A powerful and efficient two-stage method for detecting gene-to-gene interactions in GWAS.

For over a decade functional gene-to-gene interaction (epistasis) has been suspected to be a determinant in the "missing heritability" of complex traits. However, searching for epistasis on the genome-wide scale has been challenging due to the prohibitively large number of tests which result in a serious loss of statistical power as well as computational challenges. In this article, we propose a two-stage method applicable to existing case-control data sets, which aims to lessen both of these problems by pre-assessing whether a candidate pair of genetic loci is involved in epistasis before it is actually tested for interaction with respect to a complex phenotype. The pre-assessment is based on a two-locus genotype independence test performed in the sample of cases. Only the pairs of loci that exhibit non-equilibrium frequencies are analyzed via a logistic regression score test, thereby reducing the multiple testing burden. Since only the computationally simple independence tests are performed for all pairs of loci while the more demanding score tests are restricted to the most promising pairs, genome-wide association study (GWAS) for epistasis becomes feasible. By design our method provides strong control of the type I error. Its favourable power properties especially under the practically relevant misspecification of the interaction model are illustrated. Ready-to-use software is available. Using the method we analyzed Parkinson's disease in four cohorts and identified possible interactions within several SNP pairs in multiple cohorts.

Accurate prediction of a minimal region around a genetic association signal that contains the causal variant.

In recent years, genome-wide association studies have been very successful in identifying loci for complex traits. However, typically these findings involve noncoding and/or intergenic SNPs without a clear functional effect that do not directly point to a gene. Hence, the challenge is to identify the causal variant responsible for the association signal. Typically, the first step is to identify all genetic variation in the locus region, usually by resequencing a large number of case chromosomes. Among all variants, the causal one needs to be identified in further functional studies. Because the experimental follow up can be very laborious, restricting the number of variants to be scrutinized can yield a great advantage. An objective method for choosing the size of the region to be followed up would be highly valuable. Here, we propose a simple method to call the minimal region around a significant association peak that is very likely to contain the causal variant. We model linkage disequilibrium (LD) in cases from the observed single SNP association signals, and predict the location of the causal variant by quantifying how well this relationship fits the data. Simulations showed that our approach identifies genomic regions of on average ∼50 kb with up to 90% probability to contain the causal variant. We apply our method to two genome-wide association data sets and localize both the functional variant REP1 in the α-synuclein gene that conveys susceptibility to Parkinson's disease and the APOE gene responsible for the association signal in the Alzheimer's disease data set.

Resequencing three candidate genes for major depressive disorder in a Dutch cohort.

Major depressive disorder (MDD) is a psychiatric disorder, characterized by periods of low mood of more than two weeks, loss of interest in normally enjoyable activities and behavioral changes. MDD is a complex disorder and does not have a single genetic cause. In 2009 a genome wide association study (GWAS) was performed on the Dutch GAIN-MDD cohort. Many of the top signals of this GWAS mapped to a region spanning the gene PCLO, and the non-synonymous coding single nucleotide polymorphism (SNP) rs2522833 in the PCLO gene became genome wide significant after post-hoc analysis. We performed resequencing of PCLO, GRM7, and SLC6A4 in 50 control samples from the GAIN-MDD cohort, to detect new genomic variants. Subsequently, we genotyped these variants in the entire GAIN-MDD cohort and performed association analysis to investigate if rs2522833 is the causal variant or simply in linkage disequilibrium with a more associated variant. GRM7 and SLC6A4 are both candidate genes for MDD from literature. We aimed to gather more evidence that rs2522833 is indeed the causal variant in the GAIN-MDD cohort or to find a previously undetected common variant in either PCLO, GRM7, or SLC6A4 with a higher association in this cohort. After next generation sequencing and association analysis we excluded the possibility of an undetected common variant to be more associated. For neither PCLO nor GRM7 we found a more associated variant. For SLC6A4, we found a new SNP that showed a lower P-value (P = 0.07) than in the GAIN-MDD GWAS (P = 0.09). However, no evidence for genome-wide significance was found. Although we did not take into account rare variants, we conclude that our results provide further support for the hypothesis that the non-synonymous coding SNP rs2522833 in the PCLO gene is indeed likely to be the causal variant in the GAIN-MDD cohort.

Modulatory effects of the piccolo genotype on emotional memory in health and depression.

Major depressive disorder (MDD) has been associated with biased memory formation for mood-congruent information, which may be related to altered monoamine levels. The piccolo (PCLO) gene, involved in monoaminergic neurotransmission, has previously been linked to depression in a genome-wide association study. Here, we investigated the role of the PCLO risk allele on functional magnetic resonance imaging (MRI) correlates of emotional memory in a sample of 89 MDD patients (64 PCLO risk allele carriers) and 29 healthy controls (18 PCLO risk allele carriers). During negative word encoding, risk allele carriers showed significant lower activity relative to non-risk allele carriers in the insula, and trend-wise in the anterior cingulate cortex and inferior frontal gyrus. Moreover, depressed risk allele carriers showed significant lower activity relative to non-risk allele carriers in the striatum, an effect which was absent in healthy controls. Finally, amygdalar response during processing new positive words vs. known words was blunted in healthy PCLO+ carriers and in MDD patients irrespective of genotype, which may indicate that signalling of salient novel information does not occur to the same extent in PCLO+ carriers and MDD patients. The PCLO risk allele may increase vulnerability for MDD by modulating local brain function with regard to responsiveness to salient stimuli (i.e. insula) and processing novel negative information. Also, depression-specific effects of PCLO on dorsal striatal activation during negative word encoding and the absence of amygdalar salience signalling for novel positive information further suggest a role of PCLO in symptom maintenance in MDD.

Cervical dystonia and genetic common variation in the dopamine pathway.

Cervical dystonia, a late onset focal dystonia, has a complex genetic background. Multiple lines of evidence point to a role for aberrant dopamine levels in dystonia. We assessed whether common variation within genes that regulate brain dopamine levels and in key genes of the dopamine metabolic pathway, modulate the risk for cervical dystonia. DNA was collected from 363 Dutch CD patients and a cohort of Dutch control individuals. Haplotype-tagging single nucleotide polymorphisms (SNPs) complemented with selected variants of functional importance in COMT, DAT, TH, MAO-A and -B, DDC and DBH were investigated. We tested the 143 markers in single-SNP, haplotype and epistasis analyses. We did not find an association with any of the selected 143 SNPs in these key dopamine genes. Our data shows that common variations in key genes of the dopamine pathway do not contribute to dystonia risk in the Dutch population. Possibly, risk alleles in this pathway may be rarer than detectable in this study, or might be located in downstream dopamine signaling pathway. Alternatively, found dopamine level changes are secondary to the dystonia disease processes.

Use of support vector machines for disease risk prediction in genome-wide association studies: concerns and opportunities.

The success of genome-wide association studies (GWAS) in deciphering the genetic architecture of complex diseases has fueled the expectations whether the individual risk can also be quantified based on the genetic architecture. So far, disease risk prediction based on top-validated single-nucleotide polymorphisms (SNPs) showed little predictive value. Here, we applied a support vector machine (SVM) to Parkinson disease (PD) and type 1 diabetes (T1D), to show that apart from magnitude of effect size of risk variants, heritability of the disease also plays an important role in disease risk prediction. Furthermore, we performed a simulation study to show the role of uncommon (frequency 1-5%) as well as rare variants (frequency <1%) in disease etiology of complex diseases. Using a cross-validation model, we were able to achieve predictions with an area under the receiver operating characteristic curve (AUC) of ~0.88 for T1D, highlighting the strong heritable component (∼90%). This is in contrast to PD, where we were unable to achieve a satisfactory prediction (AUC ~0.56; heritability ~38%). Our simulations showed that simultaneous inclusion of uncommon and rare variants in GWAS would eventually lead to feasible disease risk prediction for complex diseases such as PD. The used software is available at http://www.ra.cs.uni-tuebingen.de/software/MACLEAPS/.

A fine-mapping study of 7 top scoring genes from a GWAS for major depressive disorder.

Major depressive disorder (MDD) is a psychiatric disorder that is characterized--amongst others--by persistent depressed mood, loss of interest and pleasure and psychomotor retardation. Environmental circumstances have proven to influence the aetiology of the disease, but MDD also has an estimated 40% heritability, probably with a polygenic background. In 2009, a genome wide association study (GWAS) was performed on the Dutch GAIN-MDD cohort. A non-synonymous coding single nucleotide polymorphism (SNP) rs2522833 in the PCLO gene became only nominally significant after post-hoc analysis with an Australian cohort which used similar ascertainment. The absence of genome-wide significance may be caused by low SNP coverage of genes. To increase SNP coverage to 100% for common variants (m.a.f.>0.1, r(2)>0.8), we selected seven genes from the GAIN-MDD GWAS: PCLO, GZMK, ANPEP, AFAP1L1, ST3GAL6, FGF14 and PTK2B. We genotyped 349 SNPs and obtained the lowest P-value for rs2715147 in PCLO at P = 6.8E-7. We imputed, filling in missing genotypes, after which rs2715147 and rs2715148 showed the lowest P-value at P = 1.2E-6. When we created a haplotype of these SNPs together with the non-synonymous coding SNP rs2522833, the P-value decreased to P = 9.9E-7 but was not genome wide significant. Although our study did not identify a more strongly associated variant, the results for PCLO suggest that the causal variant is in high LD with rs2715147, rs2715148 and rs2522833.

Extended kindred with recessive late-onset Alzheimer disease maps to locus 8p22-p21.2: a genome-wide linkage analysis.

Late-onset Alzheimer disease (LOAD) is a complex genetic disorder. Although genes involved in early-onset forms were discovered more than a decade ago, LOAD research has only been able to point out small effect loci, with the exception of APOE. We mapped the gene predisposing to LOAD in an extended inbred family coming from a genetically isolated region (24 sampled individuals, 12 of whom are affected), completing a genome-wide screen with an Affymetrix10 K single nucleotide polymorphism microarray. Genotyping results were evaluated under model-dependent (dominant and recessive) and model-free analysis. We obtained a maximum nonparametric linkage score of 3.24 (P=0.00006) on chromosome 8p22-p21.2. The same genomic position also yielded the highest multipoint heterogeneity LOD (HLOD) under a recessive model (HLOD=3.04). When we compared the results of the model-dependent analysis, a higher score was obtained in the recessive model (3.04) than in the dominant model (1.0). This is a new locus identified in LOAD, in chromosome 8p22-p21.2 and encompassing several candidate genes, among them CLU and PPP3CC that were excluded by sequencing. The finding of a recessive model of inheritance, consistent with the assumption of inbreeding as a morbidity factor in this population, supports the notion of a role of recessive genes in LOAD.

The complete automation of cell culture: improvements for high-throughput and high-content screening.

Genomic approaches provide enormous amounts of raw data with regard to genetic variation, the diversity of RNA species, and protein complement. High-throughput (HT) and high-content (HC) cellular screens are ideally suited to contextualize the information gathered from other "omic" approaches into networks and can be used for the identification of therapeutic targets. Current methods used for HT-HC screens are laborious, time-consuming, and prone to human error. The authors thus developed an automated high-throughput system with an integrated fluorescent imager for HC screens called the AI.CELLHOST. The implementation of user-defined culturing and assay plate setup parameters allows parallel operation of multiple screens in diverse mammalian cell types. The authors demonstrate that such a system is able to successfully maintain different cell lines in culture for extended periods of time as well as significantly increasing throughput, accuracy, and reproducibility of HT and HC screens.

Probing the roles of LRR RLK genes in Arabidopsis thaliana roots using a custom T-DNA insertion set.

Leucine-rich repeat receptor-like protein kinases (LRR RLKs) represent the largest group of Arabidopsis RLKs with approximately 235 members. A minority of these LRR RLKs have been assigned to diverse roles in development, pathogen resistance and hormone perception. Using a reverse genetics approach, a collection of homozygous T-DNA insertion lines for 69 root expressed LRR RLK genes was screened for root developmental defects and altered response after exposure to environmental, hormonal/chemical and abiotic stress. The obtained data demonstrate that LRR RLKs play a role in a wide variety of signal transduction pathways related to hormone and abiotic stress responses. The described collection of T-DNA insertion mutants provides a valuable tool for future research into the function of LRR RLK genes.

Genome-wide association study confirms extant PD risk loci among the Dutch.

In view of the population-specific heterogeneity in reported genetic risk factors for Parkinson's disease (PD), we conducted a genome-wide association study (GWAS) in a large sample of PD cases and controls from the Netherlands. After quality control (QC), a total of 514,799 SNPs genotyped in 772 PD cases and 2024 controls were included in our analyses. Direct replication of SNPs within SNCA and BST1 confirmed these two genes to be associated with PD in the Netherlands (SNCA, rs2736990: P = 1.63 × 10(-5), OR = 1.325 and BST1, rs12502586: P = 1.63 × 10(-3), OR = 1.337). Within SNCA, two independent signals in two different linkage disequilibrium (LD) blocks in the 3' and 5' ends of the gene were detected. Besides, post-hoc analysis confirmed GAK/DGKQ, HLA and MAPT as PD risk loci among the Dutch (GAK/DGKQ, rs2242235: P = 1.22 × 10(-4), OR = 1.51; HLA, rs4248166: P = 4.39 × 10(-5), OR = 1.36; and MAPT, rs3785880: P = 1.9 × 10(-3), OR = 1.19).

Transcriptional plasticity of a soil arthropod across different ecological conditions.

Ecological functional genomics, dealing with the responses of organisms to their natural environment is confronted with a complex pattern of variation and a large number of confounding environmental factors. For gene expression studies to provide meaningful information on conditions deviating from normal, a baseline or normal operating range (NOR) response needs to be established which indicates how an organism's transcriptome reacts to naturally varying ecological factors. Here we determine the transcriptional plasticity of a soil arthropod, Folsomia candida, exposed to various natural environments, as part of a first attempt in establishing such a NOR. Animals were exposed to 26 different field soils after which gene expression levels were measured. The main factor found to regulate gene expression was soil-type (sand or clay). Cell homeostasis and DNA replication were affected in collembolans exposed to sandy soil, indicating general stress. Multivariate analysis identified soil fertility as the main factor influencing gene expression. Regarding land-use, only forest soils showed an expression pattern deviating from the others. No significant effect of land-use, agricultural practice or soil type on fitness was observed, but arsenic concentration was negatively correlated with reproductive output. In conclusion, transcriptional responses remained within a limited range across the different land-uses but were significantly affected by soil-type. This may be caused by the contrasting soil physicochemical properties to which F. candida strongly responds. The broad range of conditions over which this soil-living detritivore is able to survive and reproduce, indicates a strategy of high plasticity, which comes with extensive gene expression regulation.

Do consanguineous parents of a child affected by an autosomal recessive disease have more DNA identical-by-descent than similarly-related parents with healthy offspring? Design of a case-control study.

BACKGROUND: The offspring of consanguineous relations have an increased risk of congenital/genetic disorders and early mortality. Consanguineous couples and their offspring account for approximately 10% of the global population. The increased risk for congenital/genetic disorders is most marked for autosomal recessive disorders and depends on the degree of relatedness of the parents. For children of first cousins the increased risk is 2-4%. For individual couples, however, the extra risk can vary from zero to 25% or higher, with only a minority of these couples having an increased risk of at least 25%. It is currently not possible to differentiate between high-and low-risk couples. The quantity of DNA identical-by-descent between couples with the same degree of relatedness shows a remarkable variation. Here we hypothesize that consanguineous partners with children affected by an autosomal recessive disease have more DNA identical-by-descent than similarly-related partners who have only healthy children. The aim of the study is thus to establish whether the amount of DNA identical-by-descent in consanguineous parents of children with an autosomal recessive disease is indeed different from its proportion in consanguineous parents who have healthy children only. METHODS/DESIGN: This project is designed as a case-control study. Cases are defined as consanguineous couples with one or more children with an autosomal recessive disorder and controls as consanguineous couples with at least three healthy children and no affected child. We aim to include 100 case couples and 100 control couples. Control couples are matched by restricting the search to the same family, clan or ethnic origin as the case couple. Genome-wide SNP arrays will be used to test our hypothesis. DISCUSSION: This study contains a new approach to risk assessment in consanguineous couples. There is no previous study on the amount of DNA identical-by-descent in consanguineous parents of affected children compared to the consanguineous parents of healthy children. If our hypothesis proves to be correct, further studies are needed to obtain different risk figure estimates for the different proportions of DNA identical-by-descent. With more precise information about their risk status, empowerment of couples can be improved when making reproductive decisions.

The role of the COMT Val(158)Met polymorphism in the phenotypic expression of obsessive-compulsive disorder.

Obsessive-Compulsive Disorder (OCD) is characterized by the presence of obsessions and compulsions, and shows considerable phenotypic variability. Family and twin studies have indicated a genetic component in the etiology of OCD, and the catechol-O-methyl transferase (COMT) gene is an important candidate gene for OCD. This study investigates the influence of the functional COMT Val158Met polymorphism on the phenotypic expression of OCD, using an item-level factor-analytic approach in a large sample. The COMT Val158Met variant was genotyped in 373 patients and 462 controls. It was tested whether there was an association between the COMT Val158Met polymorphism and OCD or dimensional phenotypes such as YBOCS severity score, age of onset of obsessive-compulsive symptoms and six symptom dimensions recently found in a large item-level factor-analytic study [Katerberg et al., submitted]. We further investigated possible sex-specific associations between the COMT Val158Met polymorphism and OCD or dimensional phenotypes. There was a trend for an association of the COMT 158Met allele with OCD in males, and an interaction between the COMT Val158Met genotype and sex on the somatic and sensory phenomena symptom dimension, with females showing lower scores. In conclusion, a dimensional approach seems fruitful in detecting genes of importance for OCD.

Variation at GRN 3'-UTR rs5848 is not associated with a risk of frontotemporal lobar degeneration in Dutch population.

BACKGROUND: A single nucleotide polymorphism (rs5848) located in the 3'- untranslated region of GRN has recently been associated with a risk of frontotemporal lobar degeneration (FTLD) in North American population particularly in pathologically confirmed cases with neural inclusions immunoreactive for ubiquitin and TAR DNA-binding protein 43 (TDP-43), but negative for tau and alpha-synuclein (FTLD-TDP). METHODOLOGY/PRINCIPAL FINDINGS: In an effort to replicate these results in a different population, rs5848 was genotyped in 256 FTLD cases and 1695 controls from the Netherlands. Single SNP gender-adjusted logistic regression analysis revealed no significant association between variation at rs5848 and FTLD. Fisher's exact test, failed to find any significant association between rs5848 and a subset of 23 pathology confirmed FTLD-TDP cases. CONCLUSIONS/SIGNIFICANCE: The evidence presented here suggests that variation at rs5848 does not contribute to the etiology of FTLD in the Dutch population.

Comprehensive mRNA expression profiling distinguishes tauopathies and identifies shared molecular pathways.

BACKGROUND: Understanding the aetiologies of neurodegenerative diseases such as Alzheimer's disease (AD), Pick's disease (PiD), Progressive Supranuclear Palsy (PSP) and Frontotemporal dementia (FTD) is often hampered by the considerable clinical and molecular overlap between these diseases and normal ageing. The development of high throughput genomic technologies such as microarrays provide a new molecular tool to gain insight in the complexity and relationships between diseases, as they provide data on the simultaneous activity of multiple genes, gene networks and cellular pathways. METHODOLOGY/PRINCIPAL FINDINGS: We have constructed genome wide expression profiles from snap frozen post-mortem tissue from the medial temporal lobe of patients with four neurodegenerative disorders (5 AD, 5 PSP, 5 PiD and 5 FTD patients) and 5 control subjects. All patients were matched for age, gender, ApoE-epsilon and MAPT (tau) haplotype. From all groups a total of 790 probes were shown to be differently expressed when compared to control individuals. The results from these experiments were then used to investigate the correlations between clinical, pathological and molecular findings. From the 790 identified probes we extracted a gene set of 166 probes whose expression could discriminate between these disorders and normal ageing. CONCLUSIONS/SIGNIFICANCE: From genome wide expression profiles we extracted a gene set of 166 probes whose expression could discriminate between neurological disorders and normal ageing. This gene set can be further developed into an accurate microarray-based classification test. Furthermore, from this dataset we extracted a disease specific set of genes and identified two aging related transcription factors (FOXO1A and FOXO3A) as possible drug targets related to neurodegenerative disease.

Adaptive differences in gene expression associated with heavy metal tolerance in the soil arthropod Orchesella cincta.

Field-selected tolerance to heavy metals has been reported for Orchesella cincta (Arthropoda: Collembola) populations occurring at metal-contaminated mining sites. This tolerance correlated with heritable increase in metal excretion efficiency, less pronounced cadmium (Cd)-induced growth reduction and overexpression of the metallothionein gene. We applied transcriptomics to determine differential gene expression caused by this abiotic stress in reference and Cd-tolerant populations. Many cDNAs responded to Cd exposure in the reference population. Significantly fewer clones were Cd responsive in tolerant animals. Analysis of variance revealed transcripts that interact between Cd exposure and population. Hierarchical cluster analysis of these clones identified two major groups. The first one contained cDNAs that were up-regulated by Cd in the reference culture but non-responsive or down-regulated in tolerant animals. This cluster was also characterized by elevated constitutive expression in the tolerant population. Gene ontology analysis revealed that these cDNAs were involved in structural integrity of the cuticle, anti-microbial defence, calcium channel-blocking, sulphur assimilation and chromatin remodelling. The second group consisted of cDNAs down-regulated in reference animals but not responding or slightly up-regulated in tolerant animals. Their functions involved carbohydrate metabolic processes, Ca(2+)-dependent stress signalling, redox state, proteolysis and digestion. The reference population showed a strong signature of stress-induced genome-wide perturbation of gene expression, whereas the tolerant animals maintained normal gene expression upon Cd exposure. We confirmed the micro-evolutionary processes occurring in soil arthropod populations and suggest a major contribution of gene regulation to the evolution of a stress-adapted phenotype.

The role of the brain-derived neurotrophic factor (BDNF) val66met variant in the phenotypic expression of obsessive-compulsive disorder (OCD).

Evidence suggests that the Val66Met variant of the brain-derived neurotrophic factor (BDNF) gene may play a role in the etiology of Obsessive-Compulsive Disorder (OCD). In this study, the role of the BDNF Val66Met variant in the etiology and the phenotypic expression of OCD is investigated. Associations between the BDNF Val66Met variant and OCD, obsessive-compulsive symptom dimensions, Yale-Brown Obsessive Compulsive Scale (YBOCS) severity scores, age of onset and family history of obsessive-compulsive symptoms were assessed. The BDNF Val66Met variant was genotyped in 419 patients with sub-/clinical OCD and 650 controls. No differences in allele or genotype frequency were observed between cases and controls. In females with OCD, the Met66Met genotype was associated with later age of onset and a trend for a negative family history, whereas the Val66Val genotype was associated with a trend for lower YBOCS severity scores. Item-level factor analysis revealed six factors: 1) Contamination/cleaning; 2) Aggressive obsessions/checking; 3) Symmetry obsessions, counting, ordering and repeating; 4) Sexual/religious obsessions; 5) Hoarding and 6) Somatic obsessions/checking. A trend was found for a positive association between Factor 4 (Sexual/religious obsessions) and the BDNF Val66Val genotype. The results suggest that BDNF function may be implicated in the mediation of OCD. We found that for the BDNF Met66Met genotype may be associated with a milder phenotype in females and a possible role for the BDNF Val66Val genotype and the BDNF Val66 allele in the sexual/religious obsessions.

Global similarity with local differences in linkage disequilibrium between the Dutch and HapMap-CEU populations.

The HapMap project has facilitated the selection of tagging single nucleotide polymorphisms (tagSNPs) for genome-wide association studies (GWAS) under the assumption that linkage disequilibrium (LD) in the HapMap populations is similar to the populations under investigation. Earlier reports support this assumption, although in most of these studies only a few loci were evaluated. We compared pair-wise LD and LD block structure across autosomes between the Dutch population and the CEU-HapMap reference panel. The impact of sampling distribution on the estimation of LD blocks was studied by bootstrapping. A high Pearson correlation (genome-wide; 0.93) between pair-wise r(2) for the Dutch and the CEU populations was found, indicating that tagSNPs from the CEU-HapMap panel capture common variation in the Dutch population. However, some genomic regions exhibited, significantly lower correlation than the genome-wide estimate. This might decrease the validity of HapMap tagSNPs in these regions and the power of GWAS. The LD block structure differed considerably between the Dutch and CEU-HapMap populations. This was not explained by demographic differences between the CEU and Dutch samples, as testing for population stratification was not significant. We also found that sampling variation had a large effect on the estimation of LD blocks, as shown by the bootstrapping analysis. Thus, in small samples, most of the observed differences in LD blocks between populations are most likely the result of sampling variation. This poor concordance in LD block structure suggests that large samples are required for robust estimations of local LD block structure in populations.

A Functional polymorphism under positive evolutionary selection in ADRB2 is associated with human intelligence with opposite effects in the young and the elderly.

Comparative genomics offers a novel approach to unravel the genetic basis of complex traits. We performed a two stage analysis where genes ascertained for enhanced protein evolution in primates are subsequently searched for the presence of non-synonymous coding SNPs in the current human population at amino acid sites that differ between humans and chimpanzee. Positively selected genes among primates are generally presumed to determine phenotypic differences between humans and chimpanzee, such as the enhanced cognitive ability of our species. Amino acid substitutions segregating in humans at positively selected amino acid sites are expected to affect phenotypic differences among humans. Therefore we conducted an association study in two family based cohorts and one population based cohort between cognitive ability and the most likely candidate gene among the five that harbored more than one such polymorphism. The derived, human-specific allele of the beta-2 adrenergic receptor Arg16Gly polymorphism was found to be the increaser allele for performance IQ in the young, family based cohort but the decreaser allele for two different measures of cognition in the large Scottish cohort of unrelated individuals. The polymorphism is known to affect signaling activity and modulation of beta-2 adrenergic signaling has been shown to adjust memory consolidation, a trait related to cognition. The opposite effect of the polymorphism on cognition in the two age classes observed in the different cohorts resembles the effect of ADRB2 on hypertension, which also has been reported to be age dependent. This result illustrates the relevance of comparative genomics to detect genes that are involved in human behavior.