getp.func: This function caculate P-value for every trio

In QidiPeng/eQTLMAPT: eQTL Mediaiton Analysis with accelerated Permutation Testing approaches

This function caculate P-value for every trio

Description

This function calculates the P-value for each trios. If Minperm=0, only the nominal P-value is calculated. If Minperm=Maxperm, the empirical P-value is calculated using a fixed number of permutation statistics; otherwise, the empirical P-value is calculated using the adaptive permutation scheme. The user can specify whether to use the GPD fit to estimate a more accurate empirical P-value, and at the same time specify how small the empirical P-value is for GPD fitting.

Usage

getp.func(
  i,
  triomatrix,
  confounders,
  Minperm = 100,
  Maxperm = 10000,
  use.gpd = FALSE,
  gpd.perm = 0.01,
  pool_cov = NULL,
  est_conf_pool_idx = NULL,
  use.PC = FALSE
)

Arguments

i	Trios index in triomatrix
triomatrix	A three-dimensional matrix of size: samples number * trios number * 3. Triomatrix[i,j,1] represents the genotype of the j-th trios at the i-th sample,and triomatrix[i,j,2] represents the feature1 data of the j-th trios at the i-th sample, triomatrix[i,j,3] represents the feature2 data of the j-th trios at the i-th sample.
Minperm	Decide whether to use the parameters of the GPD fit. If the value is 0, only the nominal P-value is calculated. If the proportion of the permutation statistic better than the original statistic to the total number of permutations exceeds this value, a more accurate empirical P value is estimated using the GPD fit. If Minperm=0, only the nominal P-value is calculated. We set Minperm=100 as default.
Maxperm	Maximum number of permutation. We set Maxperm=10000 as default.
use.gpd	Whether to use the GPD fit to estimate a more accurate empirical P-value. We set use.gpd=NULL as default.
gpd.perm	The proportion parameter for estimating the empirical P-value when using GPD fit. When the proportion of permutation better than the original statistic is greater than par, the GPD is fitted to estimate the empirical P-value. We set gpd.perm=0.01 as default.
pool_cov	Candidate Confusion Variable Pool. We set pool_cov=NULL as default, which use.PC requied.
est_conf_pool_idx	The index of the adaptively selected confunding variable. We set pool_cov=NULL as default, at this time pool_cov=NULL too.
use.PC	Whether the candidate confusion variable pool is PCs.
confunders	A confounders matrix which is adjusted in all mediation tests.

Value

The algorithm will return a list of nperm, empirical.p, empirical.p.gpd, nominal.p, std.error, t_stat, beta, beta.total, beta.change.

nperm	The actual number of permutations for testing mediation, equal to the input parameter nperm.
empirical.p	The mediation Empirical P-values with nperm times permutation.
empirical.p.gpd	The mediation Empirical P-values with nperm times permutation using GPD fit.
nominal.p	The mediation nominal P-values. A matrix with dimension of the number of trios.
std.error	The return std.error value of feature1 for fit liner models.
t_stat	The return t_stat value of feature1 for fit liner models.
beta	The return beta value of feature2 for fit liner models in the case of feature1.
beta.total	The return beta value of feature2 for fit liner models without considering feature1.
beta.change	The proportions mediated.

QidiPeng/eQTLMAPT documentation built on Jan. 25, 2023, 11:03 p.m.