search

sqjin / CellChat

R toolkit for inference, visualization and analysis of cell-cell communication from single-cell data
GNU General Public License v3.0
637 stars 145 forks source link

Cannot find some signaling pathways #317

Open sel041 opened 3 years ago

sel041 commented 3 years ago

Hi!

Thanks for providing this package! It has been working great but I have a problem with finding some signaling pathways from Cellchat database. Out of the many signaling pathways, only a few worked. For example, I could analyze SEMA3 signaling but not SEMA4,5,7. If I try SEMA7, I got this error message.

pathways.show <- c("SEMA7") netVisual_aggregate(cellchat, signaling = pathways.show, layout = "chord") Cannot find SEMA7 . Please input a correct name! Error in ...future.FUN(...future.X_jj, ...) :

Any help? Let me know if you need more information.

Thanks! Se

sqjin commented 3 years ago

@raphaela2018 For individual L-R pair, you need to use netVisual_individual

sel041 commented 3 years ago

@sqjin Thanks for your prompt reply. I am still getting the same error with netVisual_individual.

netVisual_individual(cellchat, signaling = "SEMA7", layout = "chord", vertex.size = groupSize) Cannot find SEMA7 . Please input a correct name! Error in ...future.FUN(...future.X_jj, ...) :

Again, this works for some signaling pathways (SEMA3, BMP...) but not SEMA4,5,7. SEMA4,5,7 are not individual L-R pairs, they are pathway names that I could find on http://www.cellchat.org/cellchatdb/. Any more suggestions? I appreciate your help!

yxzhangLAB commented 2 years ago

I also had same problem, there is no use for "netVisual_individual" and "computeCommunProb(cellchat1, type = "truncatedMean", trim = 0.1)". I'm wondering if this issue has since been resolved? Thank you!

sqjin commented 2 years ago

@yxzhangLAB iF IT SHOWS Cannot find SEMA7 . Please input a correct name!, it means that this signaling is not identified by cellchat.

Wangyananblink commented 2 years ago

@yxzhangLAB iF IT SHOWS Cannot find SEMA7 . Please input a correct name!, it means that this signaling is not identified by cellchat.

Sorry to disturb you. I am still puzzled about "not identified by cellchat". If it is not identified by cellchat, what algorithm identifies this signaling pathway?

sqjin commented 2 years ago

@Wangyananblink You may can consider to adjust the method for calculating the average gene expression per cell group when running computeCommunProb. e.g, USER can set type = "truncatedMean" and trim = 0.1. Please check tutorial for details. This new setting allows you to identify signaling with weak signaling.