iDEA.fit.null: We constructed a permuted null distribution by permuting gene...

In xzhoulab/iDEA: Integrative Differential Expression and Gene Set Enrichment Analysis

We constructed a permuted null distribution by permuting gene labels. Specifically, we permuted the gene labels of annotation matrix to construced the null pvakue for gene sets.

Description

We constructed a permuted null distribution by permuting gene labels. Specifically, we permuted the gene labels of annotation matrix to construced the null pvakue for gene sets.

Usage

iDEA.fit.null(
  object,
  init_beta = NULL,
  init_tau = c(-2, 0.5),
  min_degene = 5,
  em_iter = 15,
  mcmc_iter = 1000,
  fit.tol = 1e-05,
  modelVariant = F,
  numPermute = 10,
  verbose = TRUE,
  ...
)

Arguments

object	iDEA object
init_beta	Initial value for gene effect size, beta in MCMC sampling
init_tau	Initial value for annotations, including the intercept in EM procedure, default is c(-2,0.5).
min_degene	The threshold for the number of detected DE genes. For some of extremely cases, the method does not work when the number of detected DE genes is 0.
em_iter	Maximum iteration for EM algorithm, default is 15
mcmc_iter	Maximum iteration for MCMC algorithm, default is 1000
fit.tol	Tol for fitting the model, default is 1e-5
modelVariant	Model option to run, boolean variable, if FALSE, runing the main iDEA mode, which models on z score statistics. if TRUE, runing iDEA variant model which models on beta effect size.
numPermute	Number of permutation used in object.null. Default is 10.
verbose	Print the progresses
...	Ignored

Value

Returns a idea object which stores the results of iDEA under permuted null.

xzhoulab/iDEA documentation built on Oct. 8, 2022, 8:54 a.m.