lasso: Create an adjacency matrix based on LASSO

In MetNet: Inferring metabolic networks from untargeted high-resolution mass spectrometry data

Description

'lasso' infers a adjacency matrix using LASSO using the 'stabsel.matrix' function from the 'stabs' package. 'lasso' extracts the predictors from the function 'stabsel.matrix' and writes the coefficients to an adjacency matrix.

Usage

lasso(x, parallel = FALSE, ...)

Arguments

x	matrix, where columns are the samples and the rows are features (metabolites), cell entries are intensity values
parallel	logical, should computation be parallelized? If 'parallel = TRUE' the 'bplapply' will be applied if 'parallel = FALSE' the 'lapply' function will be applied.
...	parameters passed to 'stabsel.matrix'

Details

For use of the parameters used in the 'stabsel.matrix' function, refer to '?stabs::stabsel.matrix'.

Value

matrix, matrix with edges inferred from LASSO algorithm 'stabsel.matrix'

Author(s)

Thomas Naake, [e-mail]

Examples

data("x_test", package = "MetNet")
x <- x_test[1:10, 3:ncol(x_test)]
x <- as.matrix(x)
x_z <- t(apply(x, 1, function(y) (y - mean(y)) / sd(y)))
## Not run: lasso(x_z, PFER = 0.95, cutoff = 0.95)

MetNet documentation built on Nov. 8, 2020, 7:34 p.m.