epiGWAS: Runs a selection of epistasis detection methods in a joint...

In EpiSlim/epiGWAS: Robust Methods for Epistasis Detection

Description

This function is a wrapper for the different epistasis detection methods implemented in this package. If methods is "all", we run OWL and the four modified outcome approaches. Otherwise, we run a selection of those methods. In this case, the methods argument is a character vector with its entries being the names of the functions to call.

Usage

epiGWAS(
  A,
  X,
  Y,
  propensity,
  methods = "all",
  parallel = TRUE,
  shift = 0.1,
  ...
)

Arguments

A	target variant
X	rest of the genotype
Y	phenotype
propensity	propensity scores
methods	character vector for the epistasis detection methods to call
parallel	whether to perform support estimation in a parallelized fashion for the modified outcome family of methods
shift	regularization parameter for shifted_outcome
...	additional arguments to be passed to stabilityGLM or stabilityBIG

Value

list of numeric vectors. Each vector corresponds to the auc scores of a particular method in methods.

See Also

OWL, modified_outcome, shifted_outcome, normalized_outcome and robust_outcome

Examples

n <- 20
p <- 8
X <- matrix((runif(n * p) < 0.4) + (runif(n * p) < 0.4),
             ncol = p, nrow = n) # SNP matrix
A <- rbinom(n, 1, 0.3)
propensity <- runif(n, min = 0.4, max = 0.8)
Y <- rnorm(n)
aucs <- epiGWAS(A, X, Y, propensity, lambda_min_ratio = 0.01, parallel = FALSE,
                 shift = 0.2, n_subsample = 1, short = TRUE, eps = 1e-4,
                 methods = c("normalized_outcome", "robust_outcome"))

names(aucs)

EpiSlim/epiGWAS documentation built on Nov. 19, 2019, 7:15 p.m.