DJEanalyze: DJEanalyze: Test for Differential Junction Usage

In MauerLab/DJExpress: An integrated application for differential splicing analysis and visualization

Description

Normalize junction expression and performs differential junction usage analysis using limma::diffSplice methods. Returns output object for DJEplotSplice(), DJEvsTrait() and JCNAprepare() functions.

Usage

DJEanalyze(
  prepare.out,
  Group1,
  junct.annot = NULL,
  normalize.method = "none",
  calcNormFactors = FALSE,
  calMethod = "none",
  plot = FALSE,
  FDR = 0.05,
  logFC = 2,
  level = 0.9,
  topJunct = NULL,
  targetGene = NULL,
  legend.position = "right",
  volc.axes.face = "plain",
  volc.axes.col = "black",
  volc.axes.size = 11,
  volc.axes.angle = 0,
  scale_color_manual = c("#005BA2", "#D44F4F", "#B3B4B5")
)

Arguments

prepare.out	output object from DJEprepare()
Group1	vector or factor specifying basic control sample names
junct.annot	gene-annotated junction IDs
normalize.method	limma voom's normalization method to be applied to the logCPM values (default "none")
calcNormFactors	logical, should edgeR's library Size Normalization be calculated?
calMethod	Normalization method to be used. Options are: "TMM","RLE","upperquartile" or "none" (default "none")
plot	logical, should median/variance trend plot be displayed?
FDR	numeric: adjusted p-value cutoff in the linear model fit to define significance
logFC	numeric: log2-fold-change value cutoff corresponding to the effect or contrast in the linear model fit to define significance
level	numeric: confidence interval for Linear Regression of absolute vs gene-wise logFC in junction expression
topJunct	numeric: number of top significant junctions that should be labeled in volcano plot
targetGene	vector or factor specifying the genes whose significant junctions should be labeled in volcano plot
legend.position	character indicating the position of labels in volcano plot. Allowed values are: “left”, “top”, “right”, “bottom”. Default "right".
volc.axes.face	character indicating the font face of axes tick labels in volcano plot. Allowed values are: "plain", "italic", "bold", "bold.italic"
volc.axes.col	character indicating the color of axes tick labels in volcano plot
volc.axes.size	numeric indicating the size of axes tick labels in volcano plot
volc.axes.angle	numeric indicating the angle of axes tick labels in volcano plot
scale_color_manual	vector of aesthetic values (colors) to indicate significance of data values in volcano plot (default are #005BA2 for downregulated, #D44F4F for upregulated and #B3B4B5 for non-significant junctions)

Value

A list of objects containing: v.norm: log-cpm values output by limma:voom ex.norm: Differential Junction Expression analysis output dje.out: Annotated ex.norm data set with additional information, including basic statistics (e.g. median, zero counts, etc) and DJE group for each junction. dje.sig: Significant hits in dje.out based on FDR and logFC cutoffs. volcano.plot: Volcano plot of differential junction expression logFC.plot.junctions: Table with junctions shown in logFC.plot. They are defined as junctions passing FDR cutoff for differential usage as well as differential expression. logFC.plot: Regression plot of Absolute logFC ~ Relative logFC for differentially used (compared to average junction expression in the gene) and differentially expressed junctions (basal vs tested sample group)

Details

This wrapper function receives DJEprepare() output object and performs junction expression normalization and differential junction usage analysis. The differential analysis is an implementation of limma::diffSplice() method. See limma::diffSplice documentation for details.

See Also

limma

Examples

DJEprep <- system.file("extdata", "DJEprep.rds", package = "DJExpress")
prep.out <- readRDS(DJEprep)
Group1 <- colnames(prep.out$JunctExprfilt)[grep("GTEx", colnames(prep.out$JunctExprfilt))]
anlz.out <- DJExpress::DJEanalyze(prep.out, Group1)

MauerLab/DJExpress documentation built on Jan. 1, 2022, 12:57 p.m.