Genome-wide Association Studies in Ancestrally Diverse Populations: Opportunities, Methods, Pitfalls, and Recommendations.

Genome-wide association studies (GWASs) have focused primarily on populations of European descent, but it is essential that diverse populations become better represented. Increasing diversity among study participants will advance our understanding of genetic architecture in all populations and ensure that genetic research is broadly applicable. To facilitate and promote research in multi-ancestry and admixed cohorts, we outline key methodological considerations and highlight opportunities, challenges, solutions, and areas in need of development. Despite the perception that analyzing genetic data from diverse populations is difficult, it is scientifically and ethically imperative, and there is an expanding analytical toolbox to do it well.

Pattern Similarity Analyses of FrontoParietal Task Coding: Individual Variation and Genetic Influences.

Pattern similarity analyses are increasingly used to characterize coding properties of brain regions, but relatively few have focused on cognitive control processes in FrontoParietal regions. Here, we use the Human Connectome Project (HCP) N-back task functional magnetic resonance imaging (fMRI) dataset to examine individual differences and genetic influences on the coding of working memory load (0-back, 2-back) and perceptual category (Face, Place). Participants were grouped into 105 monozygotic twin, 78 dizygotic twin, 99 nontwin sibling, and 100 unrelated pairs. Activation pattern similarity was used to test the hypothesis that FrontoParietal regions would have higher similarity for same load conditions, while Visual regions would have higher similarity in same perceptual category conditions. Results confirmed this highly robust regional double dissociation in neural coding, which also predicted individual differences in behavioral performance. In pair-based analyses, anatomically selective genetic relatedness effects were observed: relatedness predicted greater activation pattern similarity in FrontoParietal only for load coding and in Visual only for perceptual coding. Further, in related pairs, the similarity of load coding in FrontoParietal regions was uniquely associated with behavioral performance. Together, these results highlight the power of task fMRI pattern similarity analyses for detecting key coding and heritability features of brain regions.

Principled distillation of UK Biobank phenotype data reveals underlying structure in human variation.

Data within biobanks capture broad yet detailed indices of human variation, but biobank-wide insights can be difficult to extract due to complexity and scale. Here, using large-scale factor analysis, we distill hundreds of variables (diagnoses, assessments and survey items) into 35 latent constructs, using data from unrelated individuals with predominantly estimated European genetic ancestry in UK Biobank. These factors recapitulate known disease classifications, disentangle elements of socioeconomic status, highlight the relevance of psychiatric constructs to health and improve measurement of pro-health behaviours. We go on to demonstrate the power of this approach to clarify genetic signal, enhance discovery and identify associations between underlying phenotypic structure and health outcomes. In building a deeper understanding of ways in which constructs such as socioeconomic status, trauma, or physical activity are structured in the dataset, we emphasize the importance of considering the interwoven nature of the human phenome when evaluating public health patterns.

Patterns of item nonresponse behaviour to survey questionnaires are systematic and associated with genetic loci.

Response to survey questionnaires is vital for social and behavioural research, and most analyses assume full and accurate response by participants. However, nonresponse is common and impedes proper interpretation and generalizability of results. We examined item nonresponse behaviour across 109 questionnaire items in the UK Biobank (N = 360,628). Phenotypic factor scores for two participant-selected nonresponse answers, 'Prefer not to answer' (PNA) and 'I don't know' (IDK), each predicted participant nonresponse in follow-up surveys (incremental pseudo-R2 = 0.056), even when controlling for education and self-reported health (incremental pseudo-R2 = 0.046). After performing genome-wide association studies of our factors, PNA and IDK were highly genetically correlated with one another (rg = 0.73 (s.e. = 0.03)) and with education (rg,PNA = -0.51 (s.e. = 0.03); rg,IDK = -0.38 (s.e. = 0.02)), health (rg,PNA = 0.51 (s.e. = 0.03); rg,IDK = 0.49 (s.e. = 0.02)) and income (rg,PNA = -0.57 (s.e. = 0.04); rg,IDK = -0.46 (s.e. = 0.02)), with additional unique genetic associations observed for both PNA and IDK (P < 5 × 10-8). We discuss how these associations may bias studies of traits correlated with item nonresponse and demonstrate how this bias may substantially affect genome-wide association studies. While the UK Biobank data are deidentified, we further protected participant privacy by avoiding exploring non-response behaviour to single questions, assuring that no information can be used to associate results with any particular respondents.

Developmental Variability in Autism Across 17 000 Autistic Individuals and 4000 Siblings Without an Autism Diagnosis: Comparisons by Cohort, Intellectual Disability, Genetic Etiology, and Age at Diagnosis.

Importance: Presence of developmental delays in autism is well established, yet few studies have characterized variability in developmental milestone attainment in this population. Objective: To characterize variability in the age at which autistic individuals attain key developmental milestones based on co-occurring intellectual disability (ID), presence of a rare disruptive genetic variant associated with neurodevelopmental disorders (NDD), age at autism diagnosis, and research cohort membership. Design: The study team harmonized data from 4 cross-sectional autism cohorts: the Autism Genetics Research Exchange (n = 3284; 1997-2015), The Autism Simplex Collection (n = 694; 2008-2011), the Simons Simplex Collection (n = 2753; 2008-2011), and the Simons Foundation Powering Autism Research for Knowledge (n = 10 367; 2016-present). The last sample further included 4145 siblings without an autism diagnosis or ID. Participants: Convenience sample of 21 243 autistic individuals or their siblings without an autism diagnosis aged 4 to 17 years. Main Outcomes and Measures: Parents reported ages at which participants attained key milestones including smiling, sitting upright, crawling, walking, spoon-feeding self, speaking words, speaking phrases, and acquiring bladder and bowel control. A total of 5295 autistic individuals, and their biological parents, were genetically characterized to identify de novo variants in NDD-associated genes. The study team conducted time-to-event analyses to estimate and compare percentiles in time with milestone attainment across autistic individuals, subgroups of autistic individuals, and the sibling sample. Results: Seventeen thousand ninety-eight autistic individuals (mean age, 9.15 years; 80.8% male) compared with 4145 siblings without autism or ID (mean age, 10.2 years; 50.2% female) showed delays in milestone attainment, with median (IQR) delays ranging from 0.7 (0.3-1.6) to 19.7 (11.4-32.2) months. More severe and more variable delays in autism were associated with the presence of co-occurring ID, carrying an NDD-associated rare genetic variant, and being diagnosed with autism by age 5 years. More severe and more variable delays were also associated with membership in earlier study cohorts, consistent with autism's diagnostic and ascertainment expansion over the last 30 years. Conclusions and Relevance: As the largest summary to date of developmental milestone attainment in autism, to our knowledge, this study demonstrates substantial developmental variability across different conditions and provides important context for understanding the phenotypic and etiological heterogeneity of autism.

The female protective effect against autism spectrum disorder.

Autism spectrum disorder (ASD) is diagnosed three to four times more frequently in males than in females. Genetic studies of rare variants support a female protective effect (FPE) against ASD. However, sex differences in common inherited genetic risk for ASD are less studied, particularly within families. Leveraging the Danish iPSYCH resource, we found siblings of female ASD cases (n = 1,707) had higher rates of ASD than siblings of male ASD cases (n = 6,270; p < 1.0 × 10-10). In the Simons Simplex and SPARK collections, mothers of ASD cases (n = 7,436) carried more polygenic risk for ASD than fathers of ASD cases (n = 5,926; 0.08 polygenic risk score [PRS] SD; p = 7.0 × 10-7). Further, male unaffected siblings under-inherited polygenic risk (n = 1,519; p = 0.03). Using both epidemiologic and genetic approaches, our findings strongly support an FPE against ASD's common inherited influences.

A phenotypic spectrum of autism is attributable to the combined effects of rare variants, polygenic risk and sex.

The genetic etiology of autism spectrum disorder (ASD) is multifactorial, but how combinations of genetic factors determine risk is unclear. In a large family sample, we show that genetic loads of rare and polygenic risk are inversely correlated in cases and greater in females than in males, consistent with a liability threshold that differs by sex. De novo mutations (DNMs), rare inherited variants and polygenic scores were associated with various dimensions of symptom severity in children and parents. Parental age effects on risk for ASD in offspring were attributable to a combination of genetic mechanisms, including DNMs that accumulate in the paternal germline and inherited risk that influences behavior in parents. Genes implicated by rare variants were enriched in excitatory and inhibitory neurons compared with genes implicated by common variants. Our results suggest that a phenotypic spectrum of ASD is attributable to a spectrum of genetic factors that impact different neurodevelopmental processes.

Statistical and functional convergence of common and rare genetic influences on autism at chromosome 16p.

The canonical paradigm for converting genetic association to mechanism involves iteratively mapping individual associations to the proximal genes through which they act. In contrast, in the present study we demonstrate the feasibility of extracting biological insights from a very large region of the genome and leverage this strategy to study the genetic influences on autism. Using a new statistical approach, we identified the 33-Mb p-arm of chromosome 16 (16p) as harboring the greatest excess of autism's common polygenic influences. The region also includes the mechanistically cryptic and autism-associated 16p11.2 copy number variant. Analysis of RNA-sequencing data revealed that both the common polygenic influences within 16p and the 16p11.2 deletion were associated with decreased average gene expression across 16p. The transcriptional effects of the rare deletion and diffuse common variation were correlated at the level of individual genes and analysis of Hi-C data revealed patterns of chromatin contact that may explain this transcriptional convergence. These results reflect a new approach for extracting biological insight from genetic association data and suggest convergence of common and rare genetic influences on autism at 16p.

Identification of shared and differentiating genetic architecture for autism spectrum disorder, attention-deficit hyperactivity disorder and case subgroups.

Attention-deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are highly heritable neurodevelopmental conditions, with considerable overlap in their genetic etiology. We dissected their shared and distinct genetic etiology by cross-disorder analyses of large datasets. We identified seven loci shared by the disorders and five loci differentiating them. All five differentiating loci showed opposite allelic directions in the two disorders and significant associations with other traits, including educational attainment, neuroticism and regional brain volume. Integration with brain transcriptome data enabled us to identify and prioritize several significantly associated genes. The shared genomic fraction contributing to both disorders was strongly correlated with other psychiatric phenotypes, whereas the differentiating portion was correlated most strongly with cognitive traits. Additional analyses revealed that individuals diagnosed with both ASD and ADHD were double-loaded with genetic predispositions for both disorders and showed distinctive patterns of genetic association with other traits compared with the ASD-only and ADHD-only subgroups. These results provide insights into the biological foundation of the development of one or both conditions and of the factors driving psychopathology discriminatively toward either ADHD or ASD.

The impact of non-additive genetic associations on age-related complex diseases.

Genome-wide association studies (GWAS) are not fully comprehensive, as current strategies typically test only the additive model, exclude the X chromosome, and use only one reference panel for genotype imputation. We implement an extensive GWAS strategy, GUIDANCE, which improves genotype imputation by using multiple reference panels and includes the analysis of the X chromosome and non-additive models to test for association. We apply this methodology to 62,281 subjects across 22 age-related diseases and identify 94 genome-wide associated loci, including 26 previously unreported. Moreover, we observe that 27.7% of the 94 loci are missed if we use standard imputation strategies with a single reference panel, such as HRC, and only test the additive model. Among the new findings, we identify three novel low-frequency recessive variants with odds ratios larger than 4, which need at least a three-fold larger sample size to be detected under the additive model. This study highlights the benefits of applying innovative strategies to better uncover the genetic architecture of complex diseases.

Genetic analyses identify widespread sex-differential participation bias.

Genetic association results are often interpreted with the assumption that study participation does not affect downstream analyses. Understanding the genetic basis of participation bias is challenging since it requires the genotypes of unseen individuals. Here we demonstrate that it is possible to estimate comparative biases by performing a genome-wide association study contrasting one subgroup versus another. For example, we showed that sex exhibits artifactual autosomal heritability in the presence of sex-differential participation bias. By performing a genome-wide association study of sex in approximately 3.3 million males and females, we identified over 158 autosomal loci spuriously associated with sex and highlighted complex traits underpinning differences in study participation between the sexes. For example, the body mass index-increasing allele at FTO was observed at higher frequency in males compared to females (odds ratio = 1.02, P = 4.4 × 10-36). Finally, we demonstrated how these biases can potentially lead to incorrect inferences in downstream analyses and propose a conceptual framework for addressing such biases. Our findings highlight a new challenge that genetic studies may face as sample sizes continue to grow.

Genome-wide association studies of alcohol dependence, DSM-IV criterion count and individual criteria.

Genome-wide association studies (GWAS) of alcohol dependence (AD) have reliably identified variation within alcohol metabolizing genes (eg, ADH1B) but have inconsistently located other signals, which may be partially attributable to symptom heterogeneity underlying the disorder. We conducted GWAS of DSM-IV AD (primary analysis), DSM-IV AD criterion count (secondary analysis), and individual dependence criteria (tertiary analysis) among 7418 (1121 families) European American (EA) individuals from the Collaborative Study on the Genetics of Alcoholism (COGA). Trans-ancestral meta-analyses combined these results with data from 3175 (585 families) African-American (AA) individuals from COGA. In the EA GWAS, three loci were genome-wide significant: rs1229984 in ADH1B for AD criterion count (P = 4.16E-11) and Desire to cut drinking (P = 1.21E-11); rs188227250 (chromosome 8, Drinking more than intended, P = 6.72E-09); rs1912461 (chromosome 15, Time spent drinking, P = 1.77E-08). In the trans-ancestral meta-analysis, rs1229984 was associated with multiple phenotypes and two additional loci were genome-wide significant: rs61826952 (chromosome 1, DSM-IV AD, P = 8.42E-11); rs7597960 (chromosome 2, Time spent drinking, P = 1.22E-08). Associations with rs1229984 and rs18822750 were replicated in independent datasets. Polygenic risk scores derived from the EA GWAS of AD predicted AD in two EA datasets (P < .01; 0.61%-1.82% of variance). Identified novel variants (ie, rs1912461, rs61826952) were associated with differential central evoked theta power (loss - gain; P = .0037) and reward-related ventral striatum reactivity (P = .008), respectively. This study suggests that studying individual criteria may unveil new insights into the genetic etiology of AD liability.

Reward-related ventral striatum activity links polygenic risk for attention-deficit/hyperactivity disorder to problematic alcohol use in young adulthood.

BACKGROUND: Problematic alcohol use in adolescence and adulthood is a common and often debilitating correlate of childhood attention-deficit/hyperactivity disorder (ADHD). Converging evidence suggests that ADHD and problematic alcohol use share a common additive genetic basis, which may be mechanistically related to reward-related brain function. In the current study, we examined whether polygenic risk for childhood ADHD is linked to problematic alcohol use in young adulthood through alterations in reward-related activity of the ventral striatum, a neural hub supporting appetitive behaviors and reinforcement learning. METHODS: Genomic, neuroimaging, and self-report data were available for 404 non-Hispanic European-American participants who completed the ongoing Duke Neurogenetics Study. Polygenic risk scores for childhood ADHD were calculated based on a genome-wide association study meta-analysis conducted by the Psychiatric Genomics Consortium and tested for association with reward-related ventral striatum activity, measured using a number-guessing functional magnetic resonance imaging paradigm, and self-reported problematic alcohol use. A mediational model tested whether ventral striatum activity indirectly links polygenic risk for ADHD to problematic alcohol use. RESULTS: Despite having no main effect on problematic alcohol use, polygenic risk for childhood ADHD was indirectly associated with problematic alcohol use through increased reward-related ventral striatum activity. CONCLUSIONS: Individual differences in reward-related brain function may, at least in part, mechanistically link polygenic risk for childhood ADHD to problematic alcohol use.

Hypothalamic-pituitary-adrenal axis genetic variation and early stress moderates amygdala function.

Early life stress may precipitate psychopathology, at least in part, by influencing amygdala function. Converging evidence across species suggests that links between childhood stress and amygdala function may be dependent upon hypothalamic-pituitary-adrenal (HPA) axis function. Using data from college-attending non-Hispanic European-Americans (n=308) who completed the Duke Neurogenetics Study, we examined whether early life stress (ELS) and HPA axis genetic variation interact to predict threat-related amygdala function as well as psychopathology symptoms. A biologically-informed multilocus profile score (BIMPS) captured HPA axis genetic variation (FKBP5 rs1360780, CRHR1 rs110402; NR3C2 rs5522/rs4635799) previously associated with its function (higher BIMPS are reflective of higher HPA axis activity). BOLD fMRI data were acquired while participants completed an emotional face matching task. ELS and depression and anxiety symptoms were measured using the childhood trauma questionnaire and the mood and anxiety symptom questionnaire, respectively. The interaction between HPA axis BIMPS and ELS was associated with right amygdala reactivity to threat-related stimuli, after accounting for multiple testing (empirical-p=0.016). Among individuals with higher BIMPS (i.e., the upper 21.4%), ELS was positively coupled with threat-related amygdala reactivity, which was absent among those with average or low BIMPS. Further, higher BIMPS were associated with greater self-reported anxious arousal, though there was no evidence that amygdala function mediated this relationship. Polygenic variation linked to HPA axis function may moderate the effects of early life stress on threat-related amygdala function and confer risk for anxiety symptomatology. However, what, if any, neural mechanisms may mediate the relationship between HPA axis BIMPS and anxiety symptomatology remains unclear.

Association Between Substance Use Disorder and Polygenic Liability to Schizophrenia.

BACKGROUND: There are high levels of comorbidity between schizophrenia and substance use disorder, but little is known about the genetic etiology of this comorbidity. METHODS: We tested the hypothesis that shared genetic liability contributes to the high rates of comorbidity between schizophrenia and substance use disorder. To do this, polygenic risk scores for schizophrenia derived from a large meta-analysis by the Psychiatric Genomics Consortium were computed in three substance use disorder datasets: the Collaborative Genetic Study of Nicotine Dependence (ascertained for tobacco use disorder; n = 918 cases; 988 control subjects), the Collaborative Study on the Genetics of Alcoholism (ascertained for alcohol use disorder; n = 643 cases; 384 control subjects), and the Family Study of Cocaine Dependence (ascertained for cocaine use disorder; n = 210 cases; 317 control subjects). Phenotypes were harmonized across the three datasets and standardized analyses were performed. Genome-wide genotypes were imputed to the 1000 Genomes reference panel. RESULTS: In each individual dataset and in the mega-analysis, strong associations were observed between any substance use disorder diagnosis and the polygenic risk score for schizophrenia (mega-analysis pseudo-R2 range 0.8-3.7%; minimum p = 4 × 10-23). CONCLUSIONS: These results suggest that comorbidity between schizophrenia and substance use disorder is partially attributable to shared polygenic liability. This shared liability is most consistent with a general risk for substance use disorder rather than specific risks for individual substance use disorders and adds to increasing evidence of a blurred boundary between schizophrenia and substance use disorder.

Imaging Genetics and Genomics in Psychiatry: A Critical Review of Progress and Potential.

Imaging genetics and genomics research has begun to provide insight into the molecular and genetic architecture of neural phenotypes and the neural mechanisms through which genetic risk for psychopathology may emerge. As it approaches its third decade, imaging genetics is confronted by many challenges, including the proliferation of studies using small sample sizes and diverse designs, limited replication, problems with harmonization of neural phenotypes for meta-analysis, unclear mechanisms, and evidence that effect sizes may be more modest than originally posited, with increasing evidence of polygenicity. These concerns have encouraged the field to grow in many new directions, including the development of consortia and large-scale data collection projects and the use of novel methods (e.g., polygenic approaches, machine learning) that enhance the quality of imaging genetic studies but also introduce new challenges. We critically review progress in imaging genetics and offer suggestions and highlight potential pitfalls of novel approaches. Ultimately, the strength of imaging genetics and genomics lies in their translational and integrative potential with other research approaches (e.g., nonhuman animal models, psychiatric genetics, pharmacologic challenge) to elucidate brain-based pathways that give rise to the vast individual differences in behavior as well as risk for psychopathology.

Associations between Polygenic Risk for Psychiatric Disorders and Substance Involvement.

Despite evidence of substantial comorbidity between psychiatric disorders and substance involvement, the extent to which common genetic factors contribute to their co-occurrence remains understudied. In the current study, we tested for associations between polygenic risk for psychiatric disorders and substance involvement (i.e., ranging from ever-use to severe dependence) among 2573 non-Hispanic European-American participants from the Study of Addiction: Genetics and Environment. Polygenic risk scores (PRS) for cross-disorder psychopathology (CROSS) were generated based on the Psychiatric Genomics Consortium's Cross-Disorder meta-analysis and then tested for associations with a factor representing general liability to alcohol, cannabis, cocaine, nicotine, and opioid involvement (GENSUB). Follow-up analyses evaluated specific associations between each of the five psychiatric disorders which comprised CROSS-attention deficit hyperactivity disorder (ADHD), autism spectrum disorder (AUT), bipolar disorder (BIP), major depressive disorder (MDD), and schizophrenia (SCZ)-and involvement with each component substance included in GENSUB. CROSS PRS explained 1.10% of variance in GENSUB in our sample (p < 0.001). After correction for multiple testing in our follow-up analyses of polygenic risk for each individual disorder predicting involvement with each component substance, associations remained between: (A) MDD PRS and non-problem cannabis use, (B) MDD PRS and severe cocaine dependence, (C) SCZ PRS and non-problem cannabis use and severe cannabis dependence, and (D) SCZ PRS and severe cocaine dependence. These results suggest that shared covariance from common genetic variation contributes to psychiatric and substance involvement comorbidity.

PER1 rs3027172 Genotype Interacts with Early Life Stress to Predict Problematic Alcohol Use, but Not Reward-Related Ventral Striatum Activity.

Increasing evidence suggests that the circadian and stress regulatory systems contribute to alcohol use disorder (AUD) risk, which may partially arise through effects on reward-related neural function. The C allele of the PER1 rs3027172 single nucleotide polymorphism (SNP) reduces PER1 expression in cells incubated with cortisol and has been associated with increased risk for adult AUD and problematic drinking among adolescents exposed to high levels of familial psychosocial adversity. Using data from undergraduate students who completed the ongoing Duke Neurogenetics Study (DNS) (n = 665), we tested whether exposure to early life stress (ELS; Childhood Trauma Questionnaire) moderates the association between rs3027172 genotype and later problematic alcohol use (Alcohol Use Disorders Identification Test) as well as ventral striatum (VS) reactivity to reward (card-guessing task while functional magnetic resonance imaging data were acquired). Initial analyses found that PER1 rs3027172 genotype interacted with ELS to predict both problematic drinking and VS reactivity; minor C allele carriers, who were also exposed to elevated ELS reported greater problematic drinking and exhibited greater ventral striatum reactivity to reward-related stimuli. When gene × covariate and environment × covariate interactions were controlled for, the interaction predicting problematic alcohol use remained significant (p < 0.05, corrected) while the interaction predicting VS reactivity was no longer significant. These results extend our understanding of relationships between PER1 genotype, ELS, and problematic alcohol use, and serve as a cautionary tale on the importance of controlling for potential confounders in studies of moderation including gene × environment interactions.

Monoacylglycerol lipase (MGLL) polymorphism rs604300 interacts with childhood adversity to predict cannabis dependence symptoms and amygdala habituation: Evidence from an endocannabinoid system-level analysis.

Despite evidence for heritable variation in cannabis involvement and the discovery of cannabinoid receptors and their endogenous ligands, no consistent patterns have emerged from candidate endocannabinoid (eCB) genetic association studies of cannabis involvement. Given interactions between eCB and stress systems and associations between childhood stress and cannabis involvement, it may be important to consider childhood adversity in the context of eCB-related genetic variation. We employed a system-level gene-based analysis of data from the Comorbidity and Trauma Study (N = 1,558) to examine whether genetic variation in six eCB genes (anabolism: DAGLA, DAGLB, NAPEPLD; catabolism: MGLL, FAAH; binding: CNR1; SNPs N = 65) and childhood sexual abuse (CSA) predict cannabis dependence symptoms. Significant interactions with CSA emerged for MGLL at the gene level (p = .009), and for rs604300 within MGLL (ΔR2 = .007, p < .001), the latter of which survived SNP-level Bonferroni correction and was significant in an additional sample with similar directional effects (N = 859; ΔR2 = .005, p = .026). Furthermore, in a third sample (N = 312), there was evidence that rs604300 genotype interacts with early life adversity to predict threat-related basolateral amygdala habituation, a neural phenotype linked to the eCB system and addiction (ΔR2 = .013, p = .047). Rs604300 may be related to epigenetic modulation of MGLL expression. These results are consistent with rodent models implicating 2-arachidonoylglycerol (2-AG), an endogenous cannabinoid metabolized by the enzyme encoded by MGLL, in the etiology of stress adaptation related to cannabis dependence, but require further replication.