Phenotypic and Genetic Overlap Between Autistic Traits at the Extremes of the General Population

University of Cambridge, Cambridge, England, United Kingdom
Journal of the American Academy of Child & Adolescent Psychiatry (Impact Factor: 7.26). 11/2006; 45(10):1206-14. DOI: 10.1097/01.chi.0000230165.54117.41
Source: PubMed

ABSTRACT To investigate children selected from a community sample for showing extreme autistic-like traits and to assess the degree to which these individual traits--social impairments (SIs), communication impairments (CIs), and restricted repetitive behaviors and interests (RRBIs)--are caused by genes and environments, whether all of them are caused by the same genes and environments, and how often they occur together (as required by an autism diagnosis).
The most extreme-scoring 5% were selected from 3,419 8-year-old pairs in the Twins Early Development Study assessed on the Childhood Asperger Syndrome Test. Phenotypic associations between extreme traits were compared with associations among the full-scale scores. Genetic associations between extreme traits were quantified using bivariate DeFries-Fulker extremes analysis.
Phenotypic relationships between extreme SIs, CIs, and RRBIs were modest. There was a degree of genetic overlap between them, but also substantial genetic specificity.
This first twin study assessing the links between extreme individual autistic-like traits (SIs, CIs, and RRBIs) found that all are highly heritable but show modest phenotypic and genetic overlap. This finding concurs with that of an earlier study from the same cohort that showed that a total autistic symptoms score at the extreme showed high heritability and that SIs, CIs, and RRBIs show weak links in the general population. This new finding has relevance for both clinical models and future molecular genetic studies.

96 Reads
  • Source
    • "Third , the internal consistency of some of the ASD dimensions was relatively low , in particular for AQ - Short Routine and Switch . This was presumably due to the low number of items or heterogeneity among these scale items . Low alphas for ASD dimensions have been reported before [ Ronald et al . , 2006 ; Hoekstra et al . , 2011 ] , and future studies should aim to optimize the collection of autistic dimensions in the ( adult ) general population ."
    [Show abstract] [Hide abstract]
    ABSTRACT: Autistic traits are characterized by social and communication problems, restricted, repetitive and stereotyped patterns of behavior, interests and activities. The relation between autistic traits and personality characteristics is largely unknown. This study focused on the relation between five specific autistic traits measured with the abridged version of the Autism Spectrum Quotient ("social problems," "preference for routine," "attentional switching difficulties," "imagination impairments," "fascination for numbers and patterns") and Experience Seeking (ES) in a general population sample of adults, and subsequently investigated the genetic and environmental etiology between these traits. Self-reported data on autistic traits and ES were collected in a population sample (n = 559) of unrelated individuals, and in a population based family sample of twins and siblings (n = 560). Phenotypic, genetic and environmental associations between traits were examined in a bivariate model, accounting for sex and age differences. Phenotypically, ES correlated significantly with "preference for routine" and "imagination impairments" in both samples but was unrelated to the other autistic traits. Genetic analyses in the family sample revealed that the association between ES and "preference for routine" and "imagination impairments" could largely be explained by a shared genetic factor (89% and 70%, respectively). Our analyses demonstrated at a phenotypic and genetic level an inverse relationship between ES and specific autistic traits in adults. ES is associated with risk taking behavior such as substance abuse, antisocial behavior and financial problems. Future research could investigate whether autistic traits, in particular strong routine preference and impaired imagination skills, serve as protective factors for such risky behaviors. © 2015 Wiley Periodicals, Inc. © 2015 Wiley Periodicals, Inc.
    American Journal of Medical Genetics Part B Neuropsychiatric Genetics 08/2015; DOI:10.1002/ajmg.b.32352 · 3.42 Impact Factor
    • "The maximum possible score was 30; a score of 15 or above maximizes sensitivity (100%) and specificity (97%) to an ASC diagnosis [Williams et al., 2005]. In-line with prior studies [Ronald et al., 2006], the CAST was divided into three subscales corresponding to DSM-IV- TR [APA, 2000] autism symptoms: social; communication; and repetitive, restricted behaviors and interests (RRBI). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Language difficulties have historically been viewed as integral to autism spectrum conditions (ASC), leading molecular genetic studies to consider whether ASC and language difficulties have overlapping genetic bases. The extent of genetic, and also environmental, overlap between ASC and language is, however, unclear. We hence conducted a twin study of the concurrent association between autistic traits and receptive language abilities. Internet-based language tests were completed by ∼3,000 pairs of twins, while autistic traits were assessed via parent ratings. Twin model fitting explored the association between these measures in the full sample, while DeFries-Fulker analysis tested these associations at the extremes of the sample. Phenotypic associations between language ability and autistic traits were modest and negative. The degree of genetic overlap was also negative, indicating that genetic influences on autistic traits lowered language scores in the full sample (mean genetic correlation = −0.13). Genetic overlap was also low at the extremes of the sample (mean genetic correlation = 0.14), indicating that genetic influences on quantitatively defined language difficulties were largely distinct from those on extreme autistic traits. Variation in language ability and autistic traits were also associated with largely different nonshared environmental influences. Language and autistic traits are influenced by largely distinct etiological factors. This has implications for molecular genetic studies of ASC and understanding the etiology of ASC. Additionally, these findings lend support to forthcoming DSM-5 changes to ASC diagnostic criteria that will see language difficulties separated from the core ASC communication symptoms, and instead listed as a clinical specifier. © 2014 Wiley Periodicals, Inc.
    American Journal of Medical Genetics Part B Neuropsychiatric Genetics 10/2014; 165(7). DOI:10.1002/ajmg.b.32262 · 3.42 Impact Factor
  • Source
    • "For example, several reports from the Twins Early Development Study in the UK that have analysed data on the social interaction, communication and restricted repetitive behaviours and preoccupations in 7-, 9-and 12-year olds Psychopharmacology found autistic traits to be highly heritable (about 0.58–0.88 at age 12). However, these report further described rather weak phenotypic correlations among these domains (Pearson correlations between 0.1 and 0.4) and also rather weak genetic correlations (between 0.18 and 0.40) (Robinson et al. 2012; Ronald et al. 2006). In other words, more than half the genes that contribute to variation in social impairment are independent from those that contribute to variation in communicative skills or restricted repetitive behaviours. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The establishment of robust and replicable behavioural testing paradigms with translational value for psychiatric diseases is a major step forward in developing and testing etiology-directed treatment for these complex disorders. Based on the existing literature, we have generated an inventory of applied rodent behavioural testing paradigms relevant to autism spectrum disorders (ASD). This inventory focused on previously used paradigms that assess behavioural domains that are affected in ASD, such as social interaction, social communication, repetitive behaviours and behavioural inflexibility, cognition as well as anxiety behaviour. A wide range of behavioural testing paradigms for rodents were identified. However, the level of face and construct validity is highly variable. The predictive validity of these paradigms is unknown, as etiology-directed treatments for ASD are currently not on the market. To optimise these studies, future efforts should address aspects of reproducibility and take into account data about the neurodevelopmental underpinnings and trajectory of ASD. In addition, with the increasing knowledge of processes underlying ASD, such as sensory information processes and synaptic plasticity, phenotyping efforts should include multi-level automated analysis of, for example, representative task-related behavioural and electrophysiological read-outs.
    Psychopharmacology 03/2014; 231:1125-1146. DOI:10.1007/s00213-013-3268-5 · 3.88 Impact Factor
Show more