updateCellChatDB
updateCellChatDBupdateCellChatDBCellChat v2 provides the function updateCellChatDB to
allow users to update CellChatDB by integrating new ligand-receptor
pairs from other resources or utilizing a custom ligand-receptor
interaction database. This function can be used as follows:
updateCellChatDB(
db,
gene_info = NULL,
other_info = NULL,
merged = FALSE,
species_target = NULL
)The minimum input of this function is a data frame with two columns
named as ligand and receptor, and other inputs
include classified pathway, complex, cofactor and gene information.
Users can check the documentation of this function for more details of
the required input data by typing help(updateCellChatDB) in
RStudio.
Here we describe how to integrate other resources by taking another database CellTalkDB as an example. After downloading the CellTalkDB from https://github.com/ZJUFanLab/CellTalkDB, follow the four steps as follows.
library(CellChat)
options(stringsAsFactors = FALSE)The first input db is a data frame with as least two
columns named as ligand and receptor. We
highly suggest users to provide a column of pathway information named
‘pathway_name’ associated with each L-R pair. Other optional columns
include ‘interaction_name’ and ‘interaction_name_2’. The default columns
of CellChatDB can be checked via
‘colnames(CellChatDB.human$interaction)’.
Warning: If no pathway information of each L-R pair is
provided, all pathway-level analysis from CellChat cannot be used, such
as computeCommunProb!
The second input gene_info is a data frame with at least
one column named as ‘Symbol’. “When setting gene_info = NULL, the input
‘species_target’ should be provided: either ‘human’ or ‘mouse’.
The third input other_info is a list consisting of other
information including a dataframe named as ‘complex’ and a dataframe
named as ‘cofactor’. This additional information is not necessary. If
other_info is provided, the ‘complex’ and ‘cofactor’ are dataframes with
defined rownames.
# Load the dataframe consisting of customized ligand-receptor pairs
db.user <- readRDS("./CellTalkDB-master/database/human_lr_pair.rds")
# Load the dataframe consisting of gene information (optional)
gene_info <- readRDS("./CellTalkDB-master/data/human_gene_info.rds")Please check above for the basic requirement of the input files. In addition, users can check the detailed database information in CellChatDB.
The main ligand-receptor interaction information is stored in
CellChatDB$interaction. Columns of this dataframe includes
information of ligands, receptors and co-factors.
CellChatDB$geneInfo contains all gene information in
mouse or other species and it should have a column named ‘Symbol’, which
is not necessary to be changed when updating CellChatDB.
Optional information is stored in CellChatDB$complex
and CellChatDB$cofactor, which include any ligand complex,
receptor complex and cofactors. Users need to make sure that
user-defined complex/cofactor names are the same in
CellChatDB$interaction.
# Modify the colnames because of incompatible with `colnames(CellChatDB.human$interaction)`
colnames(db.user) <- plyr::mapvalues(colnames(db.user), from = c("ligand_gene_symbol","receptor_gene_symbol","lr_pair"), to = c("ligand","receptor","interaction_name"), warn_missing = TRUE)updateCellChatDB# Create a new database by typing one of the following commands
# Use user-provided gene information
db.new <- updateCellChatDB(db = db.user, gene_info = gene_info)
# Use built-in gene information of CellChatDB
db.new <- updateCellChatDB(db = db.user, gene_info = NULL, species_target = "human")
# Alternatively, users can integrate the customized L-R pairs into the built-in CellChatDB
db.new <- updateCellChatDB(db = db.user, merged = TRUE, species_target = "human")
Warning: Becuase no pathway information of each L-R pair is
provided in db.user here, all pathway-level analysis from
CellChat cannot be used, such as computeCommunProb!
Re-build CellChat package by updating the database as follows
# Users can now use this new database in CellChat analysis
cellchat@DB <- db.new
# Users can save the new database for future use
save(db.new, file = "CellChatDB.human_user.rda")
# Users can also re-build CellChatDB in CellChat package
setwd("/Users/$USERS/Downloads/CellChat-master") # This is the folder of CellChat package downloaded from Github
CellChatDB.mouse <- CellChatDB
usethis::use_data(CellChatDB.mouse, overwrite = TRUE)
# If working on a human dataset, do following:
# CellChatDB.human <- CellChatDB
# usethis::use_data(CellChatDB.human, overwrite = TRUE)Since cellphonedb v5, cellphonedb also introduces signalling directionality (ligand is partner A, receptor partner B) and classification of signaling pathways, which makes it easiler to make full use of CellChat’s versatile analysis by taking CellPhoneDB as an input. Here we describe how to integrate L-R interaction resources from CellPhoneDB. After downloading the cellphonedb v5 from https://github.com/ventolab/cellphonedb-data, run the following codes.
library(CellChat)
options(stringsAsFactors = FALSE)## load the database files
interaction_input <- read.csv(file = './otherDB/cellphonedb-data-master/data/interaction_input.csv')
complex_input <- read.csv(file = './otherDB/cellphonedb-data-master/data/complex_input.csv', row.names = 1)
geneInfo <- read.csv(file = './otherDB/cellphonedb-data-master/data/gene_input.csv')
geneInfo$Symbol <- geneInfo$hgnc_symbol
geneInfo <- select(geneInfo, -c("ensembl"))
geneInfo <- unique(geneInfo)## prepare interaction_input file
# get the ligand information
idx_partnerA <- match(interaction_input$partner_a, geneInfo$uniprot)
idx.use <- !is.na(idx_partnerA)
interaction_input$ligand <- interaction_input$partner_a
interaction_input$ligand[idx.use] <- geneInfo$hgnc_symbol[idx_partnerA[idx.use]]
# get the receptor information
idx_partnerB <- match(interaction_input$partner_b, geneInfo$uniprot)
idx.use <- !is.na(idx_partnerB)
interaction_input$receptor <- interaction_input$partner_b
interaction_input$receptor[idx.use] <- geneInfo$hgnc_symbol[idx_partnerB[idx.use]]
# get other information
interaction_input$interaction_name <- interaction_input$interactors
interaction_input$interaction_name_2 <- interaction_input$interaction_name
interaction_input$pathway_name <- interaction_input$classification
interaction_input$pathway_name <- gsub(".*by ", "", interaction_input$pathway_name)
interaction_input$annotation <- interaction_input$directionality
interaction_input <- select(interaction_input, -c("partner_a","partner_b","protein_name_a","protein_name_b","interactors","classification","directionality"))
## prepare complex_input file
complexsubunits <- dplyr::select(complex_input, starts_with("uniprot"))
for (i in 1:ncol(complexsubunits)) {
idx_complex <- match(complex_input[,paste0("uniprot_",i)], geneInfo$uniprot)
idx.use <- !is.na(idx_complex)
complex_input[idx.use,paste0("uniprot_",i)] <- geneInfo$hgnc_symbol[idx_complex[idx.use]]
}
colnames(complex_input)[1:ncol(complexsubunits)] <- paste0("subunit_",seq_len(ncol(complexsubunits)))
complex_input <- dplyr::select(complex_input, starts_with("subunit"))
## prepare other information
other_info <- list(complex = complex_input)updateCellChatDBdb.new <- updateCellChatDB(db = interaction_input, gene_info = geneInfo, other_info = other_info, trim.pathway = T)
Re-build CellChat package by updating the database as follows
# Users can now use this new database in CellChat analysis
cellchat@DB <- db.new
# Users can save the new database for future use
save(db.new, file = "CellChatDB.human_user.rda")
# Users can also re-build CellChatDB in CellChat package
setwd("/Users/$USERS/Downloads/CellChat-master") # This is the folder of CellChat package downloaded from Github
CellChatDB.human <- CellChatDB
usethis::use_data(CellChatDB.human, overwrite = TRUE)
Here we outline the steps to update CellChatDB by manually adding user-defined ligand-receptor pairs. To do so, the format of the users’ lists must be compatible with the input files of CellChatDB.
library(CellChat)
options(stringsAsFactors = FALSE)Extract the database information in CellChatDB and then save them in a local computer, including four files: ‘geneInfo.csv’, ‘interaction_input_CellChat.csv’, ‘complex_input_CellChat.csv’, ‘and cofactor_input_CellChat.csv’. Users can do it by running the following codes in Rstudio:
CellChatDB <- CellChatDB.mouse # set CellChatDB <- CellChatDB.human if working on the human dataset
interaction_input <- CellChatDB$interaction
complex_input <- CellChatDB$complex
cofactor_input <- CellChatDB$cofactor
geneInfo <- CellChatDB$geneInfo
write.csv(interaction_input, file = "interaction_input_CellChatDB.csv")
write.csv(complex_input, file = "complex_input_CellChatDB.csv")
write.csv(cofactor_input, file = "cofactor_input_CellChatDB.csv")
write.csv(geneInfo, file = "geneInfo_input_CellChatDB.csv")
Update the above four .csv files by adding users’ curated ligand-receptor pairs.
The main file is ‘interaction_input_CellChatDB.csv’. Users can first update the ligands, receptors and co-factors in the corresponding columns in ‘interaction_input_CellChatDB.csv’.
Users can then update “complex_input_CellChatDB.csv” and “cofactor_input_CellChatDB.csv” if any ligand complex, receptor complex and cofactors are updated. Users need to make sure that user-defined complex/cofactor names are the same in ‘interaction_input_CellChatDB.csv’ and “complex_input_CellChatDB.csv”, ‘interaction_input_CellChatDB.csv’ and ” cofactor_input_CellChatDB.csv”.
“geneInfo_input_CellChatDB.csv” contains all gene information in mouse and it should have a column named ‘Symbol’, which does not need to be changed when updating CellChatDB.
Update CellChatDB once updating the four .csv files. Users can do it by running the following codes in Rstudio:
options(stringsAsFactors = FALSE)
interaction_input <- read.csv(file = 'interaction_input_CellChatDB.csv', row.names = 1)
complex_input <- read.csv(file = 'complex_input_CellChatDB.csv', row.names = 1)
cofactor_input <- read.csv(file = 'cofactor_input_CellChatDB.csv', row.names = 1)
geneInfo <- read.csv(file = ' geneInfo_input_CellChatDB.csv', row.names = 1)
CellChatDB <- list()
CellChatDB$interaction <- interaction_input
CellChatDB$complex <- complex_input
CellChatDB$cofactor <- cofactor_input
CellChatDB$geneInfo <- geneInfo
Re-build CellChat package by updating the database as follows
setwd("/Users/$USERS/Downloads/CellChat-master") # This is the folder of CellChat package downloaded from Github
CellChatDB.mouse <- CellChatDB
usethis::use_data(CellChatDB.mouse, overwrite = TRUE)
# If working on a human dataset, do following:
# CellChatDB.human <- CellChatDB
# usethis::use_data(CellChatDB.human, overwrite = TRUE)