DiffCorPlus: Implement differential correlation analysis based on...

In SCBIT-YYLab/DysRegSig: Gene dysregulation analysis and construction of mechanistic signature in cancer

Description

Identify regulations with differential correlation, differential expression of target, and the consistency between differential correlation and differential expression.

Usage

DiffCorPlus(exp.1, exp.2, tf2tar, 
            de.genes = NULL, de.pval = NULL, de.qval = NULL, de.logFC = NULL, 
            cor.method = "pearson", p.adj = "BH", verbose = TRUE)

Arguments

exp.1	Expression matrix of a special condition. Columns correspond to genes, rows correspond to experiments. The matrix is expected to be already normalized.
exp.2	Expression matrix of an another special condition. Columns correspond to genes, rows correspond to experiments. The matrix is expected to be already normalized.
tf2tar	The prior reference GRN containing TF-target relationships.
de.genes	A dataframe for differential expression genes. If no de.genes offered, DysReg uses the default method limma to implement differential expression analysis. If de.genes offered, The dataframe must include three columns, "GeneSymbol", "high.condition", "de.logFC". "high.condition" means which condition represents high expression level. "de.logFC" is the output logFC from differential expression analysis.
de.pval	The cutoff of pval for filtering differential expression genes. If you don't use this parameter to filter differential expression genes, this parameter could be set as NULL. If you use this parameter to filter differential expression genes, this parameter could be set as a special number, such as 0.05.
de.qval	The cutoff of qval used for filtering differential expression genes. If you don't use this parameter to filter differential expression genes, this parameter could be set as NULL. If you use this parameter to filter differential expression genes, this parameter could be set as a special number, such as 0.05.
de.logFC	The cutoff of absolute logFC used for filtering differential expression genes. If you don't use this parameter to filter differential expression genes, this parameter could be set as NULL. If you use this parameter to filter differential expression genes, this parameter could be set as a special number, such as 0.5. This parameter could be used by combining with de.pval or de.qval.
cor.method	Which correlation coefficient (or covariance) is to be computed. One of "pearson" (default) or "spearman", can be abbreviated.
p.adj	Correction method for p value adjust.
verbose	A logical value indicating whether display the computing progress.

Value

The identified regulations.

See Also

DiffCor

Examples

data(ExpData)
data(tf2tar)
data(ClinData)

group.1 <- ClinData$sample[which(ClinData$binaryResponse == 'CR/PR')]
exp.1 <- ExpData[,colnames(ExpData) %in% group.1]

group.2 <- ClinData$sample[which(ClinData$binaryResponse == 'SD/PD')]
exp.2 <- ExpData[,colnames(ExpData) %in% group.2]


## implement differential correlation analysis
diffcor.p.res <- DiffCorPlus(exp.1 = exp.1,exp.2 = exp.2, tf2tar, 
                             de.genes = NULL, de.pval = 0.05, 
                             cor.method = 'pearson', p.adj = 'BH')

## set cutoff
diffcor.p.res <- subset(diffcor.p.res,p.val < 0.05)
head(diffcor.p.res)

SCBIT-YYLab/DysRegSig documentation built on July 19, 2021, 4:38 a.m.