vignettes/guide.R

In ncchung/SeuratAddon: Unsupervised evaluation of cell identities in Seurat

## ----setup, include = FALSE----------------------------------------------
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  eval = FALSE
)

## ------------------------------------------------------------------------
#  library(Seurat)
#  library(SeuratAddon)
#  library(jackstraw)
#  # Load the PBMC dataset, change the directory if necessary
#  pbmc.data <- Read10X(data.dir = "data/pbmc3k/filtered_gene_bc_matrices/hg19/")
#  
#  # Initialize the Seurat object with the raw (non-normalized data)
#  pbmc <- CreateSeuratObject(raw.data = pbmc.data, min.cells = 3, min.genes = 200, project = "10X_PBMC")

## ------------------------------------------------------------------------
#  mito.genes <- grep(pattern = "^MT-", x = rownames(x = pbmc@data), value = TRUE)
#  percent.mito <- Matrix::colSums(pbmc@raw.data[mito.genes, ])/Matrix::colSums(pbmc@raw.data)
#  
#  pbmc <- AddMetaData(object = pbmc, metadata = percent.mito, col.name = "percent.mito")
#  
#  pbmc <- FilterCells(object = pbmc, subset.names = c("nGene", "percent.mito"),
#      low.thresholds = c(200, -Inf), high.thresholds = c(2500, 0.05))

## ------------------------------------------------------------------------
#  pbmc <- NormalizeData(object = pbmc, normalization.method = "LogNormalize", scale.factor = 10000)

## ------------------------------------------------------------------------
#  pbmc <- FindVariableGenes(object = pbmc, mean.function = ExpMean, dispersion.function = LogVMR, do.plot=FALSE, x.low.cutoff = 0.0125, x.high.cutoff = 3, y.cutoff = 0.5)

## ------------------------------------------------------------------------
#  pbmc <- ScaleData(object = pbmc, vars.to.regress = c("nUMI", "percent.mito"))

## ------------------------------------------------------------------------
#  pbmc <- RunPCA(object = pbmc, pc.genes = pbmc@var.genes, do.print = FALSE)

## ------------------------------------------------------------------------
#  # ClusterCellsKmeans is a modified version of Seurat::DoKMeans
#  pbmc <- ClusterCellsKmeans(pbmc, k.cells = 3)

## ------------------------------------------------------------------------
#  # Use optional arguments, e.g.,batch_size=100, num_init=5, max_iters=100
#  pbmc <- ClusterCellsKmeans(pbmc, k.cells =  3, minibatch = TRUE)

## ------------------------------------------------------------------------
#  pbmc <- EvaluateIdent(pbmc, clustering = "kmeans")

## ------------------------------------------------------------------------
#  pbmc <- EvaluateIdent(pbmc, clustering = "KMeans", prob.use = "p.adj", p.adjust.methods = "BH")

## ------------------------------------------------------------------------
#  pbmc <- EvaluateIdent(pbmc, clustering = "MiniBatchKMeans", ...)

## ------------------------------------------------------------------------
#  pbmc_jackstraw <- EvaluateIdent(pbmc, clustering = "KMeans", return.jackstraw = TRUE)

## ------------------------------------------------------------------------
#  #### Run Non-linear dimensional reduction (tSNE)
#  pbmc <- RunTSNE(object = pbmc, dims.use = 1:10, do.fast = TRUE)
#  # note that you can set do.label=T to help label individual clusters
#  tsne_all <- TSNEPlot(object = pbmc, do.label=T)

## ------------------------------------------------------------------------
#  ggplot(pbmc@meta.data["ident_prob"]) + geom_histogram(aes(x=ident_prob))

## ------------------------------------------------------------------------
#  tsne_piphard <- TSNEPlot2(object = pbmc, do.label=T, ident.threshold=0.9)

## ------------------------------------------------------------------------
#  tsne_pipsoft <- TSNEPlot2(object = pbmc, do.label=T, ident.threshold="soft")