Calculate The Gene Drug Sensitivity — geneDrugSensitivityPBCorr • PharmacoGx

This version of the function uses a partial correlation instead of standardized linear models, for discrete predictive features Requires at least 3 observations per group.

Usage

geneDrugSensitivityPBCorr(
  x,
  type,
  batch,
  drugpheno,
  test = c("resampling", "analytic"),
  req_alpha = 0.05,
  nBoot = 1000,
  conf.level = 0.95,
  max_perm = getOption("PharmacoGx_Max_Perm", ceiling(1/req_alpha * 100)),
  verbose = FALSE
)

Arguments

x: A numeric vector of gene expression values
type: A vector of factors specifying the cell lines or type types
batch: A vector of factors specifying the batch
drugpheno: A numeric vector of drug sensitivity values (e.g., IC50 or AUC)
test: A character string indicating whether resampling or analytic based tests should be used
req_alpha: numeric, number of permutations for p value calculation
nBoot: numeric, number of bootstrap resamplings for confidence interval estimation
conf.level: numeric, between 0 and 1. Size of the confidence interval required
max_perm: numeric the maximum number of permutations that QUICKSTOP can do before giving up and returning NA. Can be set globally by setting the option "PharmacoGx_Max_Perm", or left at the default of ceiling(1/req_alpha*100).
verbose: boolean Should the function display messages?

Value

A vector reporting the effect size (estimateof the coefficient of drug concentration), standard error (se), sample size (n), t statistic, and F statistics and its corresponding p-value.

Examples

print("TODO::")
#> [1] "TODO::"