getDEG: Get differentially expressed genes

In NMikolajewicz/scMiko: scRNAseq analysis functions. Developed using Seurat framework.

Get differentially expressed genes

Description

Differential expression analysis wrapper for presto::wilcoxauc(). Option to return differentially-expressed gene list (return.list = T) or statistics only (return.list = F)

Usage

getDEG(
  object,
  assay = DefaultAssay(object),
  data = "data",
  group_by = "seurat_clusters",
  auc.thresh = 0.6,
  fdr.thresh = 0.01,
  logFC.thresh = NA,
  pct.dif.thresh = NA,
  pct.in.thresh = NA,
  pct.out.thresh = NA,
  return.list = F,
  return.all = F,
  sig.figs = NA,
  verbose = T
)

Arguments

object	Seurat object
assay	assay. Default "SCT".
data	data slot. Default "data".
group_by	Name of grouping variable (must be present in object's meta.data)
auc.thresh	AUC threshold. Default = 0.6.
fdr.thresh	FDR threshold. Default = 0.01.
logFC.thresh	logFC threshold. Default = NA.
pct.dif.thresh	Difference in expression percentage. Default = NA.
pct.in.thresh	Expression percentages exceeding this threshold are retained. Default is NA.
pct.out.thresh	Expression percentages below this threshold are retained Default is NA.
return.list	If TRUE, return list of differentially expressed genes. If FALSE, returns table with statistics from differential expression analysis.
return.all	If TRUE, all thresholding filters are ignored, and all results are returned.
sig.figs	If specified and return.list = F, rounds statistics to specified significant figure (recommended: 3). Default is NA.
verbose	Print progress. Default is TRUE.

Value

data.frame or list. Statistics in data.frame output include:

• avgExpr: Average expression for group

• logFC: Log fold change

• statistics: Test statistics

• auc: Area under curve

• pval: p-value

• padj: adjusted p-value

• pct_in: percentage of expressing cells within group

• pct_out: percentage of expression cells outside of group

• pct.dif: difference between pct_in and pct_out

• sensitivity: pct_in/100

• specificity: (100-pct_out)/100

• PPV: positive predictive value

• NPV: negative predictive value

• ss: sensitivity x specificity

Author(s)

Nicholas Mikolajewicz

See Also

wilcoxauc

NMikolajewicz/scMiko documentation built on June 28, 2023, 1:41 p.m.