Biomedical Engineering Reference
In-Depth Information
Chapter 9
Population Stratification Analysis
in Genome-Wide Association Studies
Erika Salvi, Alessandro Orro, Guia Guffanti, Sara Lupoli, Federica Torri,
Cristina Barlassina, Steven Potkin, Daniele Cusi, Fabio Macciardi,
and Luciano Milanesi
Abstract
Differences in genetic background within two or more populations are
an important cause of disturbance in case-control association studies. In fact, when
mixing together populations of different ethnic groups, different allele frequencies
between case and control samples could be due to the ancestry rather than a real
association with the disease under study. This can easily lead to a large amount of
false positive and negative results in association study analysis. Moreover, the grow-
ing need to put together several data sets coming from different studies in order to
increase the statistical power of the analysis makes this problem particularly impor-
tant in recent statistical genetics research. To overcome these problems, different
correction strategies have been proposed, but currently there is no consensus about
a common powerful strategy to adjust for population stratification. In this chap-
ter, we discuss the state-of-the-art of strategies used for correcting the statistics for
genome-wide association analysis by taking into account the ancestral structure of
the population. After a short review of the most important methods and tools avail-
able, we will show the results obtained in two real data sets and discuss them in
terms of advantages and disadvantages of each algorithm.
9.1
Introduction
A genome-wide association study (GWAS) is defined as an examination of ge-
netic variation across the human genome aimed to identify genetic associations
with observable traits or qualitative dichotomous traits. To date, several genome-
wide association studies have been performed to identify chromosomal regions
containing disease-susceptibility loci by detecting differences in allele frequencies
between affected (cases) and unaffected individuals (controls). Mapping genetic
loci with GWAS is based on linkage disequilibrium (LD), which is defined as a
Search WWH ::
Custom Search