DJEvsTrait: DJEvsTrait: Junction relationship to external trait

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

Description

Correlation between junction expression and sample traits

Usage

DJEvsTrait(
  analize.out,
  Group1,
  traitData,
  nThreads = 2,
  select.junctions = NULL,
  coeff = 0.3,
  pvalue = 0.05,
  plot.heatmap = FALSE,
  test.type = c("Correlation", "Regression", "both"),
  cor.method = c("bicor", "pearson", "kendall", "spearman"),
  useModel = c(modelANOVA, modelLINEAR, modelLINEAR_CROSS),
  covariates = NULL
)

Arguments

analize.out	output object from DJEanalyze()
Group1	vector or factor specifying basic control sample names
traitData	a numeric vector or a matrix of external sample traits
nThreads	numeric: number of threads to allow.
select.junctions	a vector of junction IDs used to generate SpliceRadar plots when test.type = "Correlation". Junction ID vector length should be < 5
coeff	numeric: correlation coefficient cutoff for significance
pvalue	numeric: correlation p-value cutoff for significance
plot.heatmap	logical, should junction vs trait correlation heatmap plot be saved as output?. Default is FALSE
test.type	a character string specifying the type of junction-trait association test to use. Options are "Correlation", "Regression" or "both".
cor.method	a character string specifying the method to be used for correlation as in WGCNA. Options are "bicor", "pearson", "kendall" or "spearman".
useModel	integer specifying the model type to be used for junction-trait association as in MatrixEQTL. Options are modelLINEAR, modelANOVA or modelLINEAR_CROSS. Valid when test.type = "Regression" or "both".
covariates	a numeric vector or a matrix of external sample covariates to be used in regression models. Valid when test.type = "Regression" or "both".

Value

When "Correlation" is selected as test.type, DJEvsTrait returns the matrices with correlation coefficients and associated P-values for each junction-trait pair. sig.cor contains top significant associations between junctions in select.junctions and traits for SpliceRadar plotting. When selected "Regression" as test.type, DJEvsTrait returns a data frame with significant junction-trait associations based on selected model in useModel. "both" test.type returns correlation and regression outputs.

See Also

cor

Matrix_eQTL_main

Examples

data(DJEanlz)
SF <- system.file("extdata", "SF.expr.rds", package = "DJExpress")
SF.exp <- readRDS(SF)
Group1 <- colnames(DJEanlz$v.norm$E)[grep("SRR", colnames(DJEanlz$v.norm$E))]
DT.out <- DJEvsTrait(analize.out = DJEanlz, Group1 = Group1,traitData = SF.exp,
coeff = 0.2,select.junctions = c("chr1:225688773:225695652:2",
"chr1:225692756:225695652:2","chr1:225688773:225692692:2"),
test.type = "Correlation", cor.method = "bicor")

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