ZIQML.LRT: Likelihood ratio test based on ZIQML.fit()

Description Usage Arguments Value Examples

View source: R/ZIQML.LRT.R

Description

ZIQML.LRT returns the likelihood ratio test results based on the object returned by ZIQML.fit().

Usage

1
2
ZIQML.LRT(ZIQML.fit, coef = NULL, simulated.pvalue = F,
  permutation = 100)

Arguments

ZIQML.fit

Object returned by ZIQML.fit()

coef

An integer or vector indicating the coefficient(s) in design matrix to be tested. coef=1 is the intercept (i.e. baseline group effect), and should not be tested.

simulated.pvalue

If empirical p-values are computed, simulated.pvalue=TRUE. The default is FALSE.

permutation

The number of permutations used in simulating pvalues. The defualt value is 100.

Value

test.results

Likelihood ratio test results with test statistics, p-value, FDR (or adjusted pvalues). If simulated.pvalue=TRUE, test.results also includes simulated p-value and FDR.

Estimates

Estimated group effect in object ZIQML.fit.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
data('tcga.hnsc.match.edata','design')  
# 'tcga.hnsc.match.edata' contains RPKM of 1132 lncRNA genes and 80 samples. 
# 'design' is the design matrix for tumor vs normal tissue. 

fit.log=ZIQML.fit(edata=tcga.hnsc.match.edata,design.matrix=design,link='log') 
# Fit GLM by ZIQML with logarithmic link function 

LRT.results=ZIQML.LRT(fit.log,coef=2)  
# Likelihood ratio test on tumor vs normal, using observed p-values. 

# Include simulated p-values
fit.log.sim=ZIQML.fit(edata=tcga.hnsc.match.edata[1:10,],design.matrix=design,link='log') 
LRT.results.sim=ZIQML.LRT(fit.log.sim,coef=2,simulated.pvalue=TRUE,permutation=200)  

qianli10000/lncDIFF documentation built on Feb. 2, 2020, 6:46 p.m.