multiwcomp.MiRKAT: Taxa importance evaluation through weighted MiRKAT

In UVic-omics/CoDA-wKMR:

Description

For different weight values, computes for each variable the p-value using 'wcomp.MiRKAT' function and implements a linear regression model giving back its slope and the p-value for the slope.

For different weight values, computes for each variable the p-value using 'wcomp.MiRKAT' function and implements a linear regression model giving back its slope and the p-value for the slope.

Usage

multiwcomp.MiRKAT(x, y, w.seq = c(0.1, 0.4, 0.7, 1, 2, 3, 4, 5), cov = NULL,
  zero.rep = T, pseudocount = 1)

multiwcomp.MiRKAT(x, y, w.seq = c(0.1, 0.4, 0.7, 1, 2, 3, 4, 5), cov = NULL,
  zero.rep = T, pseudocount = 1)

Arguments

x	matrix with the composition for each individual
y	response variable: dichotomous or continuous
w.seq	vector of weights to consider in the analysis
cov	covariates to hace into account as a possible confounding factors
zero.rep	logical value indicating if cmultRepl() frunction from zCompositions should be used in order to replace null values.
pseudocount	a value to add to the whole matrix in order to avoid zeros. Only if zero.rep == FALSE.
x	matrix with the composition for each individual
y	response variable: dichotomous or continuous
w.seq	vector of weights to consider in the analysis
cov	covariates to hace into account as a possible confounding factors
zero.rep	logical value indicating if cmultRepl() frunction from zCompositions should be used in order to replace null values.
pseudocount	a value to add to the whole matrix in order to avoid zeros. Only if zero.rep == FALSE.

Value

a data.frame with the slope and its p-value for the linear regression model of wcomp.MiRKAT p-values considering the given weights.

a data.frame with the slope and its p-value for the linear regression model of wcomp.MiRKAT p-values considering the given weights.

Examples

# Matrix of samples
 x <- matrix(runif(100), nrow = 10)
# Response vector
 y <- rep(c(0,1),5)

# Run multiwcomp.MiRKAT function
 multiwcomp.MiRKAT(x,y)

# Matrix of samples
 x <- matrix(runif(100), nrow = 10)
# Response vector
 y <- rep(c(0,1),5)

# Run multiwcomp.MiRKAT function
 multiwcomp.MiRKAT(x,y)

UVic-omics/CoDA-wKMR documentation built on Oct. 31, 2019, 12:56 a.m.