Single-cell RNA-seq often results in a relatively small number of genes being detected per cell. This sparsity might reduce the statistical power to detect specific, low-frequency ligand-receptor interactions, especially if these interactions are not strong or common in the dataset. Therefore, the lack of specificity could be due to the limitations of single-cell RNA-seq data regarding gene detection depth and statistical power for precise ligand-receptor interaction analysis.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
what do you mean by "Isn't that related to the code you are using?"
I was thinking maybe this is related to the code you are using, in case of any error or anything like that.
No actually there are no errors. Just needed to understand the biology behind that. Any other thoughts?
The other thing I think about is maybe the ligands can bind to other receptors too (off-target receptors) or maybe false positives. Have you ever worked with CellChat within the LIANA package, or do you have any experience on it?
Yes, I used it once and I am planning actually to use CellChat in this dataset, independently from LIANA.
Nice, Do you think we can discuss more on this?
Do you have any insights or ideas to take them into consideration?