makeContrast: Make contrast matrix

In statOmics/msqrob2: Robust statistical inference for quantitative LC-MS proteomics

Make contrast matrix

Description

Construct the contrast matrix corresponding to specified contrasts of a set of parameters.

Usage

makeContrast(contrasts, parameterNames)

Arguments

contrasts

character vector specifying contrasts, i.e. the linear combination of the modelparameters that equals to zero.

parameterNames

character vector specifying the model parameters that are involved in the contrasts, e.g if we model data of three conditions using a factor condition with three levels a, b and c then our model will have 3 mean parameters named (Intercept), conditionb and conditionc. Hence the log2 fold change between b and a is conditionb. Under the null hypothesis the log2 fold change equals 0. Which is to be encoded as "conditionb=0". If we would like to test for log2 fold change between condition c and b we assess if the log2 fold change conditionc-conditionb equals 0, encoded as "condtionb-conditionc=0".

Value

A numeric contrast matrix with rownames that equal the model parameters that are involved in the contrasts

Examples

makeContrast(c("conditionb = 0"),
    parameterNames = c(
        "(Intercept)",
        "conditionb",
        "conditionc"
    )
)
makeContrast(c("conditionc=0"),
    parameterNames = c("conditionc")
)
makeContrast(c(
    "conditionb=0",
    "conditionc=0",
    "conditionc-conditionb=0"
),
parameterNames = c(
    "conditionb",
    "conditionc"
)
)

statOmics/msqrob2 documentation built on March 13, 2023, 6:24 a.m.