Admixture-matched case-control study: a practical approach for genetic association studies in admixed populations.

In case-control genetic association studies, cases are subjects with the disease and controls are subjects without the disease. At the time of case-control data collection, information about secondary phenotypes is also collected. In addition to studies of primary diseases, there has been some interest in studying genetic variants associated with secondary phenotypes. In genetic association studies, the deviation from Hardy-Weinberg proportion (HWP) of each genetic marker is assessed as an initial quality check to identify questionable genotypes. Generally, HWP tests are performed based on the controls for the primary disease or secondary phenotype. However, when the disease or phenotype of interest is common, the controls do not represent the general population. Therefore, using only controls for testing HWP can result in a highly inflated type I error rate for the disease- and/or phenotype-associated variants. Recently, two approaches, the likelihood ratio test (LRT) approach and the mixture HWP (mHWP) exact test were proposed for testing HWP in samples from case-control studies. Here, we show that these two approaches result in inflated type I error rates and could lead to the removal from further analysis of potential causal genetic variants associated with the primary disease and/or secondary phenotype when the study of primary disease is frequency-matched on the secondary phenotype. Therefore, we proposed alternative approaches, which extend the LRT and mHWP approaches, for assessing HWP that account for frequency matching. The goal was to maintain more (possible causative) single-nucleotide polymorphisms in the sample for further analysis. Our simulation results showed that both extended approaches could control type I error probabilities. We also applied the proposed approaches to test HWP for SNPs from a genome-wide association study of lung cancer that was frequency-matched on smoking status and found that the proposed approaches can keep more genetic variants for association studies.

Sustained Virologic Response to Hepatitis C Virus Therapy and Spontaneous Viral Clearance Found Linked to Genotype

A family-based study design is commonly used in gene mapping studies of complex human diseases. Most family-based studies use the transmission of alleles to assess evidence of association. It is generally believed that the transmission disequilibrium test (TDT) is robust against spurious association due to population stratification or admixture. While this is true when population stratification is due to discrete population structure, one should use the TDT-type methods with caution when they are applied to admixed populations in which population structure exists in local genomic regions. In a recently admixed population, such as African Americans and Hispanic Americans, the linkage disequilibrium coefficient between a marker and disease loci in the parental generation contains a spurious component from the admixture process. In this paper, we show that the general belief that family-based design would guard against spurious association caused by population stratification does not always hold in admixed populations. It is safe to use the TDT as a test of association when population stratification is due to global genome ancestry difference. However, when population stratification is due to local ancestry difference in certain genomic regions, the use of the TDT as a test of association can lead to spurious association. Second, we present a statistical framework for fine mapping of disease-associated genetic variants in admixed families. Unlike the TDT and other family-based association tests, this method does not rely on transmission disequilibrium and therefore can control local ancestry difference between transmitted and untransmitted alleles. Through simulations, we show that this method can control type I error rates under a wide range of population stratification mechanisms.

Loci exhibiting Hardy-Weinberg disequilibrium (HWD) are often excluded from association studies, because HWD may indicate genotyping error, population stratification or selection bias. For case-control studies, HWD can result from a genetic effect at the locus. We extend the modelling to accommodate both stratification and genetic effects. Theoretical genotype frequencies and HWD coefficients are derived under a general genetic model for a population with two strata. Maximum likelihood is used to estimate model parameters and a test for lack of fit identifies the models most consistent with the data. Simulations were used to assess the method. The technique was applied to a group of ethnically and clinically heterogeneous kidney stone formers and controls, both exhibiting HWD for the R990G SNP of the CASR gene. Results indicate the best fitting model incorporates both stratification and genetic association. The ability of our method to apportion HWD to stratification and genetic effects may well be a significant advance in dealing with heterogeneity in case-control genetic association studies.

The HapMap project has given case-control association studies a unique opportunity to uncover the genetic basis of complex diseases. However, persistent issues in such studies remain the proper quantification of, testing for, and correction for population stratification (PS). In this paper, we present the first unified paradigm that addresses all three fundamental issues within one statistical framework. Our unified approach makes use of an omnibus quantity (δ), which can be estimated in a case-control study from suitable null loci. We show how this estimated value can be used to quantify PS, to statistically test for PS, and to correct for PS, all in the context of case-control studies. Moreover, we provide guidelines for interpreting values of δ in association studies (e.g., at α = 0.05, a δ of size 0.416 is small, a δ of size 0.653 is medium, and a δ of size 1.115 is large). A novel feature of our testing procedure is its ability to test for either strictly any PS or only ‘practically important’ PS. We also performed simulations to compare our correction procedure with Genomic Control (GC). Our results show that, unlike GC, it maintains good Type I error rates and power across all levels of PS.

Background: Asthma is a complex genetic disease, caused by the interaction of multiple genetic and environmental factors. T cell immunoglobulin domain and mucin domain (Tim) genes are located in chromosome 5q31–33, a region repeatedly linked to asthma or asthma-related phenotypes in several populations. Two members of Tim families, Tim-1 and Tim-3, which are expressed on T cell surface and potentially involved in T cell proliferation and differentiation, are good candidate genes for asthma. We investigated whether genetic variants or haplotypes in Tim-1 and Tim-3 genes confer susceptibility to asthma in a Chinese population. Methods: A total of 9 polymorphisms were selected by using the HapMap Han Chinese population data and a haplotype-tagging single nucleotide polymorphism approach. Polymerase chain reaction fragment length polymorphism was adapted to determine the genotype in 118 complete Chinese trios of asthma. Then, transmission disequilibrium test (TDT), haplotypic relative risk (HRR), linkage disequilibrium and haplotype analysis were performed. Results: The single locus TDT and HRR analysis showed the 9 polymorphisms were not transmitted preferentially to asthma-affected children (all p > 0.05). However, in both the haplotypic TDT and HRR analysis, the haplotype G-A-ins-C-G consisting of 5 Tim-1 polymorphisms was found to be overtransmitted to affected offspring (χ² = 4.51, p = 0.03) and the haplotype G-G-G consisting of 3 Tim-3 polymorphisms was found to be undertransmitted to asthma children (χ² = 8.24, p = 0.004). Conclusions: We conclude that it is unlikely that the Tim-1 or Tim-3 polymorphisms influence asthma susceptibility individually, but that the haplotypes of variants may be functional or may be in linkage disequilibrium with other functional single nucleotide polymorphisms.

People with different ancestries inherit different risks and encounter different environmental exposures resulting in different somatic profiles. A lack of knowledge about ancestry-specific alterations is a major barrier to implementing precision medicine, leading to inequities in genetic testing, targeted treatment and clinical trial design for cancer patients from the underserved populations. However, a limited number of cancer cases from non-European populations have been sequenced in research setting, and paired normal samples are often not collected in routine clinical care. Here, we inferred African (AFR), South Asian (SAS), East Asian (EAS), American (AMR) and European (EUR) ancestry from comprehensive genomic profiling (CGP) of over 200,000 tumors at Foundation Medicine and identified 18717, 8588, 6594 patients with significant proportions of AFR, AMR, EAS ancestry, respectively. Using logistic regressions with age of diagnosis, gender and tumor mutation burden (TMB) as covariates, we identified 165 ancestry-associated genes across 14 common cancer types. We found that in colorectal cancer, KRAS (OR=1.5, FDR corrected p=4x10-44) and APC (OR=1.4, FDR corrected p=1x10-16) mutations were enriched in patients with AFR ancestry, whereas BRAF mutations were depleted in both AFR (OR=0.5, FDR corrected p=5x10-24) and EAS (OR=0.6, FDR corrected p=6x10-5) patients. CDK12 mutations were enriched in both AFR (OR=1.6, FDR corrected p=2x10-4) and EAS (OR=3.1, FDR corrected p=2x10-7) prostate cancer patients. For patients with AMR ancestry, increased mutation frequency was observed in SETD2 in pancreatic cancer (OR=3.4, FDR corrected p=1x10-4), and VHL in renal clear cell carcinoma (OR=1.9, FDR corrected p=2x10-4). For renal cell carcinoma patients with AFR ancestry, mutations in NF2 were enriched (OR=2.5, FDR =8x10-6). In GBM, TERT alterations were associated with EUR ancestry (OR=1.3, FDR corrected p=0.003). Interestingly, TMB-high status was associated with EUR ancestry in melanoma (OR=5.9, p=6x10-36) and anti-correlated with AFR ancestry in endometrial carcinoma (OR=0.7, p=1x10-6). Our large-scale ancestry analysis characterized the somatic landscape of non-European cancer patients. Future directions include quantifying the germline heritability of identified ancestry-associated mutations using a local ancestry risk score (LRS). Citation Format: Jian Carrot-Zhang, Justin New, Amy Xie, Garrett Frampton. Genomic analysis of patients across continental ancestry groups [abstract]. In: Proceedings of the 15th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2022 Sep 16-19; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2022;31(1 Suppl):Abstract nr C016.

African-American populations are genetically admixed. Studies performed among unrelated individuals from ethnically admixed populations may be both vulnerable to confounding by population stratification, but offer an opportunity for efficiently mapping complex traits through admixture linkage disequilibrium. By typing 42 ancestry-informative markers and estimating genetic ancestry, we assessed genetic admixture and heterogeneity among African-American participants in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort. We also assessed associations between individual genetic ancestry and several quantitative and binary traits related to cardiovascular risk. We found evidence of population sub-structure and excess inter-marker linkage disequilibrium, consistent with recent admixture. The estimated group admixture proportions were 78.1% African and 22.9% European, but differed according to geographic region. In multiple regression models, African ancestry was significantly associated with decreased total cholesterol, decreased LDL-cholesterol, and decreased triglycerides, and also with increased risk of insulin resistance. These observed associations between African ancestry and several lipid traits are consistent with the general tendency of individuals of African descent to have healthier lipid profiles compared to European-Americans. There was no association between genetic ancestry and hypertension, BMI, waist circumference, CRP level, or coronary artery calcification. These results demonstrate the potential for confounding of genetic associations with some cardiovascular disease-related traits in large studies involving US African-Americans.

This unit provides an overview of the design and analysis of population-based case-control studies of genetic risk factors for complex disease. Considerations specific to genetic studies are emphasized. The unit reviews basic study designs differentiating case-control studies from others, presents different genetic association strategies (candidate gene, genome-wide association, and high-throughput sequencing), introduces basic methods of statistical analysis for case-control data and approaches to combining case-control studies, and discusses measures of association and impact. Admixed populations, controlling for confounding (including population stratification), consideration of multiple loci and environmental risk factors, and complementary analyses of haplotypes, genes, and pathways are briefly discussed. Readers are referred to basic texts on epidemiology for more details on general conduct of case-control studies.

Case-control studies are a powerful strategy to identify candidate genes in complex diseases. In admixed populations, association studies can be affected by population stratification, leading to spurious genetic associations. Ancestry informative markers (AIMs) can be used to minimise this effect. The aim of this work was to select a set of AIMs to estimate population stratification in a Brazilian case-control study performed using a genome-wide array. A total of 345 single nucleotide polymorphism (SNP) AIMs, selected from the Cytoscan HD array and based on previously reported panels, was used to discriminate between European, African, and Amerindian populations. These SNP-AIMs were used to infer ancestry in systemic lupus erythematosus (SLE) patients (n = 23) and in healthy subjects (n = 110). Moderate population substructure was observed between SLE and control groups (Fst = 0.0113). Although patients and controls have shown a major European genomic contribution, significant differences in the European (P = 6.47 × 10-5 ) and African (P = 1.14 × 10-3 ) ancestries were detected between the two groups. We performed a two-step validation of the 345 SNP-AIMs panel estimating the ancestral contributions using a panel of 12 AIMs and approximately 70K SNPs from the array. Evaluation of population substructure in case-control studies, avoiding spurious genetic associations, can be performed using our panel of 345 SNP-AIMs.

<p>Supplementary Figure S1. T-ALL subtype classifying driver genes among individuals of European, Admixed American, and African ancestry. Supplementary Figure S2. T-ALL subtype classifying driver genes among patients with Early T-cell Precursor (ETP)-like subtype among individuals of European, Admixed American and African ancestry. Supplementary Figure S3. Distribution of 15 genomically defined T-ALL subtypes described by Pölönen and colleagues (16) among individuals of African and Non-African ancestry. Supplementary Figure S4. 5-year event free survival for NOTCH1 and FBXW7 by ancestral group. Supplementary Figure S5. Prevalence of (A) coding versus non-coding alterations and (B) types of pathogenic variants in NOTCH1 among individuals of African versus non-African ancestry. Supplementary Figure S6. Event free survival among all patients, patients of European, Admixed American, and African ancestry by NOTCH1 alterations (no mutation, favorable, unfavorable). Supplementary Figure S7. NOTCH1 coding mutation types (frameshift, missense, nonsense) on a lollipop plot in patients of European, African, and Admixed American Ancestry. Supplementary Figure S8. 5-year event free survival for genes with at least 5% frequency for each ancestral group (CDKN2A, PHF6, PTEN, DNM2). Supplementary Figure S9. 5-year event free survival for genes with at least 5% frequency for each ancestral group (LEF1, MYB, MYC, 6q loss). Supplementary Figure S10. 5-year event free survival for genes with at least 5% frequency for each ancestral group (WT1, USP7, IL7R, N/KRAS). Supplementary Figure S11. Distribution of genetic ancestral composition of individuals by categorically defined genetic ancestry.</p>

It is now possible to routinely collect DNA for the investigation of genetic risk factors in psychiatric epidemiology. However, concerns have been raised about the investigation of genetic risk factors in population surveys and in case-control studies with population controls, because of spurious associations arising through population stratification. To overcome this problem, some geneticists advocate the use of family controls and studies on isolated populations. However, such approaches may not be the best for psychiatric epidemiology. A critical analysis is made of arguments against the investigation of genetic risk factors in studies with population controls. It is argued that concerns about population stratification have been overblown. It is simply a form of confounding and can occur in studies of environmental as well as genetic risk factors. Epidemiologists have traditionally dealt with this and other types of confounding by matching on potential confounders in the design of the study, or stratification by confounders in the analysis. Family controls overcome population stratification, but not other forms of confounding. They are a good choice when the sole interest is genetic risk factors, but psychiatric epidemiologists are typically interested in both genetic and environmental factors. Family controls are poor for examining environmental risk factors because they can give rise to over-matching. Isolated populations may produce genetic associations that do not generalize to other populations, and they may have a restricted range of environments. We conclude that population surveys and case-control studies using population controls are suitable for the investigation of both genetic and environmental risk factors and have greater public health relevance than alternative designs.

Genetic case-control association studies--correcting for multiple testing.

Disparities in breast cancer biology are evident between American women of African ancestry (AA) and European ancestry (EA) and may be due, in part, to differences in immune function. To assess the potential role of constitutional host immunity on breast carcinogenesis, we tested associations between breast cancer risk and 47 single nucleotide polymorphisms (SNPs) in 26 cytokine-related genes of the adaptive immune system using 650 EA (n = 335 cases) and 864 AA (n = 458 cases) women from the Women's Circle of Health Study (WCHS). With additional participant accrual to the WCHS, promising SNPs from the initial analysis were evaluated in a larger sample size (1,307 EAs and 1,365 AAs). Multivariate logistic regression found SNPs in genes important for T helper type 1 (Th1) immunity (IFNGR2 rs1059293, IL15RA rs2296135, LTA rs1041981), Th2 immunity (IL4R rs1801275), and T regulatory cell-mediated immunosuppression (TGFB1 rs1800469) associated with breast cancer risk, mainly among AAs. The combined effect of these five SNPs was highly significant among AAs (P-trend = 0.0005). When stratified by estrogen receptor (ER) status, LTA rs1041981 was associated with ER-positive breast cancers among EAs and marginally among AAs. Only among AA women, IL15 rs10833 and IL15RA rs2296135 were associated with ER-positive tumors, and IL12RB1 rs375947, IL15 rs10833 and TGFB1 rs1800469 were associated with ER-negative tumors. Our study systematically identified genetic variants in the adaptive immune response pathway associated with breast cancer risk, which appears to differ by ancestry groups, menopausal status and ER status.

The gene encoding phosphatidylinositol-4-phosphate 5-kinase (PIP5K1C) has been recently implicated in pain regulation. Interestingly, a recent cross-tissue and cross-phenotypic epigenetic analysis identified the same gene in alcohol use disorder (AUD). Given the high comorbidity between AUD and chronic pain, we hypothesized that genetic variation in PIP5K1C might contribute to susceptibility to AUD. We conducted a case-control association study of genetic variants in PIP5K1C. Association analyses of 16 common PIP5K1C single nucleotide polymorphisms (SNPs) were conducted in cases and controls of African (427 cases and 137 controls) and European ancestry (488 cases and 324 controls) using standard methods. In addition, given the prominent role of the opioid system in pain signaling, we investigated the effects of acute alcohol exposure on PIP5K1C expression in humanized transgenic mice for the μ-opioid receptor that included the OPRM1 A118G polymorphism, a widely used mouse model to study analgesic response to opioids in pain. PIP5K1C expression was measured in the thalamus and basolateral amygdala (BLA) in mice after short-term administration (single 2g/kg dose) of alcohol or saline using immunohistochemistry and analyzed by 2-way analysis of variance. In the case-control association study using an NIAAA discovery sample, 8 SNPs in PIP5K1C were significantly associated with AUD in the African ancestry (AA) group (p<0.05 after correction; rs4807493, rs10405681, rs2074957, rs10432303, rs8109485, rs1476592, rs10419980, and rs4432372). However, a replication analysis using an independent sample (N=3,801) found no significant associations after correction for multiple testing. In the humanized transgenic mouse model with the OPRM1 polymorphism, PIP5K1C expression was significantly different between alcohol and saline-treated mice, regardless of genotype, in both the thalamus (p<0.05) and BLA (p<0.01). Our discovery sample shows that genetic variants in PIP5K1C are associated with AUD in the AA group, and acute alcohol exposure leads to up-regulation of PIP5K1C, potentially explaining a mechanism underlying the increased risk for chronic pain conditions in individuals with AUD.