GPA-package: GPA (Genetic analysis incorporating Pleiotropy and...
In dongjunchung/GPA: GPA (Genetic analysis incorporating Pleiotropy and Annotation)

Description Details Author(s) References See Also Examples

This package provides functions for fitting GPA, a statistical approach to prioritizing GWAS results by integrating pleiotropy information and annotation data, along with ShinyGPA, a visualization toolkit to investigate the pleiotropic architecture using GWAS results.

Package:	GPA
Type:	Package
Version:	1.1.0
Date:	2017-06-30
License:	GPL (>= 2)
LazyLoad:	yes

This package contains a main class, GPA, which represents GPA model fit.

This package contains four main methods for the GPA framework (Chung et al., 2014), GPA, assoc, pTest, and aTest. GPA method fits the GPA model and assoc method implements association mapping. pTest and aTest methods implement hypothesis testing for pleiotropy and annotation enrichment, respectively.

This package contains two main methods for the ShinyGPA visualization toolkit (Kortemeier et al., 2017), fitAll and shinyGPA. fitAll function generates all the intermediate results needed to run shinyGPA. shinyGPA opens the ShinyGPA interface, which takes the results generated from fitAll as input.

Dongjun Chung, Emma Kortemeier

Maintainer: Dongjun Chung <chungd@musc.edu>

Chung D*, Yang C*, Li C, Gelernter J, and Zhao H (2014), "GPA: A statistical approach to prioritizing GWAS results by integrating pleiotropy information and annotation data," PLoS Genetics, 10: e1004787. (* joint first authors)

Kortemeier E, Ramos PS, Hunt KJ, Kim HJ, Hardiman G, and Chung D (2017), "ShinyGPA: An interactive and dynamic visualization toolkit for genetic studies."

GPA, assoc, pTest, aTest, GPA, fitAll, shinyGPA.

## Not run: 
library(GPA)

# simulation setting

nBin <- 20000
pi1 <- 0.2
common <- 0.5
betaAlpha <- c( 0.6, 0.6 )
annP <- c( 0.2, 0.4, 0.4, 0.4 )
seed <- 12345

# simulation setting

nCommon <- round( pi1 * common * nBin )
nUniq <- round( pi1 * ( 1 - common ) * nBin )
nBg <- nBin - 2 * nUniq - nCommon

# M * K matrix of GWAS p-value

set.seed( seed )

pvec1 <- c( rbeta( nCommon, betaAlpha[1], 1 ), rbeta( nUniq, betaAlpha[1], 1 ), 
	runif( nUniq ), runif( nBg ) )
pvec2 <- c( rbeta( nCommon, betaAlpha[2], 1 ), runif( nUniq ),
	rbeta( nUniq, betaAlpha[2], 1 ), runif( nBg ) )
pmat <- cbind( pvec1, pvec2 )

# M * D matrix of annotation
 	
ann <- c( 
	sample( c(1,0), nCommon, replace=TRUE, prob = c( annP[4], 1 - annP[4] ) ), 
	sample( c(1,0), nUniq, replace=TRUE, prob = c( annP[2], 1 - annP[2] ) ),
	sample( c(1,0), nUniq, replace=TRUE, prob = c( annP[3], 1 - annP[3] ) ),
	sample( c(1,0), nBg, replace=TRUE, prob = c( annP[1], 1 - annP[1] ) ) )
		
# GPA without annotation data

fit.GPA.noAnn <- GPA( pmat, NULL )
cov.GPA.noAnn <- cov( fit.GPA.noAnn )
		
# GPA with annotation data

fit.GPA.wAnn <- GPA( pmat, ann )
cov.GPA.wAnn <- cov( fit.GPA.wAnn )

# GPA under pleiotropy H0

fit.GPA.pleiotropy.H0 <- GPA( pmat, NULL, pleiotropyH0=TRUE )

# association mapping

assoc.GPA.wAnn <- assoc( fit.GPA.wAnn, FDR=0.05, fdrControl="global" )

# hypothesis testing for pleiotropy

test.pleiotropy <- pTest( fit.GPA.noAnn, fit.GPA.pleiotropy.H0 )

# hypothesis testing for annotation enrichment

test.annotation <- aTest( fit.GPA.noAnn, fit.GPA.wAnn )

# simulator function

simulator <- function( risk.ind, nsnp=20000, alpha=0.6 ) {
  
  m <- length(risk.ind)
  
  p.sig <- rbeta( m, alpha, 1 )
  pvec <- runif(nsnp)
  pvec[ risk.ind ] <- p.sig
  
  return(pvec)
}

# run simulation

set.seed(12345)
nsnp <- 10000
alpha <- 0.4
pmat <- matrix( NA, nsnp, 5 )

pmat[,1] <- simulator( c(1:2000), nsnp=nsnp, alpha=alpha )
pmat[,2] <- simulator( c(501:2500), nsnp=nsnp, alpha=alpha )
pmat[,3] <- simulator( c(4001:6000), nsnp=nsnp, alpha=alpha )
pmat[,4] <- simulator( c(4501:7500), nsnp=nsnp, alpha=alpha )
pmat[,5] <- simulator( c(8001:10000), nsnp=nsnp, alpha=alpha )

# Fit GPA for all possible pairs of GWAS datasets

out <- fitAll( pmat )

# Run the ShinyGPA app using the ouput from fitAll()

shinyGPA(out)

## End(Not run)