sim_omics: Simulate & test mediation methods on omics dataset

In jdreyf/Hitman: High-throughput mediation analysis

Simulate & test mediation methods on omics dataset

Description

Simulate & test mediation methods on omics dataset.

Usage

sim_omics(
  b1 = 1,
  t2 = b1,
  t1 = 5,
  nsamp = 50,
  ngene = 100,
  FDR = 0.25,
  Sigma = diag(ngene),
  prop.consistent = 1/ngene,
  prop.inconsistent = 0,
  prop.em = 0,
  prop.my = 0,
  sd.mn = 0,
  nsim = 10^3,
  fdr.method = c("BH", "BY"),
  seed = 0,
  verbose = TRUE
)

Arguments

b1	Numeric value that beta1 ("b1"), the EM coefficient, takes for non-null mediators.
t2	Numeric value that theta2 ("t2"), the MY coefficient, takes for non-null mediators.
t1	Numeric value of theta1 i.e. the effect of the exposure on the outcome.
nsamp	Number of samples.
ngene	Number of genes.
FDR	FDR threshold to apply.
Sigma	Gene covariance matrix.
prop.consistent	Proportion of genes that are consistent mediators. Must be at least 1/ngene.
prop.inconsistent	Proportion of genes that are inconsistent mediators.
prop.em	Proportion of genes where E –> M but not M –> Y given E.
prop.my	Proportion of genes where M –> Y given E but not E –> M.
sd.mn	Numeric standard deviation of measurement noise.
nsim	Number of simulations.
fdr.method	Character string; either "BH" for Benjamini-Hochberg or "BY" for Benjamini-Yekutieli. See stats::p.adjust.
seed	Random seed for reproducibility.
verbose	Logical; should the number of simulations be printed every 100 simulations?

Value

Matrix with proportion of significant calls for every method for true and null mediators.

jdreyf/Hitman documentation built on April 12, 2025, 1:35 p.m.