We studied lifespan and lifetime reproductive success (LRS) in mice from a selection experiment that includes 4 replicate High Runner (HR) lines bred for voluntary wheel-running behavior and 4 non-selected Control (C) lines. HR mice typically run threefold farther on a daily basis, leading to substantially increased energy expenditure and food consumption, and are also more active and eat more when housed without wheels. HR and C mice have been shown to differ for many other traits that could affect reproduction, including body mass and composition, aerobic capacity, hormonal profile (including corticosterone and leptin), and other behavioral traits (e.g., maternal care, thermoregulatory nesting, anxiety-like behavior in an elevated plus maze). At generation 97, male–female pairs were maintained until end of life in standard cages without wheels, with ad lib food and water. The number of pups weaned (LRS) ranged from 0–80, with a mean of 29.1 ± 8.1 (SE) for C lines and 25.8 ± 4.4 for HR (P = 0.73). Thus, artificial selection for high voluntary exercise did not significantly alter LRS. In addition, we found no statistical difference in average lifespan for dams or sires. However, replicate lines differed significantly for several traits, presumably due primarily to random genetic drift. At the level of individual variation, correlations among reproductive traits and lifespan provided little evidence for negative associations, and LRS was not negatively related to parental longevity. Overall, long-term selection on voluntary locomotor behavior did not produce detectable reductions in reproductive output or lifespan under standard laboratory conditions.
Behavioral dysfunction is a common characteristic of many neurodevelopmental and mental health disorders. While the causes of these disorders vary, aberrant behaviors may arise from alterations in transcriptional regulation during early neural development. Because transcription factors (TFs) often belong to families of closely related members, disruption of a single TF may indirectly influence the functionality of other family members. Consequently, mutations in TFs within the same family may lead to overlapping, yet distinct, phenotypes. This feature of TF function has important implications for understanding behavioral phenotypes, but detailed analyses across a single TF family are still lacking. In this study, we present a comprehensive behavioral analysis of adult zebrafish harboring mutations in individual members of the TALE and Hox TF families, that are essential for nervous system development. Using a battery of validated behavioral assays, we uncover elevated stress responses among all TF mutant lines, as well as TF-specific dysregulation in social interaction, locomotion, and endurance. The shared behavioral abnormalities across mutants suggest TF family members converge on core developmental pathways for stress-related behavioral regulation, while mutation-specific phenotypes indicate unique roles for individual TFs in fine-tuning neural function. Our findings provide a systematic behavioral characterization of TALE and Hox mutants in a vertebrate model and provide a framework for understanding how genetic variation within TF families may differentially contribute to vulnerability for neurodevelopmental and mental health disorders.
Attention-deficit/hyperactivity disorder (ADHD) symptoms are strongly associated with lower academic performance, yet the nature of this association remains uncertain. We analyzed data from twins and siblings in the Adolescent Brain Cognitive Development (ABCD) Study® to examine the relationship between ADHD symptoms and school grades using genetically informed models. ADHD symptoms were assessed via the Child Behavior Checklist, and grades were caregiver-reported for the previous school year. Both phenotypes were modeled as ordinal variables. We applied Direction of Causation (DoC) and Mendelian randomization extensions (MR-DoC and MR-DoC2) incorporating polygenic risk scores for ADHD and educational attainment. DoC and MR-DoC comparisons did not distinguish directionality (e.g., DoC forward path estimate = − 0.16, 95
In celebration of Matt McGue’s career, we conducted a longitudinal extension of a simple and elegant study he supervised nearly two decades ago (Keyes et al. 2008). We examined the effects of teenage exposure to parental smoking on substance use and externalizing outcomes in an adoption study. In total, 1232 offspring from 617 families were longitudinally assessed at mean ages of 15, 18, 22, and 32 years, with 37
The Minnesota Transracial Adoption Study (MTRAS) has been a source of enduring controversy due to its goal of addressing genetic and environmental explanations of Black-White IQ differences. The study followed the intellectual development of 130 Black and interracial children adopted into relatively advantaged White households. However, contrary to the initial interpretations of its original researchers, MTRAS became a tool for hereditarian racial ideologies. Its findings were appropriated and publicized by figures associated with racist research agendas, such as J. Philippe Rushton. We contend that the design of MTRAS was never capable of achieving its most ambitious scientific aims and, as a result, unintentionally contributed to a body of speculative literature that leveraged its inconclusive findings to support harmful discriminatory ideologies. To the extent that MTRAS was both limited in scientific value and potentially harmful, this paper examines the question: was MTRAS abhorrent race science?
Large population databases frequently include complicated family structures that are amenable to modern biometric methods: allowing for intergenerational and extended pedigree analyses. To date, much of the latent potential of these resources remains untapped due to numerous complexities that arise in their analysis. Two difficult and critical problems are (1) finding independent extended families within larger population databases, and (2) determining coefficients of relatedness among all pairs of individuals within those extended families. If these problems were solved, researchers could more fully utilize data on extended families for biometric modeling. In this paper, we provide fast, computationally efficient algorithms for both of these problems and several more that are applicable to arbitrarily large and complex pedigrees. The algorithms rely solely on mother-child and father-child relationships that form the basis of many large population databases. These methods will be invaluable to any researcher trying to segment standard pedigree data files into independent extended family units, compute relatedness coefficients within extended families, and conduct intergenerational and other biometric modeling.
Processing speed (PS), the rate at which an individual can perform basic cognitive tasks, and executive functions (EFs), cognitive abilities regulating thoughts and action, are key cognitive abilities related to important life outcomes. We investigated the relationship between PS and multiple EFs (a common factor and two factors specific to updating working memory and shifting mental sets) across adolescence (M age = 17.26 years, SD = 0.68) and established adulthood (M age 29.34 years, SD = 1.29). 797 individual twins (MZ = 428, DZ = 369) in the Colorado Longitudinal Twin Study completed 3 PS tasks and 6–9 EF tasks. PSage16 correlated strongly with PSage29 (r = .90) and Common EF (cEF) at age 17 (r = .70). Furthermore, cEF and Updating-Specific factors in late adolescence predicted new variance in PSage29 (rs = 0.13 to 0.16), but PSage16 did not predict new variance in cEF in established adulthood. PS and cEF were highly heritable at both ages (a2 > = 0.74) and genetically correlated at both time points (ra = 0.70 to 0.73). PS was not significantly genetically correlated with the Updating- or Shifting-Specific factors in late adolescence, but was in established adulthood. Results indicate that PS and EFs are highly correlated yet distinct constructs. PS and EFs both demonstrate strong stability of genetic/environmental influences across nearly 15 years from late adolescence into established adulthood. It will be important for future developmental studies to understand why specific EF factors predict future variance in PS, but not the reverse.
Functional literacy-the ability to apply reading, mathematical, and scientific knowledge in authentic contexts as operationalized by the PISA framework-is a key predictor of educational attainment, labour-market outcomes, and economic growth. Despite extensive behavioral-genetic research on cognitive ability, the heritability of competency-based literacy measures remains largely unexamined, particularly outside Western populations. The present study addresses this gap by providing the first twin-based heritability estimates for PISA-type functional literacy from Russia. Participants were 114 twin pairs (50 monozygotic [MZ], 64 dizygotic [DZ]; age 14-15) identified within a nationally representative sample of 33,050 eighth-grade students assessed on the Russian Electronic School platform. A bifactor item-response theory model derived scores for a general cognitive ability factor and three domain-specific factors (mathematical, reading, and scientific literacy). Classical twin structural-equation modelling decomposed phenotypic variance into additive genetic (A), shared environmental (C), and non-shared environmental (E) components. The AE model provided the best fit across all phenotypes. Heritability estimates were substantial: functional literacy = 0.76 [95% CI: 0.67-0.85], mathematical literacy = 0.72 [0.59-0.84], reading literacy = 0.69 [0.56-0.82], and scientific literacy = 0.71 [0.59-0.82]. Shared environmental contributions were negligible and non-significant in all models. The twin sample was broadly representative of the source population, with only small mean differences in functional and mathematical literacy. These findings extend the behavioral-genetic architecture of cognitive ability to competency-based assessments and demonstrate its robustness in a non-Western educational context. Limitations include modest sample size, photo-based zygosity assignment, and absence of molecular genetic data.
This paper presents the preliminary results of a survey-based study investigating geneticists’ understanding of genetic causation. We are motivated by an underlying assumption that to improve interpretation and implementation of behavioral genetics research, and furthermore for that research to have a positive societal impact, it is important to reveal how geneticists themselves understand their research. Results of sixty-two survey responses from clinical and research geneticists at the University of Utah are discussed and assessed to fulfill these aims and survey the landscape of the field. Analysis of our survey results revealed some variation in responses to questions about genetic causation from geneticists across different research areas.
Matt McGue's research has made intelligence a model case for the explanatory reach of behavioral genetics. His early studies, conducted with Thomas Bouchard and other collaborators, on twins and other kinships demonstrated that cognitive ability is substantially heritable, and later analyses showed that this influence increases from childhood to adulthood and remains appreciable in late life, while shared environmental effects diminish. McGue and colleagues also extended twin and adoption methods to the study of educational attainment and social mobility, demonstrating that cognitive ability, noncognitive traits, and family background all contribute, and that higher education seems to benefit individuals across the ability spectrum. Because intelligence is uniquely scrutinized, the credibility of these findings depends on the soundness of the adoption design. A persistent critique is that adoption agencies engage in various forms of selection, leading to restriction of range in environmental quality and the confounding of genetic and environmental influences. McGue and colleagues addressed some of these concerns in their Minnesota studies, most notably showing that a correction for restriction of range did not appreciably alter the correlation between adoptive siblings in IQ. The present paper reviews McGue's accomplishments and reports new results about selective placement based on the correlations between parental phenotypes and the polygenic scores of their adopted offspring.
Early-life stress has been linked to anxiety, pessimism, and cognitive decline, all of which can have detrimental effects on individuals. One critical structure impacted by early-life stress is the hippocampus, which plays a vital role in regulating learning and memory functions. This study aims to elucidate the molecular mechanisms through which early-life psychological stress (ELPS) affects the learning and memory capabilities of the hippocampus in rats. In this study, ELPS model was applied on juvenile rats for 14 days. To evaluate the spatial learning and memory abilities of rats, the Morris Water Maze (MWM) test was adopted in this study. This study employed two-dimensional gel electrophoresis (2DE) and ultrahigh-performance liquid chromatographic-electrospray ionization quadrupole time-of-flight mass spectrometry (UPLC-ESI-Q-TOF-MS) techniques to reveal the proteomic map of the rat hippocampus. Subsequently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and direct protein-protein interaction (PPI) network construction were performed to identify the biological pathways associated with the differentially expressed proteins (DEPs). ELPS treatment induced age-dependent spatial cognitive deficits. In juvenile rats, ELPS caused mild impairments in spatial learning, as reflected by increased escape latency during specific training days, but did not affect spatial memory. In contrast, adult rats exhibited significant and robust impairments in both spatial learning and spatial memory, with longer escape latency across training days, fewer platform crossings, and reduced time and distance in the target quadrant. These findings demonstrate that ELPS-induced cognitive dysfunction is far more pronounced in adult animals, representing persistent and progressive spatial learning and memory deficits. After therapy, 71 proteins were found in the hippocampal region, including 10 DEPs across the juvenile and adult groups post-treatment. Subsequent pathway analysis indicated the involvement of 4 DEPs in various pathways, including Biosynthesis of amino acids, Prion disease, HIF-1 signaling pathway, Distal axon, Alzheimer disease, and Necroptosis pathways. Our study revealed significant proteomic changes in the hippocampus of rats as they transitioned from juveniles to adults, including 10 consistently differentially expressed proteins. These proteomic alterations correspond to the age-dependent spatial cognitive deficits induced by ELPS-mild spatial learning impairment in juveniles and more pronounced deficits in both spatial learning and memory in adults-collectively supporting the lasting effects of ELPS on spatial cognitive function.
Fisher’s principle proposes that the population sex ratio remains approximately equal (around 1:1) through a process of negative frequency-dependent selection. However, large-scale studies show that human offspring sex ratio (OSR) is not heritable in humans, raising doubts about whether Fisher’s principle is a valid explanatory framework for human OSR evolution. Song and Zhang recently proposed a mechanism where OSR propensity is genetically influenced but produces an OSR phenotype with negligible observable heritability due to probabilistic sex determination and small family sizes. Here, we conducted an evaluation of this proposal by simulating the proposed mechanism under a broader range of assumptions, revealing that it only produces undetectable OSR heritability under a specific set of parameters. We further explored whether Fisher’s principle could operate under such low OSR heritability and found, by reanalysing simulation outcomes, that there is only a very narrow parameter space where Fisher’s principle operated and OSR heritability was realistically low. We also make multiple corrections to Song and Zhang’s original analyses which weaken their claim. In sum, our assessment is that evidence for a genetic mechanism that yields undetectable OSR heritability, but allows the operation of Fisher’s Principle, is weak. The proposed mechanism lacks sufficient evidence to overturn the simpler conclusion that OSR is not heritable in humans, and consequently that Fisher’s principle is not a valid explanatory framework for human OSR.
Matt is an unusually talented scholar, with seminal contributions in many areas (cognitive ability, environmental influences, substance use, etc.). One key theme that undergirds and unites his work is his development of novel research designs and experiments of nature capable of estimating causal inferences with unusual precision. In this contribution to his Festshrift, I illuminate key models and experiments developed by Matt and their contributions to the field. In lieu of new empirical data, I also present a new idiographic behavioral genetic research design - to be called 'narrative non-shared environmental identification' - in which we leverage idiographic, person-specific measurement tools to meaningfully study the effective non-shared environment for the first time.
Polygenic scores have potential for predicting health outcomes based on genomic data. However, their portability across diverse populations is hindered by the overrepresentation of European ancestries in discovery genome-wide association (GWAS) studies. The current study evaluated the transferability of substance use polygenic scores across European, African, American, and East Asian populations. European-based scores demonstrated reduced predictive accuracy in non-European populations with mean relative accuracies of 36% in American, 22.5% in East Asian, and 3.8% in African ancestries. Beyond discrete genetic ancestry groups, we found that European-based score prediction declined as a function of genetic distance from the European-stratified GWAS. That is, predictive accuracy declines as genetic distance, derived from continuous measures of ancestry, from the discovery GWAS increases. We evaluated the extent to which differences across populations in linkage disequilibrium (LD), allele frequency, and heritability explained the reduction in accuracy. Across phenotypes we found that approximately 56.0% of accuracy decline in American ancestries was explained by LD and allele frequency differences, with differences in LD appearing to be the primary factor. For East Asian ancestries this was 67.1% and was 84.0% in African ancestries. Differences across populations in heritability accounted for only small portions of reduced polygenic score transferability. This suggests that LD and frequency differences explain significant portions of accuracy loss, but substantial reductions persist, suggesting contributions from other factors like causal effect size differences. The current study provides important information on the causes and extent of the polygenic score transferability problem and has important implications for refinement of polygenic scores. Expanding the diversity of GWAS and target samples remains essential for improving accuracy and equity in prediction.
Wellbeing—the balance of positive and negative affect and life satisfaction—is closely tied to mental health. However, the sources of stability and change and how life events shape wellbeing are not well understood. Data from the Oslo University Adolescent and Young Adult Project (2,879 individuals from 1,483 families) provided wellbeing and life‑event data across three waves 2 years apart (ages 12–18 at Wave 1). We used a trivariate Cholesky decomposition to estimate genetic and environmental contributions to wellbeing, and genetically informative random intercept cross‑lagged panel models (RI‑CLPMs) to test prospective effects of life events on wellbeing. Wellbeing showed moderate stability, predominantly driven by genetic influences (81
The inclusion of polygenic scores (PGS) from genetically correlated traits such as Major Depressive Disorder (MDD) and alcohol use disorder (AUD) may improve the prediction of these outcomes and their comorbidity. Despite the importance of this work, few studies have evaluated the efficacy of this possibility. The current study evaluates the use of MDD and AUD PGS individually and together to improve the prediction of MDD, AUD, and comorbid MDD-AUD using a sample of European, African, or Admixed American Ancestry participants from the National Longitudinal Study of Adolescent to Adult Health (N = 7,965). Cross-ancestry MDD and AUD PGS were created using PRS-CSx. The best fitting model of comorbid MDD-AUD in the whole sample included PGS for MDD and AUD (PGSMDD OR: 1.26, 95
For half a century, scholars have debated the societal implications of research that suggests genetic explanations for socioeconomic disparities. However, studies of whether and how such genetic explanations affect the attitudes of citizens are surprisingly limited. Using two identical, pre-registered survey experiments fielded in the United States (N=2003) and Sweden (N=2005), this study tests whether a short, twin study-based genetic explanation for income and education differences leads to citizens adopting a lower perceived possibility, as well as a lower perceived fairness, of the government reducing socioeconomic differences. The results do not provide systematic evidence to this effect. Rather, there are indications that citizens increase their belief in the effectiveness, as well as in the fairness of government intervention. Most clearly, the treatment increased citizens’ support for a set of redistributive policy proposals, particularly in Sweden. Overall, however, the treatment effects appear to have been limited. This is likely attributable to the experimental manipulation being inadequate in magnitude. This study suggests that genetic explanations for socioeconomic differences do not necessarily have to cause a naturalization of inequality among citizens, but further research on this important topic is needed.
The Danish Twin Registry has been expanded several times since its establishment in 1954. Here we describe the inclusion of the Danish 2001-2021 twin birth cohorts and the merging with the Danish 1870-2000 twin birth cohorts that enabled us to analyze changes in both twin birth frequency and survival over three centuries. Through Danish nationwide registers, 22,680 twin pairs born in Denmark in the period 2001-2021 were identified and contacted. Response to an electronic questionnaire including similarity questions for zygosity classification was obtained from 15,005 pairs (66.2%). Genetic markers for a subsample revealed a zygosity misclassification rate of 5.3%. From the mid-19th to the mid-20th century, twinning rate was relatively stable followed by an unexplained one third reduction from 1950 to 1970. A doubling occurred from the 1980s to the 2000s due to increasing maternal age at birth and, most importantly, Medically Assisted Reproduction (MAR). Changes in MAR procedures have since resulted in a marked decline, bringing the contemporary twinning rate back to the level of the 19th century. In the 19th century, 43% of twins died before age 6; in the 1970s, the number was 5.4% and in the most recent 21st century cohorts, it was 2.0%. In all three centuries studied, the twin mortality before age 6 was 2-5 times higher in twins compared to singletons. However, large studies of the Danish and other Scandinavian twin cohorts suggest that twin-singleton differences generally vanish with age and, for more recent birth cohorts, are few and generally small after childhood.
Socioeconomic status (SES) predicts age-related changes in health status; however, the source of SES-health associations is heavily debated. Twin studies allow tests of causal hypotheses by modeling within and between twin pair differences in longitudinal latent growth curve models (LGCM) of physical aging, examining level of functioning and rate of change with age. Three longitudinal twin studies of aging (mean age at baseline = 71.36, SD = 10.7) from the Swedish Twin Registry (N = 1369) included up to 27 years of follow-up on a Functional Aging Index (FAI) consisting of lung function, grip strength, walking speed, and self-report sensory functioning. SES indicators included education, financial strain, and occupation-based socioeconomic position. Pair means (between family effect) and within pair differences (within family effect) for SES were included as covariates of both intercept and slopes in a two-slope LGCM (intercept at age 75); models were corrected for sex and parental SES. LGCM results were compared across the full sample, then separately for both monozygotic and dizygotic twin pairs. Results indicated genetic confounding in the associations between multiple SES indicators and the FAI intercept. Additionally, the relationship between education longitudinal change in physical aging up to age 75 was subject to genetic confounding. These patterns were replicated among men. In contrast, findings for women pointed to shared environmental influences rather than genetic confounding, although statistical power was reduced in sex-stratified analyses. Results highlight the importance of considering the timing of socioeconomic exposures and gendered life-course trajectories when examining health inequalities in aging.
