Heritability across the distribution: an application of quantile regression

The Differential Relations Between ADHD and Reading Comprehension: A Quantile Regression and Quantile Genetic Approach

This paper extends the understanding of the relation between ADHD and reading comprehension, through examining how this relation differs depending on the quantile an individual falls in for each. Samples from three twin projects around the United States were used (Florida Twin Project, Colorado component of International Longitudinal Twin Study of Early Reading Development, & Western Reserve Reading and Math Projects). Phenotypic analysis using quantile regression showed relations between ADHD related behaviors and reading comprehension to be stronger in the lower quantiles of reading comprehension in two of three samples. A new method was developed extending this analysis into the bivariate genetic space. Results of this quantile genetic analysis revealed that overlapping common environmental influences accounted for a larger proportion of variance in the lower quantiles of these variables in two of three samples. Finally, in all three samples the phenotypic relation was strongest when shared environmental influences accounted for a larger proportion of the overall variance.

Heritability curves: A local measure of heritability in family models

Classical heritability models for family data split the phenotype variance into genetic and environmental components. For instance, the ACE model in twin studies assumes the phenotype variance decomposes as a²  + c²  + e² , representing (additive) genetic effects, common (shared) environment, and residual environment, respectively. However, for some phenotypes it is biologically plausible that the genetic and environmental components may vary over the range of the phenotype. For instance, very large or small values of the phenotype may be caused by "sporadic" environmental factors, whereas the mid-range phenotype variation may be more under the control of common genetic factors. This article introduces a "local" measure of heritability, where the genetic and environmental components are allowed to depend on the value of the phenotype itself. Our starting point is a general formula for local correlation between two random variables. For estimation purposes, we use a multivariate Gaussian mixture, which is able to capture nonlinear dependence and respects certain distributional constraints. We derive an analytical expression for the associated correlation curve, and show how to decompose the correlation curve into genetic and environmental parts, for instance, a² (y) + c² (y) + e² (y) for the ACE model, where we estimate the components as functions of the phenotype y. Furthermore, our model allows switching, for instance, from the ACE model to the ADE model within the range of the same phenotype. When applied to birth weight (BW) data on Norwegian mother-father-child trios, we conclude from the model that low and high BW are less heritable traits than medium BW. We also demonstrate switching between the ACE and ADE model when studying body mass index in adult monozygotic and dizygotic twins.

Exploring the Influence of Early Childhood Education and Care on the Etiology of Achievement

The present study used a genetically-sensitive quantile regression approach to examine the relation between participation in early childhood education and care (ECEC) and subsequent school performance in literacy and numeracy at grades 3, 5, 7, and 9. The sample consisted of 1255 twin pairs (596 MZ; 659 DZ) with information on both ECEC and the National Assessment Program-Literacy and Numeracy (NAPLAN) scores from the Twin Study of NAPLAN. Results indicated variation in heritability estimates across the distributions of achievement, suggesting that different patterns of etiological influences may exist among children of different ability levels. Additionally, the results provided no evidence that ECEC significantly influenced achievement, and in the genetically-sensitive analyses, no evidence that ECEC moderated the influences of heritability of achievement for typically advantaged children. These results suggest that ECEC may not provide the levels of environmental support for later achievement that advocates claim, although we acknowledge that ECEC quality, which was not measured in the current study, may make a difference in whether or not ECEC influences achievement.

Differential Influences of Genes and Environment Across the Distribution of Reading Ability

We partitioned early childhood reading into genetic and environmental sources of variance and examined the full distribution of ability levels from low through normal to high as computed by quantile regression. The full sample comprised twin pairs measured at preschool (n = 977), kindergarten (n = 1028), grade 1 (n = 999), and grade 2 (n = 1000). Quantile regression analyses of the full distribution of literacy ability showed genetic influence in all grades from preschool to grade 2. At preschool, the low end of the distribution had higher genetic influence than the high end of the distribution and the shared environment influence was the opposite. These shared environment influences of preschool became insignificant with formal schooling. This suggests that higher scores in pre-literacy skills (preschool) are more influenced by shared environment factors, though these are short-lived. This study discusses the factors that may be influencing the results.

Assessing the evidence for shared genetic risks across psychiatric disorders and traits

Genetic influences play a significant role in risk for psychiatric disorders, prompting numerous endeavors to further understand their underlying genetic architecture. In this paper, we summarize and review evidence from traditional twin studies and more recent genome-wide molecular genetic analyses regarding two important issues that have proven particularly informative for psychiatric genetic research. First, emerging results are beginning to suggest that genetic risk factors for some (but not all) clinically diagnosed psychiatric disorders or extreme manifestations of psychiatric traits in the population share genetic risks with quantitative variation in milder traits of the same disorder throughout the general population. Second, there is now evidence for substantial sharing of genetic risks across different psychiatric disorders. This extends to the level of characteristic traits throughout the population, with which some clinical disorders also share genetic risks. In this review, we summarize and evaluate the evidence for these two issues, for a range of psychiatric disorders. We then critically appraise putative interpretations regarding the potential meaning of genetic correlation across psychiatric phenotypes. We highlight several new methods and studies which are already using these insights into the genetic architecture of psychiatric disorders to gain additional understanding regarding the underlying biology of these disorders. We conclude by outlining opportunities for future research in this area.

A Quantile Regression Approach to Understanding the Relations Among Morphological Awareness, Vocabulary, and Reading Comprehension in Adult Basic Education Students

The purpose of this study was to investigate the joint and unique contributions of morphological awareness and vocabulary knowledge at five reading comprehension levels in adult basic education (ABE) students. We introduce the statistical technique of multiple quantile regression, which enabled us to assess the predictive utility of morphological awareness and vocabulary knowledge at multiple points (quantiles) along the continuous distribution of reading comprehension. To demonstrate the efficacy of our multiple quantile regression analysis, we compared and contrasted our results with a traditional multiple regression analytic approach. Our results indicated that morphological awareness and vocabulary knowledge accounted for a large portion of the variance (82%-95%) in reading comprehension skills across all quantiles. Morphological awareness exhibited the greatest unique predictive ability at lower levels of reading comprehension whereas vocabulary knowledge exhibited the greatest unique predictive ability at higher levels of reading comprehension. These results indicate the utility of using multiple quantile regression to assess trajectories of component skills across multiple levels of reading comprehension. The implications of our findings for ABE programs are discussed.

Nonparametric Estimates of Gene × Environment Interaction Using Local Structural Equation Modeling

Gene × environment (G × E) interaction studies test the hypothesis that the strength of genetic influence varies across environmental contexts. Existing latent variable methods for estimating G × E interactions in twin and family data specify parametric (typically linear) functions for the interaction effect. An improper functional form may obscure the underlying shape of the interaction effect and may lead to failures to detect a significant interaction. In this article, we introduce a novel approach to the behavior genetic toolkit, local structural equation modeling (LOSEM). LOSEM is a highly flexible nonparametric approach for estimating latent interaction effects across the range of a measured moderator. This approach opens up the ability to detect and visualize new forms of G × E interaction. We illustrate the approach by using LOSEM to estimate gene × socioeconomic status interactions for six cognitive phenotypes. Rather than continuously and monotonically varying effects as has been assumed in conventional parametric approaches, LOSEM indicated substantial nonlinear shifts in genetic variance for several phenotypes. The operating characteristics of LOSEM were interrogated through simulation studies where the functional form of the interaction effect was known. LOSEM provides a conservative estimate of G × E interaction with sufficient power to detect statistically significant G × E signal with moderate sample size. We offer recommendations for the application of LOSEM and provide scripts for implementing these biometric models in Mplus and in OpenMx under R.

Heritability of Autism Spectrum Disorder in a UK Population-Based Twin Sample

IMPORTANCE

Most evidence to date highlights the importance of genetic influences on the liability to autism and related traits. However, most of these findings are derived from clinically ascertained samples, possibly missing individuals with subtler manifestations, and obtained estimates may not be representative of the population.

OBJECTIVES

To establish the relative contributions of genetic and environmental factors in liability to autism spectrum disorder (ASD) and a broader autism phenotype in a large population-based twin sample and to ascertain the genetic/environmental relationship between dimensional trait measures and categorical diagnostic constructs of ASD.

DESIGN, SETTING, AND PARTICIPANTS

We used data from the population-based cohort Twins Early Development Study, which included all twin pairs born in England and Wales from January 1, 1994, through December 31, 1996. We performed joint continuous-ordinal liability threshold model fitting using the full information maximum likelihood method to estimate genetic and environmental parameters of covariance. Twin pairs underwent the following assessments: the Childhood Autism Spectrum Test (CAST) (6423 pairs; mean age, 7.9 years), the Development and Well-being Assessment (DAWBA) (359 pairs; mean age, 10.3 years), the Autism Diagnostic Observation Schedule (ADOS) (203 pairs; mean age, 13.2 years), the Autism Diagnostic Interview-Revised (ADI-R) (205 pairs; mean age, 13.2 years), and a best-estimate diagnosis (207 pairs).

MAIN OUTCOMES AND MEASURES

Participants underwent screening using a population-based measure of autistic traits (CAST assessment), structured diagnostic assessments (DAWBA, ADI-R, and ADOS), and a best-estimate diagnosis.

RESULTS

On all ASD measures, correlations among monozygotic twins (range, 0.77-0.99) were significantly higher than those for dizygotic twins (range, 0.22-0.65), giving heritability estimates of 56% to 95%. The covariance of CAST and ASD diagnostic status (DAWBA, ADOS and best-estimate diagnosis) was largely explained by additive genetic factors (76%-95%). For the ADI-R only, shared environmental influences were significant (30% [95% CI, 8%-47%]) but smaller than genetic influences (56% [95% CI, 37%-82%]).

CONCLUSIONS AND RELEVANCE

The liability to ASD and a more broadly defined high-level autism trait phenotype in this large population-based twin sample derives primarily from additive genetic and, to a lesser extent, nonshared environmental effects. The largely consistent results across different diagnostic tools suggest that the results are generalizable across multiple measures and assessment methods. Genetic factors underpinning individual differences in autismlike traits show considerable overlap with genetic influences on diagnosed ASD.

Thinking positively: The genetics of high intelligence

High intelligence (general cognitive ability) is fundamental to the human capital that drives societies in the information age. Understanding the origins of this intellectual capital is important for government policy, for neuroscience, and for genetics. For genetics, a key question is whether the genetic causes of high intelligence are qualitatively or quantitatively different from the normal distribution of intelligence. We report results from a sibling and twin study of high intelligence and its links with the normal distribution. We identified 360,000 sibling pairs and 9000 twin pairs from 3 million 18-year-old males with cognitive assessments administered as part of conscription to military service in Sweden between 1968 and 2010. We found that high intelligence is familial, heritable, and caused by the same genetic and environmental factors responsible for the normal distribution of intelligence. High intelligence is a good candidate for "positive genetics" - going beyond the negative effects of DNA sequence variation on disease and disorders to consider the positive end of the distribution of genetic effects.

It depends: conditional correlation between frequency of storybook reading and emergent literacy skills in children with language impairments

The current study examined the association between frequency of storybook reading and emergent literacy in 212 children at risk for language impairment, assessed during the fall semester of kindergarten. Measures included parent-reported storybook reading, as well as direct assessments of print knowledge, letter awareness, and expressive vocabulary. Results suggested nonsignificant to moderate (r = .11 to .25) correlations between frequency of storybook reading and child emergent literacy across the entire range of environment and ability. Quantile regression results suggested that the association was highest at low frequency of storybook reading, particularly for print knowledge, approaching r = .50. Moreover, the association between frequency of storybook reading and emergent literacy was highest at higher levels of emergent literacy for print knowledge, but particularly for letter naming, approaching r = .80. These results suggest that in children with language difficulties, the relationship between aspects of the home environment and emergent literacy is conditional on the quality of the home environment as well as the child's proficiency in emergent literacy skills.

A twin study of heritable and shared environmental contributions to autism

The present study examined genetic and shared environment contributions to quantitatively-measured autism symptoms and categorically-defined autism spectrum disorders (ASD). Participants included 568 twins from the Interactive Autism Network. Autism symptoms were obtained using the Social Communication Questionnaire and Social Responsiveness Scale. Categorically-defined ASD was based on clinical diagnoses. DeFries-Fulker and liability threshold models examined etiologic influences. Very high heritability was observed for extreme autism symptom levels ([Formula: see text]). Extreme levels of social and repetitive behavior symptoms were strongly influenced by common genetic factors. Heritability of categorically-defined ASD diagnosis was comparatively low (.21, 95 % CI 0.15-0.28). High heritability of extreme autism symptom levels confirms previous observations of strong genetic influences on autism. Future studies will require large, carefully ascertained family pedigrees and quantitative symptom measurements.

Quantile regression in the study of developmental sciences

Linear regression analysis is one of the most common techniques applied in developmental research, but only allows for an estimate of the average relations between the predictor(s) and the outcome. This study describes quantile regression, which provides estimates of the relations between the predictor(s) and outcome, but across multiple points of the outcome's distribution. Using data from the High School and Beyond and U.S. Sustained Effects Study databases, quantile regression is demonstrated and contrasted with linear regression when considering models with: (a) one continuous predictor, (b) one dichotomous predictor, (c) a continuous and a dichotomous predictor, and (d) a longitudinal application. Results from each example exhibited the differential inferences which may be drawn using linear or quantile regression.