| Title: | Multi-SNP Mediation Intersection-Union Test |
|---|---|
| Description: | Testing the mediation effect of multiple SNPs on an outcome through a mediator. |
| Authors: | Wujuan Zhong |
| Maintainer: | Wujuan Zhong <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 1.3 |
| Built: | 2025-03-07 05:35:33 UTC |
| Source: | https://github.com/wjzhong/smut |
Matrix multiplication using RcppEigen.
eigenMapMatMult(A,B)eigenMapMatMult(A,B)
A, B
|
numeric (double) complex matrices or vectors. |
The matrix product. The value is the same as A %*% B
library(SMUT) A=matrix(1:9,3,3) A=A+0 B=as.matrix(c(5.0, 2.0, 0.0)) eigenMapMatMult(A,B) # the result is the same as A %*% B # Thanks for using our R package SMUTlibrary(SMUT) A=matrix(1:9,3,3) A=A+0 B=as.matrix(c(5.0, 2.0, 0.0)) eigenMapMatMult(A,B) # the result is the same as A %*% B # Thanks for using our R package SMUT
Testing coefficient of mediator, namely theta, in the outcome model. The outcome model is the following.
outcome ~ intercept + G*gamma + mediator*theta + error
Generalized_Testing_coefficient_of_mediator(G,mediator,outcome, covariates=NULL,outcome_type, approxi=TRUE,verbose=FALSE)Generalized_Testing_coefficient_of_mediator(G,mediator,outcome, covariates=NULL,outcome_type, approxi=TRUE,verbose=FALSE)
G |
n by p matrix (n rows and p columns). Each row is one individual; each column is one SNP. |
mediator |
a vector length of n. It is the mediator variable. |
outcome |
a vector length of n. It is the outcome variable. |
covariates |
n by r matrix (n rows and r columns). Each row is one individual; each column is one covariate. |
outcome_type |
Type of the outcome variable. "continuous" for a continuous outcome; "binary" for a binary outcome; "count" for a count outcome; "survival" for a survival outcome. |
approxi |
a boolean value. This is an indicator whether the approximation of computing derivatives is applied to save computing time. Default is TRUE. |
verbose |
a boolean value. If TRUE a lot of computing details is printed. Default is FALSE. |
p_value |
P value for testing the coefficient of mediator in the outcome model. |
theta_hat |
The point estimate of theta (coefficient of mediator) in the outcome model. |
Wujuan Zhong
Example genotype data for SMUT. It is a matrix with 100 rows and 200 columns. Each row is an individual; each column is a SNP.
It is a matrix with 100 rows and 200 columns. Each row is an individual; each column is a SNP.
Testing the mediation effect of multiple SNPs on an outcome following an exponential family distribution or a survival outcome through a continuous mediator.
GSMUT(G,mediator,outcome,covariates=NULL,outcome_type, approxi=TRUE,verbose=FALSE)GSMUT(G,mediator,outcome,covariates=NULL,outcome_type, approxi=TRUE,verbose=FALSE)
G |
n by p matrix (n rows and p columns). Each row is one individual; each column is one SNP. |
mediator |
a vector length of n. It is the mediator variable. |
outcome |
a vector length of n. It is the outcome variable. |
covariates |
n by r matrix (n rows and r columns). Each row is one individual; each column is one covariate. |
outcome_type |
Type of the outcome variable. "continuous" for a continuous outcome; "binary" for a binary outcome; "count" for a count outcome; "survival" for a survival outcome. |
approxi |
a boolean value. This is an indicator whether the approximation of computing derivatives is applied to save computing time. Default is TRUE. |
verbose |
a boolean value. If TRUE a lot of computing details is printed. Default is FALSE. |
p_value_IUT |
The p value for testing the mediation effect (theta*beta) based on intersection-union test. |
p_value_theta |
The p value for testing theta in the outcome model.
The outcome model is the following. |
theta_hat |
The point estimate of theta (coefficient of mediator) in the outcome model. |
p_value_beta |
The p value for testing beta in the mediator model.
The mediator model is the following. |
Wujuan Zhong
library(SMUT) ##### for a binary outcome ##### set.seed(1) # generate two covariates covariate_1=rnorm(nrow(Genotype_data),0,1) covariate_2=sample(c(0,1),size=nrow(Genotype_data),replace = TRUE) covariates=cbind(covariate_1,covariate_2) # generate a mediator beta=rnorm(ncol(Genotype_data),0,0.5) tau_M=c(-0.3,0.2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + eigenMapMatMult(covariates, tau_M) + e1 #### generate a binary outcome #### theta=1 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator pi=1/(1+exp( -(eta ) )) outcome=rbinom(length(pi),size=1,prob=pi) result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="binary") print(result) # p_value_IUT is the p value for the mediation effect. ## Not run: ##### generate a count outcome ##### theta=1 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator mu_param=exp(eta) # the mean parameter phi_param=10 # the shape parameter outcome=rnbinom(length(mu_param),size=phi_param,mu=mu_param) result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="count") print(result) # p_value_IUT is the p value for the mediation effect. ##### generate a survival outcome ##### theta=2 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator v=runif(nrow(Genotype_data)) lambda=0.01; rho=1; rateC=0.001 Tlat=(- log(v) / (lambda * exp( eta )))^(1 / rho) # censoring times C= rexp(nrow(Genotype_data), rate=rateC) # follow-up times and event indicators time= pmin(Tlat, C) status= as.numeric(Tlat <= C) outcome=cbind(time,status) colnames(outcome)=c("time","status") result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="survival") print(result) # p_value_IUT is the p value for the mediation effect. ## End(Not run)library(SMUT) ##### for a binary outcome ##### set.seed(1) # generate two covariates covariate_1=rnorm(nrow(Genotype_data),0,1) covariate_2=sample(c(0,1),size=nrow(Genotype_data),replace = TRUE) covariates=cbind(covariate_1,covariate_2) # generate a mediator beta=rnorm(ncol(Genotype_data),0,0.5) tau_M=c(-0.3,0.2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + eigenMapMatMult(covariates, tau_M) + e1 #### generate a binary outcome #### theta=1 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator pi=1/(1+exp( -(eta ) )) outcome=rbinom(length(pi),size=1,prob=pi) result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="binary") print(result) # p_value_IUT is the p value for the mediation effect. ## Not run: ##### generate a count outcome ##### theta=1 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator mu_param=exp(eta) # the mean parameter phi_param=10 # the shape parameter outcome=rnbinom(length(mu_param),size=phi_param,mu=mu_param) result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="count") print(result) # p_value_IUT is the p value for the mediation effect. ##### generate a survival outcome ##### theta=2 gamma=rnorm(ncol(Genotype_data),0,0.5) tau=c(-0.2,0.2) eta=1 + eigenMapMatMult(Genotype_data, gamma) + eigenMapMatMult(covariates, tau) + theta * mediator v=runif(nrow(Genotype_data)) lambda=0.01; rho=1; rateC=0.001 Tlat=(- log(v) / (lambda * exp( eta )))^(1 / rho) # censoring times C= rexp(nrow(Genotype_data), rate=rateC) # follow-up times and event indicators time= pmin(Tlat, C) status= as.numeric(Tlat <= C) outcome=cbind(time,status) colnames(outcome)=c("time","status") result=GSMUT(G=Genotype_data,mediator=mediator,outcome=outcome, covariates=covariates,outcome_type="survival") print(result) # p_value_IUT is the p value for the mediation effect. ## End(Not run)
Testing the mediation effect of multiple SNPs on an outcome through a mediator.
SMUT(G, mediator, outcome, covariates=NULL, outcome_type="continuous", method="score",approxi=TRUE, debug=FALSE)SMUT(G, mediator, outcome, covariates=NULL, outcome_type="continuous", method="score",approxi=TRUE, debug=FALSE)
G |
n by p matrix (n rows and p columns). Each row is one individual; each column is one SNP. |
mediator |
a vector length of n. It is the mediator variable. |
outcome |
a vector length of n. It is the outcome variable. |
covariates |
n by r matrix (n rows and r columns). Each row is one individual; each column is one covariate. |
outcome_type |
Type of the outcome variable. For now, this package only deals with continuous outcome. Default is "continuous". |
method |
The method of testing coefficient of mediator in the outcome model. The score test is used. Default is "score". |
approxi |
a boolean value. This is an indicator whether the approximation of the score statistic is applied to save computing time. Default is TRUE. |
debug |
a boolean value. If TRUE a lot of computing details is printed; otherwise the function is completely silent. Default is FALSE. |
p_value_IUT |
The p value for testing the mediation effect (theta*beta) based on intersection-union test. |
p_value_theta |
The p value for testing theta in the outcome model.
The outcome model is the following. |
p_value_beta |
The p value for testing beta in the mediator model.
The mediator model is the following. |
Wujuan Zhong
Zhong, W., Spracklen, C. N., Mohlke, K. L., Zheng, X., Fine, J., & Li, Y. (2019). Multi-SNP mediation intersection-union test. Bioinformatics.
library(SMUT) # generate one mediator and one outcome # first example, the mediation effect is significant set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.001655787). # second example, the mediation effect is non-significant set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have non-significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.3281677). # third example, the mediation effect is non-significant set.seed(1) beta=rep(0,ncol(Genotype_data)) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have non-significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.5596977). # Thanks for using our R package SMUTlibrary(SMUT) # generate one mediator and one outcome # first example, the mediation effect is significant set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.001655787). # second example, the mediation effect is non-significant set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have non-significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.3281677). # third example, the mediation effect is non-significant set.seed(1) beta=rep(0,ncol(Genotype_data)) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=SMUT(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # p_value_IUT is the p value for the mediation effect. # we have non-significant(at alpha level 0.05) mediation effects (p_value_IUT = 0.5596977). # Thanks for using our R package SMUT
Testing coefficient of mediator, namely theta, in the outcome model. The outcome model is the following.
outcome ~ intercept + G*gamma + mediator*theta + error
Testing_coefficient_of_mediator(G, mediator, outcome, covariates=NULL, outcome_type="continuous", method="score", approxi=TRUE, debug=FALSE)Testing_coefficient_of_mediator(G, mediator, outcome, covariates=NULL, outcome_type="continuous", method="score", approxi=TRUE, debug=FALSE)
G |
n by p matrix (n rows and p columns). Each row is one individual; each column is one SNP. |
mediator |
a vector length of n. It is the mediator variable. |
outcome |
a vector length of n. It is the outcome variable. |
covariates |
n by r matrix (n rows and r columns). Each row is one individual; each column is one covariate. |
outcome_type |
Type of the outcome variable. For now, this package only deals with continuous outcome. Default is "continuous". |
method |
The method of testing coefficient of mediator in the outcome model. The score test is used. Default is "score". |
approxi |
a boolean value. This is an indicator whether the approximation of the score statistic is applied to save computing time. Default is TRUE. |
debug |
a boolean value. If TRUE a lot of computing details is printed; otherwise the function is completely silent. Default is FALSE. |
P value for testing the coefficient of mediator in the outcome model.
Wujuan Zhong
library(SMUT) # load the Genotype data included in this R package data("Genotype_data") # generate one mediator and one outcome set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=Testing_coefficient_of_mediator(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # Thanks for using our R package SMUTlibrary(SMUT) # load the Genotype data included in this R package data("Genotype_data") # generate one mediator and one outcome set.seed(1) beta=rnorm(ncol(Genotype_data),1,2) e1 = rnorm(nrow(Genotype_data), 0, 1) mediator = 1 + eigenMapMatMult(Genotype_data,beta) + e1 theta=0.8 gamma=rnorm(ncol(Genotype_data),0.5,2) e2 = rnorm(nrow(Genotype_data), 0, 1) outcome = 2 + eigenMapMatMult(Genotype_data,gamma) + theta*mediator + e2 p_value=Testing_coefficient_of_mediator(G=Genotype_data,mediator=mediator,outcome=outcome) print(p_value) # Thanks for using our R package SMUT