A Machine Learning Tool for Genetic Association Studies

Portable, scalable and highly computationally efficient tool for genetic association studies."VariantScan" provides a set of machine learning methods (Linear, Local Polynomial Regression Fitting and Generalized Additive Model with Local Polynomial Smoothing) for genetic association studies that test for disease or trait association with genetic variants (biomarkers, e.g.,genomic (genetic loci), transcriptomic (gene expressions), epigenomic (methylations), proteomic (proteins), metabolomic (metabolites)). It is particularly useful when local associations and complex nonlinear associations exist.

VariantScan: a machine leaning tool for variant association testing

Introduction

This package provides a set of tools for performing association testing to identify QTLs in genome-wide association studies (GWAS,MWAS,EWAS,PWAS). It integrates three methods, Linear Model, Local Polynomial Fitting (Nonlinear Model) and Generalized Additive Model (GAM) to carry out genome-wide scanning. These methods can be also applied to case-control studies, where the ROC is used to assess the model performance.

Welcome any feedback and pull request.

Installation

Install the package from github:

library(devtools)

install_github("xinghuq/VariantScan")

library("VariantScan")

Testing the association between phenotypes and genotypes using genomic data

Get example file

f <- system.file('extdata',package='VariantScan')
infile <- file.path(f, "sim1.csv")
## read genotype file
geno=read.csv(infile)

# traits
traitq=geno[,14]
genotype=geno[,-c(1:14)]

# get PCs as covariates

PCs=prcomp(genotype)
PCs$x[,1:2]

## do Vscan using local polynomial regression fitting without specifying covariates

loessW=VScan(x=genotype,y=(traitq),methods ="loess")

## do Vscan using local polynomial regression fitting using PCs as covariates

loessWcv=VScan(x=genotype,y=(traitq),U=PCs$x[,1:2],methods ="loess")

## try linear model

lmW=VScan(x=genotype,y=(traitq),methods ="lm")
lmWcv=VScan(x=genotype,y=(traitq),U=PCs$x[,1:2],methods ="lm")

Visualizing the association signatures

Plot Manhattan plot

## 
Loci<-rep("Neutral", 1000)
Loci[c(201,211,221,231,241,251,261,271,281,291)]<-"QT"
Selected_Loci<-Loci[-which(Loci=="Neutral")]

library(ggplot2)
## Manhattan plot

g1=ggplot() +
  geom_point(aes(x=which(Loci=="Neutral"), y=-log10(lmWcv$p_norm$p.value[-which(Loci!="Neutral")])), col = "gray83") +
  geom_point(aes(x=which(Loci!="Neutral"), y=-log10(lmWcv$p_norm$p.value[-which(Loci=="Neutral")]), colour = Selected_Loci)) +
  xlab("SNPs") + ylab("-log10(p-value)") +ylim(c(0,35))+theme_bw()

g1


g2=ggplot() +
  geom_point(aes(x=which(Loci=="Neutral"), y=-log10(loessWcv$p_norm$p.value[-which(Loci!="Neutral")])), col = "gray83") +
  geom_point(aes(x=which(Loci!="Neutral"), y=-log10(loessWcv$p_norm$p.value[-which(Loci=="Neutral")]), colour = Selected_Loci)) +xlab("SNPs") + ylab("-log10(p-value)") +ylim(c(0,35))+theme_bw()

g3


```