runLASSOregression: Main function for running LASSO regression to calculate...

In Huber-group-EMBL/DepInfeR: Inferring tumor-specific cancer dependencies through integrating ex-vivo drug response assays and drug-protein profiling

Main function for running LASSO regression to calculate protein dependence coefficient

Description

This function performs multivariate linear regression with LASSO penalty and cross-validation to infer per-sample protein dependence coefficients. Please refer to the package vignette for more detailed information about this function. For the mathematical model behind this function, please refer to our preprint on bioRxiv: https://doi.org/10.1101/2022.01.11.475864.

Usage

runLASSORegression(
  TargetMatrix,
  ResponseMatrix,
  repeats = 100,
  BPPARAM = bpparam()
)

Arguments

TargetMatrix	Pre-processed drug-protein affinity matrix. Each row is a drug and each column is a protein target.
ResponseMatrix	Pre-processed drug-response viability matrix. Each row is a drug and each column is a sample (cell line or tumor sample).
repeats	A integer variable specifying the number of regression repeats. The default value is 100. A higher number can result in better stability but also takes longer time.
BPPARAM	The BiocParallel back-end. If not specified, the default MulticoreParam will be used.

Value

A list of 6 elements:

coefMat	A matrix containing the inferred protein dependency coefficient matrix. Each row is a protein and each column is a sample.
freqMat	A matrix containing the selection frequency a protein (row) for each sample (column).
lambdaList	A vector containing the lambda values selected for all repeated runs.
varExplain.all	A vector containing the R2 (variance explained) values for all repeated runs.
inputX	A copy of the input drug-protein affinity matrix.
inputY	A copy of the input drug-repsonse viability matrix.

Examples

data(responseInput) #load drug response matrix
data(targetInput) #load drug-target affinity matrix
runLASSORegression(TargetMatrix = targetInput, ResponseMatrix = responseInput, repeats = 5)

Huber-group-EMBL/DepInfeR documentation built on April 7, 2023, 7:40 a.m.