SCdcv: Single-cell level deconvolution

In LeonSong1995/mthodtest: Estimate Cell-type Proportions of Bulk RNA-Seq Data with Linear Mixed Model

Description

Single-cell level deconvolution

Usage

SCdcv(
  bulk,
  sce,
  select.ct,
  ncpu = NULL,
  smoothing = TRUE,
  gene = NULL,
  op = 0.2,
  maxgene = 1000,
  nbin = 0.1,
  SF = 1000
)

Arguments

bulk	A matrix containing bulk RNA-Seq data. Each row corresponds to a certain gene and each column to a certain sample.
sce	A 'Seurat' object containing the single-cell RNA-Seq data. Meta data of the 'Seurat' object must includes 'cellType' and 'space'.
select.ct	A character (or character vector) of the names of the target cell-types. The default value is NULL. With default value, all cell-types in the single-cell data will be used.
ncpu	The number of CPU cores to be used.
smoothing	A Boolean variable to determine whether to smooth the BLUP along cell-space or not. The default value is TRUE.
gene	A character vector of the gene names to use as signature for the deconvolution.The default value is NULL. With default, MLM will select genes to differentiate cells in the target cell-type with ANOVA.
op	A numeric variable to determine the overlapping between cell-clusters. The default value is 0.2.
maxgene	A numeric variable to determine the maximum number of genes to be selected. The default value is 1e+3.
nbin	A numeric variable to determine the number of cell-clusters. The default value is 0.2.
SF	Scaling factor. The default value is 1e+3. #Gene selecting

Value

A list with cell_density-cell_name matrix:

Examples

library(MLM)
SCdcv(bulk = example.bulk,sce = example.sce,select.ct = 'alpha')

LeonSong1995/mthodtest documentation built on Jan. 1, 2021, 1:56 p.m.