View source: R/QPCorr.matrix.R
QPCorr.matrix | R Documentation |
A robust gene-environment interaction identification approach using the quantile partial correlation technique. This approach is a marginal analysis approach built on the quantile regression technique, which can accommodate long-tailed or contaminated outcomes. For response with right censoring, Kaplan-Meier (KM) estimator-based weights are adopted to easily accommodate censoring. In addition, it adopts partial correlation to identify important interactions while properly controlling for the main genetic (G) and environmental (E) effects.
QPCorr.matrix(G, E, Y, tau, w = NULL, family = c("continuous", "survival"))
G |
Input matrix of |
E |
Input matrix of |
Y |
Response variable. A quantitative vector for |
tau |
Quantile. |
w |
Weight for accommodating censoring if |
family |
Response type of |
Matrix of (censored) quantile partial correlations for interactions.
Yaqing Xu, Mengyun Wu, Qingzhao Zhang, and Shuangge Ma. Robust identification of gene-environment interactions for prognosis using a quantile partial correlation approach. Genomics, 111(5):1115-1123, 2019.
QPCorr.pval
method.
alpha=matrix(0,5,1) alpha[1:2]=1 beta=matrix(0,6,100) beta[1,1:5]=1 beta[2:3,1:5]=2 beta[4:6,6:7]=2 sigmaG<-AR(rho=0.3,100) sigmaE<-AR(rho=0.3,5) G<-MASS::mvrnorm(200,rep(0,100),sigmaG) E<-MASS::mvrnorm(200,rep(0,5),sigmaE) e1<-rnorm(200*.05,50,1);e2<-rnorm(200*.05,-50,1);e3<-rnorm(200*.9) e<-c(e1,e2,e3) # continuous y1=simulated_data(G=G,E=E,alpha=alpha,beta=beta,error=e,family="continuous") cpqcorr_stat1<-QPCorr.matrix(G,E,y1,tau=0.5,w=NULL,family="continuous") # survival y2=simulated_data(G,E,alpha,beta,rnorm(200,0,1),family="survival",0.7,0.9) cpqcorr_stat<-QPCorr.matrix(G,E,y2,tau=0.5,w=NULL,family="survival")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.