Genotype x adiposity interaction linkage analyses reveal a locus on chromosome 1 for lipoprotein-associated phospholipase A2, a marker of inflammation and oxidative stress

Genotype-by-socioeconomic status interaction influences heart disease risk scores and carotid artery thickness in Mexican Americans: the predominant role of education in comparison to household income and socioeconomic index

Background: Socioeconomic status (SES) is a potent environmental determinant of health. To our knowledge, no assessment of genotype-environment interaction has been conducted to consider the joint effects of socioeconomic status and genetics on risk for cardiovascular disease (CVD). We analyzed Mexican American Family Studies (MAFS) data to evaluate the hypothesis that genotype-by-environment interaction (GxE) is an important determinant of variation in CVD risk factors. Methods: We employed a linear mixed model to investigate GxE in Mexican American extended families. We studied two proxies for CVD [Pooled Cohort Equation Risk Scores/Framingham Risk Scores (FRS/PCRS) and carotid artery intima-media thickness (CA-IMT)] in relation to socioeconomic status as determined by Duncan's Socioeconomic Index (SEI), years of education, and household income. Results: We calculated heritability for FRS/PCRS and carotid artery intima-media thickness. There was evidence of GxE due to additive genetic variance heterogeneity and genetic correlation for FRS, PCRS, and CA-IMT measures for education (environment) but not for household income or SEI. Conclusion: The genetic effects underlying CVD are dynamically modulated at the lower end of the SES spectrum. There is a significant change in the genetic architecture underlying the major components of CVD in response to changes in education.

Identification of Genes Associated with the Impairment of Olfactory and Gustatory Functions in COVID-19 via Machine-Learning Methods

The coronavirus disease 2019 (COVID-19), as a severe respiratory disease, affects many parts of the body, and approximately 20-85% of patients exhibit functional impairment of the senses of smell and taste, some of whom even experience the permanent loss of these senses. These symptoms are not life-threatening but severely affect patients' quality of life and increase the risk of depression and anxiety. The pathological mechanisms of these symptoms have not been fully identified. In the current study, we aimed to identify the important biomarkers at the expression level associated with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection-mediated loss of taste or olfactory ability, and we have suggested the potential pathogenetic mechanisms of COVID-19 complications. We designed a machine-learning-based approach to analyze the transcriptome of 577 COVID-19 patient samples, including 84 COVID-19 samples with a decreased ability to taste or smell and 493 COVID-19 samples without impairment. Each sample was represented by 58,929 gene expression levels. The features were analyzed and sorted by three feature selection methods (least absolute shrinkage and selection operator, light gradient boosting machine, and Monte Carlo feature selection). The optimal feature sets were obtained through incremental feature selection using two classification algorithms: decision tree (DT) and random forest (RF). The top genes identified by these multiple methods (H3-5, NUDT5, and AOC1) are involved in olfactory and gustatory impairments. Meanwhile, a high-performance RF classifier was developed in this study, and three sets of quantitative rules that describe the impairment of olfactory and gustatory functions were obtained based on the optimal DT classifiers. In summary, this study provides a new computation analysis and suggests the latent biomarkers (genes and rules) for predicting olfactory and gustatory impairment caused by COVID-19 complications.

A Robust Method Uncovers Significant Context-Specific Heritability in Diverse Complex Traits

Gene-environment interactions (GxE) can be fundamental in applications ranging from functional genomics to precision medicine and is a conjectured source of substantial heritability. However, unbiased methods to profile GxE genome-wide are nascent and, as we show, cannot accommodate general environment variables, modest sample sizes, heterogeneous noise, and binary traits. To address this gap, we propose a simple, unifying mixed model for gene-environment interaction (GxEMM). In simulations and theory, we show that GxEMM can dramatically improve estimates and eliminate false positives when the assumptions of existing methods fail. We apply GxEMM to a range of human and model organism datasets and find broad evidence of context-specific genetic effects, including GxSex, GxAdversity, and GxDisease interactions across thousands of clinical and molecular phenotypes. Overall, GxEMM is broadly applicable for testing and quantifying polygenic interactions, which can be useful for explaining heritability and invaluable for determining biologically relevant environments.

The genetic and environmental etiology of child maltreatment in a parent-based extended family design

Child maltreatment has been associated with various cumulative risk factors. However, little is known about the extent to which genetic and environmental factors contribute to individual differences between parents in perpetrating child maltreatment. To estimate the relative contribution of genetic and environmental factors to perpetrating maltreatment we used a parent-based extended family design. Child-reported perpetrated maltreatment was available for 556 parents (283 women) from 63 families. To explore reporter effects (i.e., child perspective on maltreatment), child reports were compared to multi-informant reports. Based on polygenic model analyses, most of the variance related to the perpetration of physical abuse and emotional neglect was explained by common environmental factors (physical abuse: c2 = 59%, SE = 12%, p = .006; emotional neglect: c2 = 47%, SE = 8%, p < .001) whereas genetic factors did not significantly contribute to the model. For perpetrated emotional abuse, in contrast, genetic factors did significantly contribute to perpetrated emotional abuse (h2 = 33%, SE = 8%, p < .001), whereas common environment factors did not. Multi-informant reports led to similar estimates of genetic and common environmental effects on all measures except for emotional abuse, where a multi-informant approach yielded higher estimates of the common environmental effects. Overall, estimates of unique environment, including measurement error, were lower using multi-informant reports. In conclusion, our findings suggest that genetic pathways play a significant role in perpetrating emotional abuse, while physical abuse and emotional neglect are transmitted primarily through common environmental factors. These findings imply that interventions may need to target different mechanisms dependings on maltreatment type.

Genetic and environmental (physical fitness and sedentary activity) interaction effects on cardiometabolic risk factors in Mexican American children and adolescents

Knowledge on genetic and environmental (G × E) interaction effects on cardiometabolic risk factors (CMRFs) in children is limited.  The purpose of this study was to examine the impact of G × E interaction effects on CMRFs in Mexican American (MA) children (n = 617, ages 6-17 years). The environments examined were sedentary activity (SA), assessed by recalls from "yesterday" (SAy) and "usually" (SAu) and physical fitness (PF) assessed by Harvard PF scores (HPFS). CMRF data included body mass index (BMI), waist circumference (WC), fat mass (FM), fasting insulin (FI), homeostasis model of assessment-insulin resistance (HOMA-IR), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), systolic (SBP) and diastolic (DBP) blood pressure, and number of metabolic syndrome components (MSC). We examined potential G × E interaction in the phenotypic expression of CMRFs using variance component models and likelihood-based statistical inference. Significant G × SA interactions were identified for six CMRFs: BMI, WC, FI, HOMA-IR, MSC, and HDL, and significant G × HPFS interactions were observed for four CMRFs: BMI, WC, FM, and HOMA-IR. However, after correcting for multiple hypothesis testing, only WC × SAy, FM × SAy, and FI × SAu interactions became marginally significant. After correcting for multiple testing, most of CMRFs exhibited significant G × E interactions (Reduced G × E model vs. Constrained model). These findings provide evidence that genetic factors interact with SA and PF to influence variation in CMRFs, and underscore the need for better understanding of these relationships to develop strategies and interventions to effectively reduce or prevent cardiometabolic risk in children.

Bayesian Variable Selection in Multilevel Item Response Theory Models with Application in Genomics

The goal of this paper is to present an implementation of stochastic search variable selection (SSVS) to multilevel model from item response theory (IRT). As experimental settings get more complex and models are required to integrate multiple (and sometimes massive) sources of information, a model that can jointly summarize and select the most relevant characteristics can provide better interpretation and a deeper insight into the problem. A multilevel IRT model recently proposed in the literature for modeling multifactorial diseases is extended to perform variable selection in the presence of thousands of covariates using SSVS. We derive conditional distributions required for such a task as well as an acceptance-rejection step that allows for the SSVS in high dimensional settings using a Markov Chain Monte Carlo algorithm. We validate the variable selection procedure through simulation studies, and illustrate its application on a study with genetic markers associated with the metabolic syndrome.

Sex-specific genetic effects in physical activity: results from a quantitative genetic analysis

BACKGROUND

The objective of this study is to present a model to estimate sex-specific genetic effects on physical activity (PA) levels and sedentary behaviour (SB) using three generation families.

METHODS

The sample consisted of 100 families covering three generations from Portugal. PA and SB were assessed via the International Physical Activity Questionnaire short form (IPAQ-SF). Sex-specific effects were assessed by genotype-by-sex interaction (GSI) models and sex-specific heritabilities. GSI effects and heterogeneity were tested in the residual environmental variance. SPSS 17 and SOLAR v. 4.1 were used in all computations.

RESULTS

The genetic component for PA and SB domains varied from low to moderate (11% to 46%), when analyzing both genders combined. We found GSI effects for vigorous PA (p = 0.02) and time spent watching television (WT) (p < 0.001) that showed significantly higher additive genetic variance estimates in males. The heterogeneity in the residual environmental variance was significant for moderate PA (p = 0.02), vigorous PA (p = 0.006) and total PA (p = 0.001). Sex-specific heritability estimates were significantly higher in males only for WT, with a male-to-female difference in heritability of 42.5 (95% confidence interval: 6.4, 70.4).

CONCLUSIONS

Low to moderate genetic effects on PA and SB traits were found. Results from the GSI model show that there are sex-specific effects in two phenotypes, VPA and WT with a stronger genetic influence in males.

Genotype by sex and genotype by age interactions with sedentary behavior: the Portuguese Healthy Family Study

Sedentary behavior (SB) expression and its underlying causal factors have been progressively studied, as it is a major determinant of decreased health quality. In the present study we applied Genotype x Age (GxAge) and Genotype x Sex (GxSex) interaction methods to determine if the phenotypic expression of different SB traits is influenced by an interaction between genetic architecture and both age and sex. A total of 1345 subjects, comprising 249 fathers, 327 mothers, 334 sons and 325 daughters, from 339 families of The Portuguese Healthy Family Study were included in the analysis. SB traits were assessed by means of a 3-d physical activity recall, the Baecke and IPAQ questionnaires. GxAge and GxSex interactions were analyzed using SOLAR 4.0 software. Sedentary behaviour heritability estimates were not always statistically significant (p>0.05) and ranged from 3% to 27%. The GxSex and GxAge interaction models were significantly better than the single polygenic models for TV (min/day), EEsed (kcal/day), personal computer (PC) usage and physical activty (PA) tertiles. The GxAge model is also significantly better than the polygenic model for Sed (min/day). For EEsed, PA tertiles, PC and Sed, the GxAge interaction was significant because the genetic correlation between SB environments was significantly different from 1. Further, PC and Sed variance heterogeneity among distinct ages were observed. The GxSex interaction was significant for EEsed due to genetic variance heterogeneity between genders and for PC due to a genetic correlation less than 1 across both sexes. Our results suggest that SB expression may be influenced by the interactions between genotype with both sex and age. Further, different sedentary behaviors seem to have distinct genetic architectures and are differentially affected by age and sex.

Genotype by energy expenditure interaction and body composition traits: The Portuguese Healthy Family Study

Background and Aims. Energy expenditure has been negatively correlated with fat accumulation. However, this association is highly variable. In the present study we applied a genotype by environment interaction method to examine the presence of Genotype x by Total Daily Energy Expenditure and Genotype x by Daily Energy Expenditure interactions in the expression of different body composition traits. Methods and Results. A total of 958 subjects from 294 families of The Portuguese Healthy Family Study were included in the analysis. TDEE and DEE were assessed using a physical activity recall. Body fat percentages were measured with a bioelectrical impedance scale. GxTDEE and GxDEE examinations were performed using SOLAR 4.0 software. All BC traits were significantly heritable, with heritabilities ranging from 21% to 34%. The GxTDEE and GxDEE interaction models fitted the data better than the polygenic model for all traits. For all traits, a significant GxTDEE and GxDEE interaction was due to variance heterogeneity among distinct levels of TDEE and DEE. For WC, GxTDEE was also significant due to the genetic correlation function. Conclusions. TDEE and DEE are environmental constraints associated with the expression of individuals' BC genotypes, leading to variability in the phenotypic expression of BC traits.

Genotype by energy expenditure interaction with metabolic syndrome traits: the Portuguese healthy family study

Moderate-to-high levels of physical activity are established as preventive factors in metabolic syndrome development. However, there is variability in the phenotypic expression of metabolic syndrome under distinct physical activity conditions. In the present study we applied a Genotype X Environment interaction method to examine the presence of GxEE interaction in the phenotypic expression of metabolic syndrome. A total of 958 subjects, from 294 families of The Portuguese Healthy Family study, were included in the analysis. Total daily energy expenditure was assessed using a 3 day physical activity diary. Six metabolic syndrome related traits, including waist circumference, systolic blood pressure, glucose, HDL cholesterol, total cholesterol and triglycerides, were measured and adjusted for age and sex. GxEE examination was performed on SOLAR 4.3.1. All metabolic syndrome indicators were significantly heritable. The GxEE interaction model fitted the data better than the polygenic model (p<0.001) for waist circumference, systolic blood pressure, glucose, total cholesterol and triglycerides. For waist circumference, glucose, total cholesterol and triglycerides, the significant GxEE interaction was due to rejection of the variance homogeneity hypothesis. For waist circumference and glucose, GxEE was also significant by the rejection of the genetic correlation hypothesis. The results showed that metabolic syndrome traits expression is significantly influenced by the interaction established between total daily energy expenditure and genotypes. Physical activity may be considered an environmental variable that promotes metabolic differences between individuals that are distinctively active.

Clustering of body composition, blood pressure and physical activity in Portuguese families

AIM

The purposes of this study were: (i) to identify familial resemblances in body fat, blood pressure (BP) and total physical activity (TPA); (ii) to estimate the magnitude of their genetic and environmental influences; and (iii) to investigate shared familial aggregation among these phenotypes.

SUBJECTS AND METHODS

The sample comprised 260 nuclear families from Portugal. Body fat was assessed by bioelectrical impedance. BP was measured by an oscillometric device. TPA was estimated by the Baecke questionnaire. Familial correlation analyses were performed using Generalized Estimating Equations. Quantitative genetic modelling was used to estimate maximal heritability, genetic and environmental correlations.

RESULTS

Familial intra-trait correlations ranged from 0.15-0.38. Genetic and common environmental factors explained from 30%--44% of fat mass depots and BP and 24% of TPA. Genetic correlations were significant between BP and the fat mass traits (p < 0.05). Environmental correlations were statistically significant between diastolic BP and total body fat, trunk fat and arm fat (p < 0.05) and TPA and other phenotypes.

CONCLUSIONS

The results suggest familial resemblance in the variation of body fat, BP and TPA, showing partial pleiotropic effects in the variation in body fat phenotypes and BP. TPA only shares common environmental influences with BP and body fat traits.

Significant genotype by diet (G × D) interaction effects on cardiometabolic responses to a pedigree-wide, dietary challenge in vervet monkeys (Chlorocebus aethiops sabaeus)

Nutrient composition of a diet (D) has been shown to interact with genetic predispositions (G) to affect various lipid phenotypes. Our aim in this study was to confirm G × D interaction and determine whether the interaction extends to other cardiometabolic risk factors such as glycemic measures and body weight. Subjects were vervet monkeys (Chlorocebus aethiops sabaeus; n = 309) from a multigenerational pedigreed colony initially fed with a plant-based diet, standard primate diet (18% calories from protein, 13% from fat, and 69% from carbohydrates), and subsequently challenged for 8 weeks with a diet modeled on the typical American diet (18% calories from protein, 35% from fat, and 47% from carbohydrates). Our results showed that although exposure to the challenge diet did not result in significant changes in weight, most lipid and glycemic biomarkers moved in an adverse direction (P < 0.01). Quantitative genetic analyses showed that cardiometabolic phenotypes were significantly heritable under both dietary conditions (P < 0.05), and there was significant evidence of G × D interaction for these phenotypes. We observed significant differences in the additive genetic variances for most lipid phenotypes (P < 10(-4) ), indicating that the magnitude of genetic effects varies by diet. Furthermore, genetic correlations between diets differed significantly from 1 with respect to insulin, body weight, and some lipid phenotypes (P < 0.01). This implied that distinct genetic effects are involved in the regulation of these phenotypes under the two dietary conditions. These G × D effects confirm and extend previous observations in baboons (Papio sp.) and suggest that mimicking the typical human nutritional environment can reveal genetic influences that might not be observed in animals consuming standard, plant-based diets.

Sex-specific genetic architecture of human fatness in Chinese: the SAPPHIRe Study

To dissect the genetic architecture of sexual dimorphism in obesity-related traits, we evaluated the sex-genotype interaction, sex-specific heritability and genome-wide linkages for seven measurements related to obesity. A total of 1,365 non-diabetic Chinese subjects from the family study of the Stanford Asia-Pacific Program of Hypertension and Insulin Resistance were used to search for quantitative trait loci (QTLs) responsible for the obesity-related traits. Pleiotropy and co-incidence effects from the QTLs were also examined using the bivariate linkage approach. We found that sex-specific differences in heritability and the genotype-sex interaction effects were substantially significant for most of these traits. Several QTLs with strong linkage evidence were identified after incorporating genotype by sex (G × S) interactions into the linkage mapping, including one QTL for hip circumference [maximum LOD score (MLS) = 4.22, empirical p = 0.000033] and two QTLs: for BMI on chromosome 12q with MLS 3.37 (empirical p = 0.0043) and 3.10 (empirical p = 0.0054). Sex-specific analyses demonstrated that these linkage signals all resulted from females rather than males. Most of these QTLs for obesity-related traits replicated the findings in other ethnic groups. Bivariate linkage analyses showed several obesity traits were influenced by a common set of QTLs. All regions with linkage signals were observed in one gender, but not in the whole sample, suggesting the genetic architecture of obesity-related traits does differ by gender. These findings are useful for further identification of the liability genes for these phenotypes through candidate genes or genome-wide association analysis.

Mapping quantitative traits in unselected families: algorithms and examples

Linkage analysis has been widely used to identify from family data genetic variants influencing quantitative traits. Common approaches have both strengths and limitations. Likelihood ratio tests typically computed in variance component analysis can accommodate large families but are highly sensitive to departure from normality assumptions. Regression-based approaches are more robust but their use has primarily been restricted to nuclear families. In this paper, we develop methods for mapping quantitative traits in moderately large pedigrees. Our methods are based on the score statistic, which in contrast to the likelihood ratio statistic can use nonparametric estimators of variability to achieve robustness of the false-positive rate against departures from the hypothesized phenotypic model. Because the score statistic is easier to calculate than the likelihood ratio statistic, our basic mapping methods utilize relatively simple computer code that performs statistical analysis on output from any program that computes estimates of identity by descent. This simplicity also permits development and evaluation of methods to deal with multivariate and ordinal phenotypes, and with gene-gene and gene-environment interaction. We demonstrate our methods on simulated data and on fasting insulin, a quantitative trait measured in the Framingham Heart Study.

Genotype-phenotype associations: modulation by diet and obesity

Changes in diet are likely to reduce chronic disorders, but after decades of active research and heated discussion, the question still remains: what is the optimal diet to achieve this elusive goal? Is it a low-fat diet, as traditionally recommended by multiple medical societies? Or a high monounsaturated fat (MUFA) diet as predicated by the Mediterranean diet? Perhaps a high polyunsaturated fat (PUFA) diet based on the cholesterol-lowering effects? The right answer may be all of the above but not for everybody. A well-known phenomenon in nutrition research and practice is the dramatic variability in interindividual response to any type of dietary intervention. There are many other factors influencing response, and they include, among many others, age, sex, physical activity, alcohol, and smoking as well as genetic factors that will help to identify vulnerable populations/individuals that will benefit from a variety of more personalized and mechanistic-based dietary recommendations. This potential could and needs to be developed within the context of nutritional genomics that in conjunction with systems biology may provide the tools to achieve the holy grail of dietary prevention and therapy of chronic diseases and cancer. This approach will break with the traditional public health approach of "one size fits all." The current evidence based on nutrigenetics has begun to identify subgroups of individuals who benefit more from a low-fat diet, whereas others appear to benefit more from high MUFA or PUFA diets. The continuous progress in nutrigenomics will allow some time in the future to provide targeted gene-based dietary advice.

Clinical and genetic factors associated with lipoprotein-associated phospholipase A2 in the Framingham Heart Study

OBJECTIVE

To conduct an investigation of clinical and genetic correlates of lipoprotein-associated phospholipase (Lp-PLA(2)) activity and mass in a large community-based cohort. Higher circulating Lp-PLA(2) predicts cardiovascular disease risk, but sources of inter-individual variability are incompletely understood.

METHODS

We conducted stepwise regression of clinical correlates of Lp-PLA(2) in four Framingham Heart Study cohorts (n=8185; mean age 50+/-14 years, 53.8% women, 9.8% ethnic/racial minority cohort). We also conducted heritability and linkage analyses in Offspring and Generation 3 cohorts (n=6945). In Offspring cohort participants we performed association analyses (n=1535 unrelated) with 1943 common tagging SNPs in 233 inflammatory candidate genes.

RESULTS

Sixteen clinical variables explained 57% of the variability in Lp-PLA(2) activity; covariates associated with Lp-PLA(2) mass were similar but only explained 27% of the variability. Multivariable-adjusted heritability estimates for Lp-PLA(2) activity and mass were 41% and 25%, respectively. A linkage peak was observed for Lp-PLA(2) activity (chromosome 6, LOD score 2.4). None of the SNPs achieved experiment-wide statistical significance, though 12 had q values <0.50, and hence we expect at least 50% of these associations to be true positives. The strongest multivariable-association with Lp-PLA(2) activity was found for MEF2A (rs2033547; nominal p=3.20 x 10(-4)); SNP rs1051931 in PLA2G7 was nominally associated (p=1.26 x 10(-3)). The most significant association to Lp-PLA(2) mass was in VEGFC (rs10520358, p=9.14 x 10(-4)).

CONCLUSIONS

Cardiovascular risk factors and genetic variation contribute to variability in Lp-PLA(2) activity and mass. Our genetic association analyses need replication, which will be facilitated by web posting of our genetic association results.

A pleiotropic QTL on 2p influences serum Lp-PLA2 activity and LDL cholesterol concentration in a baboon model for the genetics of atherosclerosis risk factors

Lipoprotein-associated phospholipase A(2) (Lp-PLA(2)), the major portion of which is bound to low-density lipoprotein, is an independent biomarker of cardiovascular disease risk. To search for common genetic determinants of variation in both Lp-PLA(2) activity and LDL cholesterol (LDL-C) concentration, we assayed these substances in serum from 679 pedigreed baboons. Using a maximum likelihood-based variance components approach, we detected significant evidence for a QTL affecting Lp-PLA(2) activity (LOD=2.79, genome-wide P=0.039) and suggestive evidence for a QTL affecting LDL-C levels (LOD=2.16) at the same location on the baboon ortholog of human chromosome 2p. Because we also found a significant genetic correlation between the two traits (rho(G)=0.50, P<0.00001), we conducted bivariate linkage analyses of Lp-PLA(2) activity and LDL-C concentration. These bivariate analyses improved the evidence (LOD=3.19, genome-wide P=0.015) for a QTL at the same location on 2p, corresponding to the human cytogenetic region 2p24.3-p23.2. The QTL-specific correlation between the traits (rho(Q)=0.62) was significantly different from both zero and 1 (P[rho(Q)=0]=0.047; P[rho(Q)=1]=0.022), rejecting the hypothesis of co-incident linkage and consistent with incomplete pleiotropy at this locus. We conclude that polymorphisms at the QTL described in this study exert some genetic effects that are shared between Lp-PLA(2) activity and LDL-C concentration.

Gene-environment interactions in the pathogenesis of type 2 diabetes and metabolism

PURPOSE OF REVIEW

Gene-environment interaction can be viewed as a departure from an otherwise expected additivity of genetic and environmental factors on a given outcome measure. Important genetic and environmental factors contribute to the pathogenesis of type 2 diabetes and intermediary traits, probably modulated by their complex interaction. This paper provides an update on the current literature investigating gene-environment interactions of type 2 diabetes and metabolic phenotypes, and discusses the future perspectives of this research.

RECENT FINDINGS

Recent advances in gene-environment interaction studies of metabolism have involved LIPC, APOA5 and PPARG variation, and nutrition and physical activity, of which the most consistently replicated observations have been obtained for APOA5. Also, intervention studies of the promising TCF7L2 type 2 diabetes gene and possible future strategies are discussed.

SUMMARY

Possibly as a result of the complexity of these multifactorial diseases, recent years have seen only limited success in unravelling significant gene-environment interactions, but important insights have been gained and they hold promise for implementation in lifestyle intervention strategies. We need to evolve to more complex, but realistic, scenarios involving several genes and environmental factors. Recent progress in statistical methods allowing for higher-order interactions may make this possible.