iteration_over_clusters_parallelized_wilcox_boostrap: Marker gene selection with parallelization using wilcox...

In crhisto/SCDC: Bulk RNA-seq deconvolution by scRNA-seq with multi-reference datasets

Description

Marker gene selection with parallelization using wilcox method from the Seurat library

Usage

iteration_over_clusters_parallelized_wilcox_boostrap(
  sc.eset,
  ct.group,
  core_number = NULL,
  LFC.lim = 0.2,
  iteration.minimun_number_markers = 28,
  iteration.use_maximum = TRUE,
  iteration.maximo_genes = 35,
  iteration.use_final_foldchange = FALSE,
  bootstrap.sample_size = NULL,
  bootstrap.number = NULL,
  ...
)

Arguments

sc.eset	ExpressionSet object for single cells
ct.group	List of clusters that will be analyzed
core_number	Number of cores that will be used for the process.
LFC.lim	Fa threshold of log fold change when selecting genes as input to perform Wilcoxon's test.
iteration.use_final_foldchange	TRUE/FALSE. If at the end the cluster has zero genes if this parameter is true, the boostraping is going to be calculated over the foldchange with <0.05, not with zero.
minimun_number_markers	If in the first step with the wilcox text there are at least this genes, the process is not going to use outlier detection with dbscan
use_maximum	TRUE/FALSE. If the process selects just a fixed number of genes.
maximo_genes	If we want to limitate the maximum number of marker genes at the end of the process.

Value

List with the marker genes for all selected clusters

crhisto/SCDC documentation built on Dec. 19, 2021, 6:19 p.m.