run.inte.metaClust: run the integration pipeline
In Japrin/scPip: pipeline for scRNA-seq data analysis

run.inte.metaClust

R Documentation

run the integration pipeline

Description

run the integration pipeline

Usage

run.inte.metaClust(
  exp.list.table,
  out.prefix,
  gene.exclude.file,
  nGene.common = 1500,
  nGene.specific = 0,
  ncores = 12,
  npc = 15,
  TH.gene.occ = 1,
  res.hi = 50,
  method.clustering = "louvain",
  cor.var = c("S.Score", "G2M.Score", "DIG.Score1"),
  use.harmony = F,
  contamination.vec = NULL,
  gene.informative.file = NULL
)

Arguments

`exp.list.table`	data.table; one line for a dataset
`out.prefix`	character; output prefix
`gene.exclude.file`	character; file contains the genes to be excluded
`nGene.common`	integer; number of common genes. (default: 1500)
`nGene.specific`	integer; number of dataset specific genes. (default: 0)
`ncores`	integer; number of CPU cores to use. (default: 12)
`npc`	integer; number of principal components to use. (default: 15)
`TH.gene.occ`	double; range from 0 to 1. genes present in >= TH.gene.occ datasets are used. (default: 1)
`res.hi`	integer; high resolution used for mini-clusters identification. (default: 50)
`method.clustering`	character; clustering method for mini-clusters identification. (default: "louvain")
`cor.var`	character vector; subset of c("S.Score","G2M.Score","DIG.Score1","ISG.Score1","score.MALAT1")
`use.harmony`	logical; use harmony to correct for batch effect (batches are defined in column batchV). (default: FALSE)
`contamination.vec`	character vector; cells to be excluded. (default: NULL)
`gene.informative.file`	character; file contains informative genes which are to be used for integration. column geneSymbol is required. (default: NULL)

Details

For each dataset, the function first identify mini-clusters, then calculate the average expressions of mini-clusters. The gene by mini-cluster expression data will pass the pipeline: PCA, harmony, UMAP/Clustering.