This chapter provides an overview of statistical methods for
genome-wide association studies. Here we describe how to
perform basic statistical analysis in a population-based genetic asso-
ciation case-control study. The steps described involve (1) deter-
mining the disease of interest for GWAS, (2) collecting study
samples and genotyping, (3) quality control for individuals and
SNPs, (4) adjustment for multiple testing, and (5) designing repli-
cation studies.2 Materials
A standard GWAS analysis requires genotypic information, pheno-
typic information, and a model that describes the specification of
prior distributions.- Computer workstations with Unix/Linux and Windows
operating systems:
PLINK for association analysis: [8]
http://pngu.mgh.harvard.edu/~purcell/plink/download.shtml. - Product for chip-based microarray from (seeNote 1):
Illumina platform:
http://www.ocimumbio.com/illumina-platform/.
Affymetrix platform:http://www.affymetrix.com/catalog/131
406/AFFY/Complete+GeneChip+Instrument+System#1_1. - Statistical software for data analysis and graphing such as:
R:http://cran.r-project.org/.
Haploview:https://www.broadinstitute.org/haploview/haplo
view. - 1000 Genomes Project resources for access to current data,
including genotypes, sequences, and genome mapping:http://
http://www.1000genomes.org/. - Program to calculate the effective number of independent SNPs
among a collection of SNPs in LD with each other, SNPSpD:
http://fraser.qimr.edu.au/general/daleN/SNPSpD/. - Program to convert HapMap format files to pedigree and map
format files:
http://bioinformatics.well.ox.ac.uk/resources.shtml. - Genetic Power Calculator for two-stage GWA case-control
studies, CaTS:
http://www.sph.umich.edu/csg/abecasis/CaTS/index.html. - Files: HapMap genotypic and phenotypic information, to be
downloaded fromhttp://www.hapmap.org.
100 Michelle Chang et al.