China
Africa
Explore chapters and articles related to this topic
The Meta-Analysis of Genetic Studies
Published in Christopher H. Schmid, Theo Stijnen, Ian R. White, Handbook of Meta-Analysis, 2020
Cosetta Minelli, John Thompson
Population stratification is a type of confounding typical of genetic epidemiology, and it is caused by a mix of different ethnic groups within a single study population when the frequency of the genetic variant and the incidence of disease both vary across ethnicities within a single study (Cardon and Palmer, 2003). Population stratification is a theoretically important confounder of genetic associations, although its practical impact in real studies has been debated (Thomas and Witte, 2002; Wacholder et al., 2002). Investigation and correction for population stratification should be performed at the individual study-level prior to any meta-analysis of genetic association studies.
Prevalence and Genetic Epidemiology of Developmental Disabilities
Published in Merlin G. Butler, F. John Meaney, Genetics of Developmental Disabilities, 2019
Coleen A. Boyle, Kim Van Naarden Braun, Marshalyn Yeargin-Allsopp
Traditionally, case–control study designs have been used to study the associations between genetic variants and specific disorders, with the variant allele being the exposure. However, application of the case–control study design to evaluate a potential genetic association raises the problem of population stratification bias, defined by confounding attributable to differences in genetic backgrounds between cases and controls (138). While adjustment for factors possibly related to differing allelic distributions is possible using information on ethnicity, social background, and geographical origin of the parents, these factors are also often related to the disorder under study for other reasons (139). Without adequate adjustment of these systematic differences between cases and controls, this bias can obscure identification of a weak association of genes with modest effects. Some argue that the effects of population stratification bias are not a major threat to the validity of results from case–control studies of polymorphisms and disease risk (140). Using hypothetical simulations of the effects of ethnicity on the association between a genetic factor and disease risk, Wacholder et al. (141) suggested that in well-designed case-control studies in the United States, the risk ratio is biased less than 10% when ignoring the role of ethnicity. Aside from the presence or magnitude of population stratification bias, using nonrelated subjects in a case–control study requires large, and often unrealistic, sample sizes to detect interactions.
Design Issues in Case-Control Studies
Published in Ørnulf Borgan, Norman E. Breslow, Nilanjan Chatterjee, Mitchell H. Gail, Alastair Scott, Christopher J. Wild, Handbook of Statistical Methods for Case-Control Studies, 2018
The problem of population stratification refers to variation in genotype distributions across subpopulations with different ancestral origins. If these subpopulations also have differing disease risks for reasons apart from the genotypes, confounding of the association of disease with genotypes can result. These subpopulations may not be readily identifiable by self-report. Some studies have asked about race and/or ethnic origins of subjects’ parents and grandparents, but not only is this information often inaccurate, it can also be particularly challenging for multi-racial subjects. Failure to account for population stratification can lead to both uncontrolled confounding and over-dispersion of test statistics, so that conventional significance tests will have size above the nominal significance level. For genome-wide association studies, a variety of statistical adjustment procedures have been developed under the general heading of “genomic control” (Section 2.5.2). Family-based designs completely avoid the problem by making comparisons within each family, as the offspring all have identical ancestral origins. There are two main variants of this design, one using unaffected relatives as controls and one relying on gene transmissions from parents to offspring.
Polygenic risk scores for predicting outcomes and treatment response in psychiatry: hope or hype?
Published in International Review of Psychiatry, 2022
Laura Fusar-Poli, Bart P. F. Rutten, Jim van Os, Eugenio Aguglia, Sinan Guloksuz
Another important limitation of PRSs is the lack of population diversity in the current studies. It has long been known that population stratification is a major confounder in genetic research. European ancestry consistently represents the largest population: 96% of participants in GWAS by 2009 were of European descent (Popejoy & Fullerton, 2016). Current PRSs perform poorly in ancestries different from those of the GWAS training dataset. The active non-inclusion of a vast proportion of the population worldwide inevitably limits the validity and thus the predictive performance of PRSs in non-European samples. To improve the generalisability and applicability of the genomic-based prediction models, there is urgent need to achieve ‘racial’ and ethnic diversity in psychiatric genetics (Burkhard et al., 2021).
Family-based analysis of -675 4G/5G polymorphism in the PAI-1 gene of polycystic ovary syndrome in Chinese population
Published in Journal of Obstetrics and Gynaecology, 2022
Xiaocui Song, Li Ge, Dongsha Wang, Li Li, Dongmei Ma, Xiu Li, Xiaocui Song
The discrepancy of the previous studies might ascribe to the population stratification and environmental factors. To avoid these possible impact factors, we conducted the family study covering 285 family trios by the TDT to verify the linkage between PAI-1 4G/5G polymorphism and PCOS. Our data showed that the HWEp was 0.4742 and the MAF of 4G/5G was 0.325 by the TDT with no statistical significance (χ2=0.388, p = .5336). We failed to find the association between 4G/5G polymorphism of PAI-1 and PCOS. The conclusion was consistent with the result of Zhang et al. (2015), while our previous case-control studies showed that 4G/5G polymorphism of PAI-1 was associated with PCOS women in Han Chinese. The difference reminded us that we needed multicentre and collected more samples for further validation.
Association between genetic polymorphisms of cadherin 23 and noise-induced hearing loss: a meta-analysis
Published in Annals of Human Biology, 2022
Zhi-Dan Wu, Jun-Qi Lu, Wen-Jing Du, Shan Wu
To explore the sources of heterogeneity among the studies, we performed a subgroup analysis for rs1227049 and rs3802711 based on the results of HWE test. For rs3802711, subgroup analysis revealed an association in recessive, super-dominant, homozygote, and allele models. Although the controls of the two studies involved in Exon7 did not meet HWE, we did not exclude them, because Cosetta Minelli et al. suggested that studies that departed from HWE should be analysed the limitations on the research design rather than exclude them (Minelli et al. 2008). The most common reason for HWE departure was the genotyping error, attributed to difficulty in identifying heterozygotes in many genotyping platforms (Hosking et al. 2004). In addition, although all Chinese were included in our study, the researchers did not stratify the subjects according to ethnicity, which may lead to poor matching of genetic origins between groups and deviation from HWE (Yang M et al. 2006; Yang ZY et al. 2006; Wang et al. 2012; Jiao et al. 2020). Population stratification should be considered as an important confounding factor in the meta-analysis of genetic association studies (Salanti et al. 2005). Furthermore, the possibility of extreme deviation from HWE decreases as the sample size increases, so HWE deviation may be caused by insufficient sample size. In a word, we should increase the sample size and consider racial differences to reduce bias during epidemiological studies.