Thursday 12 December 2013

ISIR Genetics

For those of you who are really interested in Australian cuisine, the conference proceedings includes suggested lunch places within 5 minutes walk of the lecture hall. Establishments such as Starbucks, Pizza Hut, and Subway; those sorts of autochthonous providers of culinary delights.

SES &Heritability of IQ;

When Does Socioeconomic Status Moderate the Heritability of IQ? Data from
Australia and the USA. When is the heritability of intelligence dependent on socio economic status and when it is not? US studies largely suggest that cognitive ability is more heritable among those raised in higher socioeconomic status (SES) families. However, the mechanism of this effect is unclear, and the effect may not be universal. We tested for gene × SES interaction effects on Full-scale IQ in 2,307 adolescent Australian twins (mean age 16.2 years). While mean scores in were higher among those from higher SES backgrounds, the magnitude of genetic influences on IQ was constant across the range of SES.
The heritability of intelligence was high, and unrelated to social status.
This suggests that during the development of intelligence, genes multiply cultural inputs supportive of intellectual growth. It suggests also, however, that this interaction can be decoupled from parental SES, possibly via factors such as quality of school provision.

Do Functional SNPs Show an Enriched Association for Intelligence?;

In the present study we examined the association of intelligence with
neuronally-expressed functional Single Nucleotide Polymorphisms (SNPs), i.e. SNPs
that alter gene expression and function in the brain. These include promoter SNPs,
SNPs that change gene expression via microRNA binding, methylated SNPs and
eQTLs: SNPs that correlate with quantitative measures of gene expression. Methylated SNPs and eQTLs were further divided according to their region of activation within the brain, including frontal cortex, temporal cortex, cerebellum and pons giving a total of 10 functional SNP sets. Using subjects in the family-based Generation Scotland sample, we tested whether any of these 10 categories of SNP show significant enrichment for fluid or crystallised intelligence by comparison with non-functional SNPs. Results to date support the idea that functional SNPs significantly enriched for association with both gf and gc compared to SNPs which do not alter gene expression in the brain. We will also present data testing whether this minority of SNPs accounts for the majority of heritability.

Intelligence is heritable and predicts wealth, health and mortality. However, whilst twin studies have demonstrated that a heritable component accounts
for around 50% of the variation in intelligence differences, Genome Wide Association
Studies (GWAS) carried out to identify individual variants are currently too small to
reliably reveal such variants. The next challenge is to identify which genetic variants
contribute to variation in human intelligence. If SNPs can be pruned according to their functional role, the power of genetic studies will be enhanced, bringing the prospect of identifying small but true genetic effects on intelligence closer.
Methods. Experimentally validated promoter SNPs were extracted from dbQSNP
( with the functional categories of microRNA,
methylated SNPs, and eQTL SNPs being based on published literature. These functional SNP sets were analysed using robust statistics using empirical tests of significance based on simulations which account for both average background signal enrichment and preserve the linkage disequilibrium (LD) between SNP statistics. This matching strategy ensures that it is the biological group to which the SNPs belong, rather than extraneous genomic features such patterns of LD which link these functional categories to cognitive abilities.

A Pilot Study of Rare Genetic Variants and G

Various genetic linkage, genome-wide SNP association, and genetic copy
number variant studies have investigated genetic variation in cognitive ability, but no
studies have investigated genetic rare variants. These may be a potentially important
source of genetic variation; exome sequencing studies have already met with success in discovering novel trait-gene associations. Here, we investigate the effects of rare
variants on general cognitive ability. 150 unrelated individuals from the Generation
Scotland: Scottish Family Health Study were selected for high scores (>2.3 SD from themean ability score) on a general component of intelligence (g) based on Logical
Memory immediate and delayed, Digit Symbol, Verbal Fluency, and Mill Hill
Vocabulary test. The DNA of these individuals was exome-sequenced and the
frequency of rare genetic variants compared with those from a control sample who
scored in the lower to middle range of the g distribution. The results of single-SNP and multi-SNP tests will be presented, including a discussion of the merits of these
approaches and their power.

This is one of the first genetic rare variant studies of general cognitive
ability and increases our understanding of the genetic architecture of general cognition.

Methods. Genetic sequencing is expensive, so we were limited to genotyping the
high extreme of our sample. But these methods are becoming more affordable.
Therefore, these preliminary results will be able to inform our future larger genetic rare variant study design. And the sequencing data can be integrated with existing SNP data to improve genome coverage and potential to identify genetic variants associated with cognitive ability.

Low IQ and Mild Mental Retardation are Heritable But Severe Mental
Retardation is Not: a Genetic Analysis of 740,000 Siblings and 18,000 Twins

Despite the obvious societal importance of mental retardation (MR) in our
increasingly technological world, the most fundamental question about the genetic and environmental origins of MR has not been definitively addressed. We will report new results showing that diagnosed mild MR (IQ 50-70) and low IQ in the general
population are caused by the same genetic and environmental factors responsible for the normal range of intelligence. In contrast, most severe MR (IQ <50) is not inherited, despite hundreds of known single-gene causes of MR. Because severe MR is not inherited, its most likely causes are environmental factors, some of which are well known, such as prenatal trauma, neurotoxicity, and infections. Although severe MR is not inherited, MR is a burgeoning area of genetic research using new techniques (exome and whole genome sequencing) to identify non-inherited (de novo) mutations.
Importance. Molecular genetic research attempting to identify de novo mutations
responsible for MR will benefit from focusing on severe MR (IQ < 50), which is not
inherited. Although research on de novo mutations is an exciting new direction for
identifying causes of severe MR, severe MR is fortunately rare (.001). Much more of
the societal burden of intellectual disability lies with mild MR, which is the low extreme of the same genetic and environmental factors responsible for the normal distribution of intelligence. In other words, genes responsible for the substantial heritability of intelligence are the same genes responsible for MR. Finding genes for MR other than severe MR will involve the same problems faced throughout the life sciences for common disorders and complex traits: many genes of very small effect and ‘missing heritability’. A possible advantage for IQ is that it has a high end as well as a low end: It might be easier finding genes associated with high IQ than low IQ, even though these genes are expected to be associated with low IQ as well as MR.
Methods. We used data from 3 million 18-year-old males assessed for cognitive
abilities as part of compulsory military service in Sweden 1950-1990, which included
370,000 sibling pairs and 9000 twin pairs. We then linked these individuals to the
Swedish National Patient Register to identify 308 siblings with a diagnosis of severe
MR and 813 siblings with a diagnosis of mild MR. Two major findings emerged from
our analyses of sibling and twin correlations and concordances and model-fitting. First, severe MR is not heritable in that siblings of individuals with severe MR had IQ means and variances not significantly different from the population. Second, mild MR as well as the lowest 3% of IQ scores in the population is familial, heritable, and caused by the same genetic and environmental factors responsible for the normal distribution of intelligence.


After that there will be an interview with Prof Nick Mackintosh carried out by David Lubinski. These are always recorded, and will be available later on the ISIR website.

1 comment:

  1. The mental retardation one sounds very interesting. Is there a material effect from babies being born very prematurely? Or from being part of multiple births?

    If so, how do they incorporate those into their analysis?