Population stratification

Pritchard JK, Rosenberg NA. Use of unlinked genetic markers to detect population stratification in association studies. Am J Hum Genet 1999 65[l] 220-228. 

The two major statistical issues in asthma pharmacogenomics relate to population stratification and statistical power. 

Nearly all pharmacogenomic studies conducted to date in asthma have been case control studies. A major criticism of genetic case control studies of human diseases has been potentially undetected population stratification. Population stra-... 

Reich DE, Goldstein DB. Detecting association in a case-control study while correcting for population stratification. Genet Epidemiol 2001 20 4—16. 

Subheading 3.4.2) can correct for spurious association. These methods were recommended to adjust for population stratification even when a genetically homogeneous population is used in association studies. 

Several approaches have been proposed to deal with population stratification by using unlinked markers. In general, these methods fall into two categories model-based and non-model-based approaches. We briefly describe and explain three major methods and discuss their advantages as well as disadvantages. The basic understanding of these model-based and non-model-based methods is necessary and helpful when users apply them to analyze genetic data. 

In summary, GC adjusts for population stratification without the assumption or estimation of parameters such as the number of subpopulations involved in the study. It provides control of false-positive results caused by population structure as well as by multiple testing. One possible drawback of this method is that the correction of the test statistic is constant across the genome. As a result, GC may have less power in certain situations. 

In addition to these three major methods mentioned, several other computational approaches can also be used to deal with population stratification. For example, ADMIXMAP (22-26) is a model-based method that estimates the individual history of admixture. It can be applied to an admixed population with two or more ancestral populations. It also tests the association of a trait with ancestry at a marker locus with control for population structure. Wu et al. developed a software package in R (PSMIX) for the inference of population stratification and admixture (27). PSMIX is based on the maximum likelihood method. It performs as well as model-based methods such as STRUCTURE and is more computationally efficient. 

The recent emphasis on use of only family based association/linkage tests has ignored the readily available resource of case/control data (57), where sampling and then study of many thousands of samples are feasible with new techniques. Provided the patient and control groups are carefully matched for ethnicity, population stratification effects creating spurious associations are eliminated. The large collections of multiplex families now available for linkage studies in many complex diseases are obviously also a valuable resource for association screens (65,80). 

Whereas the impact of population admixture, or population stratification, to false positives in case-control studies is still under debate, every effort to minimize it should be taken. These efforts include the use of homogeneous populations or the use of family relatives as controls (e.g., siblings in case-sib studies). When these approaches are not possible, results should always be controlled for ethnicity. In... 

C. Ahaghotu, A. Jackson, C.A. Adebamowo et al. (2002). CYP3A4-V and prostate cancer in African Americans Causal or confounding association because of population stratification . Hum. Genet. 110, 553-560. 

---