Add seed to netEmbedding function

[ycl6]

The PR adds a seed paramter to the netEmbedding function to allow generating reproducible UMAPs, as requested in #155, #196 and #435. This works on both umap-learn and uwot methods. The default setting is seed = 42. One can use seed = NULL to reset and produce a different set of embeddings.

Below is an example running using the uwot method.

library(CellChat)

####################
# load data
####################
load("data_humanSkin_CellChat.rda")
data.input = data_humanSkin$data
meta = data_humanSkin$meta
cell.use = rownames(meta)[meta$condition == "LS"]
data.input = data.input[, cell.use]
meta = meta[cell.use, ]

####################
# Create a CellChat object
####################
cellchat <- createCellChat(object = data.input, meta = meta, group.by = "labels")
#> [1] "Create a CellChat object from a data matrix"
#> Set cell identities for the new CellChat object 
#> The cell groups used for CellChat analysis are  APOE+ FIB FBN1+ FIB COL11A1+ FIB Inflam. FIB cDC1 cDC2 LC Inflam. DC TC Inflam. TC CD40LG+ TC NKT
cellchat <- addMeta(cellchat, meta = meta)
cellchat <- setIdent(cellchat, ident.use = "labels")

####################
# Set the ligand-receptor interaction database
####################
CellChatDB <- CellChatDB.human
CellChatDB.use <- subsetDB(CellChatDB, search = "Secreted Signaling")
cellchat@DB <- CellChatDB.use

####################
# Preprocessing the expression data
####################
cellchat <- subsetData(cellchat)
cellchat <- identifyOverExpressedGenes(cellchat)
cellchat <- identifyOverExpressedInteractions(cellchat)

####################
# Inference of cell-cell communication network
####################
cellchat <- computeCommunProb(cellchat)
#> triMean is used for calculating the average gene expression per cell group. 
#> [1] ">>> Run CellChat on sc/snRNA-seq data <<< [2023-09-01 17:16:25]"
#> [1] ">>> CellChat inference is done. Parameter values are stored in `object@options$parameter` <<< [2023-09-01 17:17:47]"
cellchat <- computeCommunProbPathway(cellchat)
cellchat <- aggregateNet(cellchat)

####################
# Visulaise manifold of signaling networks
####################

# seed = 1
cellchat <- computeNetSimilarity(cellchat, type = "functional")
cellchat <- netEmbedding(cellchat, type = "functional", umap.method = "uwot", seed = 1)
#> Manifold learning of the signaling networks for a single dataset
cellchat <- netClustering(cellchat, type = "functional", do.parallel = FALSE)
#> Classification learning of the signaling networks for a single dataset
netVisual_embedding(cellchat, type = "functional", label.size = 3.5)

# re-run seed = 1
cellchat <- computeNetSimilarity(cellchat, type = "functional")
cellchat <- netEmbedding(cellchat, type = "functional", umap.method = "uwot", seed = 1)
#> Manifold learning of the signaling networks for a single dataset
cellchat <- netClustering(cellchat, type = "functional", do.parallel = FALSE)
#> Classification learning of the signaling networks for a single dataset
netVisual_embedding(cellchat, type = "functional", label.size = 3.5)

# seed = NULL
cellchat <- computeNetSimilarity(cellchat, type = "functional")
cellchat <- netEmbedding(cellchat, type = "functional", umap.method = "uwot", seed = NULL)
#> Manifold learning of the signaling networks for a single dataset
cellchat <- netClustering(cellchat, type = "functional", do.parallel = FALSE)
#> Classification learning of the signaling networks for a single dataset
netVisual_embedding(cellchat, type = "functional", label.size = 3.5)

# re-run seed = NULL
cellchat <- computeNetSimilarity(cellchat, type = "functional")
cellchat <- netEmbedding(cellchat, type = "functional", umap.method = "uwot", seed = NULL)
#> Manifold learning of the signaling networks for a single dataset
cellchat <- netClustering(cellchat, type = "functional", do.parallel = FALSE)
#> Classification learning of the signaling networks for a single dataset
netVisual_embedding(cellchat, type = "functional", label.size = 3.5)

^{Created on 2023-09-01 with reprex v2.0.2}