Missing Heritability - Reading PDF

Missing Heritability - Reading 25 November 2023 18:44 Source Notes Genetics and intelligence differences: five special findings Introduction (Plomin and Deary, 2015) Intelligence was one of the first behavioral traits studied using quantitative genetic designs. Intelligence has become a target of mo...

Missing Heritability - Reading 25 November 2023 18:44 Source Notes Genetics and intelligence differences: five special findings Introduction (Plomin and Deary, 2015) Intelligence was one of the first behavioral traits studied using quantitative genetic designs. Intelligence has become a target of molecular genetic studies attempting to identify genes responsible for its heritability. This review will focus on genetic findings that are specific to intelligence. THREE ‘LAWS’ OF THE GENETICS OF COMPLEX TRAITS (INCLUDING INTELLIGENCE) ○ All traits show significant genetic influence. ○ No traits are 100% heritable. ○ Heritability is caused by many genes of small effect. FIVE FINDINGS FROM GENETIC RESEARCH THAT ARE SPECIFIC TO INTELLIGENCE ○ Intelligence is highly heritable. ○ The same genes influence intelligence across different populations. ○ There is substantial genetic overlap between intelligence and other cognitive abilities. ○ Intelligence is influenced by both common and rare variants. ○ Environmental factors play a significant role in intelligence. CONCLUSION ○ Genetic research has made significant progress in understanding the genetic basis of intelligence. ○ There is still much to learn about the specific genes that influence intelligence. ○ Intelligence is a complex trait that is influenced by both genetic and environmental factors. What is intelligent and why is it important? Defining Intelligence ○ Intelligence is a general cognitive ability that represents individual differences in brain processes working together to solve problems. ○ Intelligence is also known as general cognitive ability or g. ○ Intelligence is measured using cognitive ability tests, which assess a variety of cognitive skills, such as reasoning, problem-solving, and abstract thinking. ○ Intelligence is a hierarchical trait, with a general intelligence factor at the top, followed by group factors, and then specific tests and their associated narrow cognitive skills. Importance of Intelligence ○ Intelligence is important scientifically because it is central to systems approaches to brain structure and function and to understanding how cognitive abilities decline with age. ○ Intelligence is important socially because it is one of the best predictors of key outcomes such as education, occupational status, mental and physical health, and longevity. Five Genetic Findings Specific to Intelligence Differences ○ Dramatic increases in heritability during the life course: Intelligence becomes more heritable as people age. ○ High genetic correlations among diverse cognitive abilities: There is a strong genetic correlation between intelligence and other cognitive abilities. ○ High assortative mating: People tend to mate with others of similar intelligence. ○ The positive genetics of high intelligence: There are genetic factors that contribute to high intelligence. ○ The impact of intelligence on 'social epidemiology': Intelligence can influence health and disease outcomes through social factors. Heritability increases dramatically from infancy through adulthood despite genetic stability Introduction ▪ Heritability is the proportion of the variance in a trait that is due to genetic factors. ▪ The heritability of intelligence increases dramatically from childhood to adulthood. ▪ This is despite the fact that the same genes affect intelligence from age to age. Possible Explanations ▪ Genotype-environment correlation: Individuals may select, modify, and create environments that are correlated with their genetic predispositions. ▪ Genetic amplification: Small genetic differences may become magnified as individuals develop. Evidence ▪ A recent twin study found that the genetic correlation of intelligence between ages 7 and 12 was 0.75, despite increasing heritability from 0.36 to 0.49. ▪ A 60-year longitudinal study of intelligence found that the genetic correlation of intelligence between ages 11 and 69 was 0.62. The first new quantitative genetic method in a century: Genome -wide Complex Trait Analysis (GCTA) Introduction ▪ A new method called GCTA (or Genomic-Relatedness-Matrix Restricted Maximum Likelihood) can estimate the net effect of genetic influence using DNA of unrelated individuals. ▪ GCTA uses genetic similarity to predict phenotypic similarity. ▪ GCTA requires samples of thousands of individuals because the method attempts to extract a small signal of genetic similarity from the noise of hundreds of thousands of SNPs. ▪ GCTA detects only those genetic effects tagged by the common SNPs that have until recently been incorporated in commercially available DNA arrays used in GWA studies. Comparison to Twin Studies ▪ GCTA heritability represents the upper limit for detection of SNP associations in GWA studies. ▪ GCTA heritability represents the lower limit for heritability estimated in twin studies. ▪ GCTA heritability estimates for intelligence are about half the heritability estimates from twin studies. Future Directions ▪ GCTA can be extended beyond univariate analysis to the bivariate analysis of the covariance between two traits. ▪ GCTA can be used to study the genetic architecture of complex traits, including intelligence. Intelligence indexes general genetic effects across diverse cognitive and learning abilities ○ Intelligence is a general cognitive ability that is influenced by a small number of genes. ○ These genes also influence other cognitive abilities, such as spatial ability, vocabulary, processing speed, executive function, and memory. ○ The genetic correlation between intelligence and other cognitive abilities is typically greater than 0.6. ○ This means that the same genes are responsible for the heritabilities of these tests. ○ General genetic effects also permeate education-related learning abilities such as reading and arithmetic. ○ The genetic correlation between intelligence and learning abilities is uniformly high. ○ This means that the same genes are responsible for the heritabilities of these abilities. ○ GCTA estimates of genetic correlation between intelligence and learning abilities are highly similar to twin study estimates. ○ This means that GCTA is a valid method for studying the genetic architecture of complex traits. Assortative mating is greater for intelligence than for other traits ○ Assortative mating is the tendency for people to marry others who are similar to them in terms of intelligence, personality,and other traits. ○ Assortative mating for intelligence is about 0.40, which is higher than for most other traits. ○ Assortative mating for verbal intelligence is about 0.50, while assortative mating for nonverbal intelligence is about 0.30. ○ Assortative mating for intelligence is caused by initial selection of a mate rather than by couples becoming more similar toeach other after living together. ○ Assortative mating for intelligence is important because it increases additive genetic variance in offspring. ○ Additive genetic variance refers to the independent effects of alleles or loci that 'add up'. ○ Assortative mating increases additive genetic variance because offspring receive a random sampling of half of each parent's genes. ○ The extra additive genetic variance for intelligence induced by assortative mating is important for three reasons: ▪ Parents share only additive genetic variance with their offspring. ▪ GCTA heritability should be greater for intelligence than for traits that show less assortative mating. ▪ Intelligence is a good target for GWA studies because GCTA estimates that a substantial amount of additive genetic variance influences intelligence. Thinking positively: the genetics of high intelligence Positive Distribution of Intelligence ▪ Intelligence exhibits a normal distribution, with high performance at one end and intellectual disability at the other. ▪ High intelligence correlates with exceptional societal achievements, documented in longitudinal studies. ▪ Exploration of positive traits' genetics highlights differences from the notion that genetic differences primarily cause disorders. Quantitative Genetic Research on Intelligence ▪ Genetic causes of high intelligence are quantitatively, not qualitatively, different from the rest of the distribution. ▪ Studies suggest that high intelligence (top 15% or 4%) is as heritable as the rest of the distribution. ▪ Genetic factors contributing to high intelligence are strongly correlated with normal intelligence variation. Genetic Discontinuity Hypothesis ▪ The Genetic Discontinuity Hypothesis suggests extreme high intelligence might have a different genetic basis. ▪ The concept of emergenesis proposes unique gene combinations contributing to high intelligence. Epistatic Traits and Twin Studies ▪ Contrary to the emergenesis hypothesis, twin studies haven't shown nonadditive genetic effects for high intelligence. ▪ Studies focused on the top 15% or 4% of intelligence, falling short of examining extreme genius (top 0.1%). Comparison with Intellectual Disability ▪ Most intellectual disability falls within the low end of the normal intelligence distribution. ▪ Severe intellectual disability appears to have a distinct etiology separate from normal intelligence. Genetic Integration of Intelligence Extremes ▪ A hypothesis suggests disruptions in normal intelligence development result from various mutations, while high intelligence requires a convergence of positive alleles. ▪ Genome-wide association studies aimed at extremely high intelligence haven't shown consistent associations with rare genetic variants. Significance for Psychiatric Genetics ▪ Polygenic scores from psychiatric disorder studies exhibit a normal distribution, implying positive extremes akin to negative ones. ▪ This prompts the consideration of individuals at the positive end—whether they merely possess low risk or potentially harbour unique abilities. Thinking Positively and Quantitatively ▪ Encouraging a shift in thinking from 'disorders' to 'dimensions' and from genetic 'risk' to genetic 'variability' is crucial for understanding positive genetics. Intelligence brings (some) genetics to ‘social’ epidemiology Intelligence, Education, and Social Class ▪ Intelligence, education, and social class are all correlated with each other. ▪ The causes of these correlations are complex and not fully understood. ▪ There is some evidence that these correlations are due to shared genetic and environmental causes. Twin and Family Studies ▪ Twin and family studies have shown that educational attainment and social class are somewhat heritable. ▪ The genetic correlation between intelligence and education is moderate to high. ▪ The genetic correlation between intelligence and social class is moderate. GCTA Studies ▪ GCTA studies have confirmed that there is a genetic basis for intelligence, education, and social class. ▪ The GCTA-estimated genetic correlation between intelligence and education is high. ▪ The GCTA-estimated genetic correlation between intelligence and social class is moderate. PSYC0036 Genes and Behaviour Page 1 Five special findings and polygenic scores 1. Heritability of intelligence increases dramatically from infancy through adulthood despite genetic stability. 2. The genetic correlation between intelligence and other cognitive abilities is high. 3. The genetic correlation between intelligence and educational attainment is high. 4. The genetic correlation between intelligence and social class is moderate. 5. There is evidence for non-additive genetic variance in intelligence. Polygenic scores ○ Polygenic scores are created by adding genotypic values across loci. ○ Polygenic scores can be used to aggregate genotypic scores for DNA variants known to be associated with a trait. ○ Genome-wide polygenic scores (GPS) include thousands of SNPs or even all SNPs on a DNA array weighted by the strength of their association. ○ GPS can theoretically account for all the heritability shown in GCTA. ○ GWA studies of intelligence and other traits have resulted in GPS that fall far short of GCTA estimates of heritability. ○ More variance in intelligence is likely to be explained with GPS derived from larger samples, whole-genome sequencing and more novel strategies. Applications of Polygenic Scores ▪ Polygenic scores can be used in the same way that candidate genes have been used. ▪ A GPS for intelligence would be like the other GPS (global positioning system) making it possible to triangulate on the genetics of intelligence from all domains of the life sciences. Intelligence and General Genetic Effects ○ Intelligence indexes general genetic effects across diverse cognitive and learning abilities. ○ A GPS for intelligence should predict other cognitive abilities. ○ A GPS for intelligence should predict better than a GPS for any other trait. ○ A pleiotropic GPS that targets the covariance among cognitive and learning abilities will be better than a GPS based on a single measure of intelligence. Assortative Mating ○ Assortative mating is greater for intelligence than for any other trait. ○ GPS could provide evidence that assortative mating for intelligence is mediated genetically by correlating GPS between spouses. ○ It is an open question whether GPS assortative mating is greater for verbal than for nonverbal ability. Genetics of High Intelligence ○ The same genes affect high intelligence to the same extent as the rest of the normal distribution. ○ A GPS for intelligence from unselected samples can also be used to predict high intelligence. Intelligence and Social Epidemiology ○ The same genes influence intelligence and social epidemiologists' "environmental" variables of education, social class, and height. ○ GPS scores for intelligence might contribute to health outcomes and mortality. ○ GPS scores might account for some of the associations between education and class and mortality. Novel Test of Genetic Influence on Social Mobility ○ GPS scores can be used to test the extent to which genetic factors influence social mobility. Research Review: Polygenic methods and their application to psychiatric traits (Wray et al., 2014) Introduction Twin and family studies have shown that genetic factors play a significant role in the development of psychiatric disorders. Recent advances in technology have made it possible to test large numbers of genetic variants for association with psychiatri c disorders. Genome-wide association studies (GWAS) have been conducted for a number of psychiatric disorders, but the results have been disappoi nting. The first phase of GWAS for the major psychiatric disorders found few or no genome -wide significant hits. These results raised the question whether common variants are of sufficient relevance in the development of psychiatric disor ders to pursue with GWAS. Methods ○ This review describes methods that use data from GWAS to provide critical empirical evidence of an important polygenic contribution to common psychiatric disorders. ○ These methods include: ▪ Heritability estimates: These estimates measure the proportion of variation in a trait that is due to genetic factors. ▪ Genome-wide complex trait analysis (GCTA): This method uses GWAS data to estimate the proportion of phenotypic variance that is explained by all common SNPs in the genome. ▪ Linkage disequilibrium-score (LDS) mapping: This method uses GWAS data to identify regions of the genome that are likely to harbour risk variants for a trait. Applications ○ These methods have been applied to a number of psychiatric disorders and related phenotypes, including: ▪ Schizophrenia: GCTA studies have shown that common SNPs account for about 18% of the variance in schizophrenia. ▪ Bipolar disorder: GCTA studies have shown that common SNPs account for about 23% of the variance in bipolar disorder. ▪ Major depressive disorder (MDD): GCTA studies have shown that common SNPs account for about 33% of the variance in MDD. ▪ Autism spectrum disorders (ASD): GCTA studies have shown that common SNPs account for about 50% of the variance in ASD. ▪ Attention deficit hyperactivity disorder (ADHD): GCTA studies have shown that common SNPs account for about 73% of the variance in ADHD. Heritability Heritability of Psychiatric Disorders ○ Increased risk of psychiatric disorders in relatives of affected individuals provides evidence for a genetic contribution. ○ Estimates of risks to relatives are used to estimate heritability on the liability scale (h2), which quantifies the proportion of variance of liability to disease attributable to inherited genetic factors. ○ Non-genetic factors include identifiable environmental factors, measurement error, and unidentifiable factors that form an intrinsic stochastic noise. ○ Heritability estimates on the liability scale depend on knowledge of baseline risk of disease in the population and may vary between populations, across ages, and depending on whether nongenetic factors have been recorded and included in the analysis. ○ Heritability estimates should be viewed as pragmatic benchmarks representing evidence for low, moderate, or high contributions of genetic effects. ○ Heritability on the liability scale tells nothing about the underlying genetic architecture of the disease, such as the number, frequency, and effect sizes of individual causal variants, or the mode of action of causal loci. ○ Under a polygenic model, liability to disease is assumed to reflect multiple genetic and nongenetic effects acting additively. ○ Liabilities are assumed to be normally distributed because such a distribution results from many additively acting effects. ○ Most individuals in the population carry some genetic risk variants and likely experience some nongenetic risk factors, but most individuals in the population are not affected. ○ Disease status results when the cumulative load exceeds a burden of risk threshold. Missing heritability GWAS GWAS identify associations between SNPs and disease. Reported results from association analyses include risk allele frequency (RAF), effect size (expressed for disease as the odd s ratio, OR), and p-value of association. The contribution of these genetic variants to variance can be calculated on the liability scale (Risch, 2000; Sham, 1998; So, Gui, Cherny, & Sham, 2011) to allow direct comparison of the contribution to risk of each locus on the same scale as heritabi lity is reported. Assuming independence, the contribution of each genome -wide significant (GWS) locus can be summed to determine the proportion of variance in liability explained by these loci toget her, thus quantifying the effects of all genome -wide significant SNPs. This is denoted by h2_GWS. Given the stringent significance threshold applied, the ability to detect risk loci (i.e., the power) depends on whether the sample size is sufficient given the effect sizes. When the first GWAS were planned, the distribution of expected effect sizes was unknown, and sample sizes were powered to det ect OR > ~1.3. These GWAS yielded few GWS results, with h2_GWS much less than h2. This difference has been termed "missing heritability" (Manolio et al., 2009). As sample sizes have increased, the number of GWS variants has increased for both quantitative traits and diseases (see figur e 2 in Visscher et al. (2012)), providing empirical evidence that common variants do play a role in complex genetic traits. Currently, GWS variants explain