CalcGlmnetPvals: Calculate regression coefficients and p-values via...

In yanwu2014/perturbLM: A lightweight R package that uses linear models to analyze the effects of chemical/genetic perturbations, conditions, and disease states on single gene expression.

Calculate regression coefficients and p-values via permutation testing

Description

Calculate regression coefficients and p-values via permutation testing

Usage

CalcGlmnetPvals(
  design.matrix,
  metadata,
  y,
  alpha,
  lambda.use,
  family,
  ctrl = NULL,
  n.rand = 20,
  n.cores = 16,
  n.bins = 10,
  use.quantiles = T,
  output.name = "cf"
)

Arguments

design.matrix	Design matrix for regression
metadata	Additional covariates to take into account (will not be permuted)
y	Linear model response
alpha	Elasticnet ratio: (0 is fully L2, 1 fully is L1)
lambda.use	Coefficient regularization parameter
family	Regression family to use
ctrl	Control variable to compare against (optional)
n.rand	Number of permutations for calculating coefficient significance
n.cores	Number of cores to use
n.bins	If binning genes, number of bins to use
use.quantiles	If binning genes, whether or not to bin by quantile
output.name	Column name of regression coefficient

Value

Returns a dataframe with Gene, Perturbation, log-FC, p-value

yanwu2014/perturbLM documentation built on Aug. 24, 2023, 2:28 p.m.