We explore the prediction of individuals' phenotypes for complex traits using genomic data. We compare several widely used prediction models, including Ridge Regression, LASSO and Elastic Nets estimated from cohort data, and polygenic risk scores constructed using published summary statistics from genome-wide association meta-analyses (GWAMA). We evaluate the interplay between relatedness, trait architecture and optimal marker density, by predicting height, body mass index (BMI) and high-density lipoprotein level (HDL) in two data cohorts, originating from Croatia and Scotland. We empirically demonstrate that dense models are better when all genetic effects are small (height and BMI) and target individuals are related to the training samples, while sparse models predict better in unrelated individuals and when some effects have moderate size (HDL). For HDL sparse models achieved good across-cohort prediction, performing similarly to the GWAMA risk score and to models trained within the same cohort, which indicates that, for predicting traits with moderately sized effects, large sample sizes and familial structure become less important, though still potentially useful. Finally, we propose a novel ensemble of whole-genome predictors with GWAMA risk scores and demonstrate that the resulting meta-model achieves higher prediction accuracy than either model on its own. We conclude that although current genomic predictors are not accurate enough for diagnostic purposes, performance can be improved without requiring access to large-scale individual-level data. Our methodologically simple meta-model is a means of performing predictive meta-analysis for optimizing genomic predictions and can be easily extended to incorporate multiple population-level summary statistics or other domain knowledge.

Menopause timing has a substantial impact on infertility and risk of disease, including breast cancer, but the underlying mechanisms are poorly understood. We report a dual strategy in ∼70,000 women to identify common and low-frequency protein-coding variation associated with age at natural menopause (ANM). We identified 44 regions with common variants, including two regions harboring additional rare missense alleles of large effect. We found enrichment of signals in or near genes involved in delayed puberty, highlighting the first molecular links between the onset and end of reproductive lifespan. Pathway analyses identified major association with DNA damage response (DDR) genes, including the first common coding variant in BRCA1 associated with any complex trait. Mendelian randomization analyses supported a causal effect of later ANM on breast cancer risk (∼6% increase in risk per year; P = 3 × 10(-14)), likely mediated by prolonged sex hormone exposure rather than DDR mechanisms.

Variants associated with blood lipid levels may be population-specific. To identify low-frequency variants associated with this phenotype, population-specific reference panels may be used. Here we impute nine large Dutch biobanks (~35,000 samples) with the population-specific reference panel created by the Genome of The Netherlands Project and perform association testing with blood lipid levels. We report the discovery of five novel associations at four loci (P value <6.61 × 10(-4)), including a rare missense variant in ABCA6 (rs77542162, p.Cys1359Arg, frequency 0.034), which is predicted to be deleterious. The frequency of this ABCA6 variant is 3.65-fold increased in the Dutch and its effect (βLDL-C=0.135, βTC=0.140) is estimated to be very similar to those observed for single variants in well-known lipid genes, such as LDLR.

Genome-wide association studies (GWAS) have provided valuable insights into the genetic basis of complex traits. However, they have explained relatively little trait heritability. Recently, we proposed a new analytical approach called regional heritability mapping (RHM) that captures more of the missing genetic variation. This method is applicable both to related and unrelated populations. Here, we demonstrate the power of RHM in comparison with single-SNP GWAS and gene-based association approaches under a wide range of scenarios with variable numbers of quantitative trait loci (QTL) with common and rare causal variants in a narrow genomic region. Simulations based on real genotype data were performed to assess power to capture QTL variance, and we demonstrate that RHM has greater power to detect rare variants and/or multiple alleles in a region than other approaches. In addition, we show that RHM can capture more accurately the QTL variance, when it is caused by multiple independent effects and/or rare variants. We applied RHM to analyze three biometrical eye traits for which single-SNP GWAS have been published or performed to evaluate the effectiveness of this method in real data analysis and detected some additional loci which were not detected by other GWAS methods. RHM has the potential to explain some of missing heritability by capturing variance caused by QTL with low MAF and multiple independent QTL in a region, not captured by other GWAS methods. RHM analyses can be implemented using the software REACTA (http://www.epcc.ed.ac.uk/projects-portfolio/reacta).

Individuals with exceptional longevity and their offspring have significantly larger high-density lipoprotein concentrations (HDL-C) particle sizes due to the increased homozygosity for the I405V variant in the cholesteryl ester transfer protein (CETP) gene. In this study, we investigate the association of CETP and HDL-C further to identify novel, independent CETP variants associated with HDL-C in humans.

Late-onset retinal degeneration (L-ORD) is a rare autosomal dominant retinal dystrophy, characterised by extensive sub-retinal pigment epithelium (RPE) deposits, RPE atrophy, choroidal neovascularisation and photoreceptor cell death associated with severe visual loss. L-ORD shows striking phenotypic similarities to age-related macular degeneration (AMD), a common and genetically complex disorder, which can lead to misdiagnosis in the early stages. To date, a single missense mutation (S163R) in the C1QTNF5 gene, encoding C1q And Tumor Necrosis Factor Related Protein 5 (C1QTNF5) has been shown to cause L-ORD in a subset of affected families. Here, we describe the identification and characterisation of three novel pathogenic mutations in C1QTNF5 in order to elucidate disease mechanisms. In silico and in vitro characterisation show that these mutations perturb protein folding, assembly or polarity of secretion of C1QTNF5 and, importantly, all appear to destabilise the wildtype protein in co-transfection experiments in a human RPE cell line. This suggests that the heterozygous mutations in L-ORD show a dominant negative, rather than a haploinsufficient, disease mechanism. The function of C1QTNF5 remains unclear but this new insight into the pathogenetic basis of L-ORD has implications for future therapeutic strategies such as gene augmentation therapy.

Age-related macular degeneration (AMD) is a common cause of blindness in older individuals. To accelerate the understanding of AMD biology and help design new therapies, we executed a collaborative genome-wide association study, including >17,100 advanced AMD cases and >60,000 controls of European and Asian ancestry. We identified 19 loci associated at P < 5 × 10(-8). These loci show enrichment for genes involved in the regulation of complement activity, lipid metabolism, extracellular matrix remodeling and angiogenesis. Our results include seven loci with associations reaching P < 5 × 10(-8) for the first time, near the genes COL8A1-FILIP1L, IER3-DDR1, SLC16A8, TGFBR1, RAD51B, ADAMTS9 and B3GALTL. A genetic risk score combining SNP genotypes from all loci showed similar ability to distinguish cases and controls in all samples examined. Our findings provide new directions for biological, genetic and therapeutic studies of AMD.

It is a longstanding puzzle why non-coding variants in the complement factor H (CFH) gene are more strongly associated with age-related macular degeneration (AMD) than functional coding variants that directly influence the alternative complement pathway. The situation is complicated by tight genetic associations across the region, including the adjacent CFH-related genes CFHR3 and CFHR1, which may themselves influence the alternative complement pathway and are contained within a common deletion (CNP147) which is associated with protection against AMD. It is unclear whether this association is mediated through a protective effect of low plasma CFHR1 concentrations, high plasma CFH or both. We examined the triangular relationships of CFH/CFHR3/CFHR1 genotype, plasma CFH or CFHR1 concentrations and AMD susceptibility in combined case-control (1256 cases, 1020 controls) and cross-sectional population (n = 1004) studies and carried out genome-wide association studies of plasma CFH and CFHR1 concentrations. A non-coding CFH SNP (rs6677604) and the CNP147 deletion were strongly correlated both with each other and with plasma CFH and CFHR1 concentrations. The plasma CFH-raising rs6677604 allele and raised plasma CFH concentration were each associated with AMD protection. In contrast, the protective association of the CNP147 deletion with AMD was not mediated by low plasma CFHR1, since AMD-free controls showed increased plasma CFHR1 compared with cases, but it may be mediated by the association of CNP147 with raised plasma CFH concentration. The results are most consistent with a regulatory locus within a 32 kb region of the CFH gene, with a major effect on plasma CFH concentration and AMD susceptibility.

Age-related macular degeneration (AMD) is the commonest cause of blindness in Western populations. Risk is influenced by age, genetic and environmental factors. Complement activation appears to be important in the pathogenesis and associations have been found between AMD and genetic variations in complement regulators such as complement factor H. We therefore investigated other complement regulators for association with AMD.

A single founder mutation resulting in a Ser163Arg substitution in the C1QTNF5 gene product causes autosomal dominant late-onset retinal macular degeneration (L-ORMD) in humans, which has clinical and pathological features resembling age-related macular degeneration. We generated and characterised a mouse "knock-in" model carrying the Ser163Arg mutation in the orthologous murine C1qtnf5 gene by site-directed mutagenesis and homologous recombination into mouse embryonic stem cells. Biochemical, immunological, electron microscopic, fundus autofluorescence, electroretinography and laser photocoagulation analyses were used to characterise the mouse model. Heterozygous and homozygous knock-in mice showed no significant abnormality in any of the above measures at time points up to 2 years. This result contrasts with another C1qtnf5 Ser163Arg knock-in mouse which showed most of the features of L-ORMD but differed in genetic background and targeting construct.

The male-to-female sex ratio at birth is constant across world populations with an average of 1.06 (106 male to 100 female live births) for populations of European descent. The sex ratio is considered to be affected by numerous biological and environmental factors and to have a heritable component. The aim of this study was to investigate the presence of common allele modest effects at autosomal and chromosome X variants that could explain the observed sex ratio at birth. We conducted a large-scale genome-wide association scan (GWAS) meta-analysis across 51 studies, comprising overall 114 863 individuals (61 094 women and 53 769 men) of European ancestry and 2 623 828 common (minor allele frequency >0.05) single-nucleotide polymorphisms (SNPs). Allele frequencies were compared between men and women for directly-typed and imputed variants within each study. Forward-time simulations for unlinked, neutral, autosomal, common loci were performed under the demographic model for European populations with a fixed sex ratio and a random mating scheme to assess the probability of detecting significant allele frequency differences. We do not detect any genome-wide significant (P < 5 × 10(-8)) common SNP differences between men and women in this well-powered meta-analysis. The simulated data provided results entirely consistent with these findings. This large-scale investigation across ~115 000 individuals shows no detectable contribution from common genetic variants to the observed skew in the sex ratio. The absence of sex-specific differences is useful in guiding genetic association study design, for example when using mixed controls for sex-biased traits.

The insulin-degrading enzyme gene (IDE) is a strong functional and positional candidate for late onset Alzheimer's disease (LOAD).

Homozygosity has long been associated with rare, often devastating, Mendelian disorders, and Darwin was one of the first to recognize that inbreeding reduces evolutionary fitness. However, the effect of the more distant parental relatedness that is common in modern human populations is less well understood. Genomic data now allow us to investigate the effects of homozygosity on traits of public health importance by observing contiguous homozygous segments (runs of homozygosity), which are inferred to be homozygous along their complete length. Given the low levels of genome-wide homozygosity prevalent in most human populations, information is required on very large numbers of people to provide sufficient power. Here we use runs of homozygosity to study 16 health-related quantitative traits in 354,224 individuals from 102 cohorts, and find statistically significant associations between summed runs of homozygosity and four complex traits: height, forced expiratory lung volume in one second, general cognitive ability and educational attainment (P < 1 × 10(-300), 2.1 × 10(-6), 2.5 × 10(-10) and 1.8 × 10(-10), respectively). In each case, increased homozygosity was associated with decreased trait value, equivalent to the offspring of first cousins being 1.2 cm shorter and having 10 months' less education. Similar effect sizes were found across four continental groups and populations with different degrees of genome-wide homozygosity, providing evidence that homozygosity, rather than confounding, directly contributes to phenotypic variance. Contrary to earlier reports in substantially smaller samples, no evidence was seen of an influence of genome-wide homozygosity on blood pressure and low density lipoprotein cholesterol, or ten other cardio-metabolic traits. Since directional dominance is predicted for traits under directional evolutionary selection, this study provides evidence that increased stature and cognitive function have been positively selected in human evolution, whereas many important risk factors for late-onset complex diseases may not have been.

Reduced glomerular filtration rate defines chronic kidney disease and is associated with cardiovascular and all-cause mortality. We conducted a meta-analysis of genome-wide association studies for estimated glomerular filtration rate (eGFR), combining data across 133,413 individuals with replication in up to 42,166 individuals. We identify 24 new and confirm 29 previously identified loci. Of these 53 loci, 19 associate with eGFR among individuals with diabetes. Using bioinformatics, we show that identified genes at eGFR loci are enriched for expression in kidney tissues and in pathways relevant for kidney development and transmembrane transporter activity, kidney structure, and regulation of glucose metabolism. Chromatin state mapping and DNase I hypersensitivity analyses across adult tissues demonstrate preferential mapping of associated variants to regulatory regions in kidney but not extra-renal tissues. These findings suggest that genetic determinants of eGFR are mediated largely through direct effects within the kidney and highlight important cell types and biological pathways.

Elevated serum uric acid levels cause gout and are a risk factor for cardiovascular disease and diabetes. To investigate the polygenetic basis of serum uric acid levels, we conducted a meta-analysis of genome-wide association scans from 14 studies totalling 28,141 participants of European descent, resulting in identification of 954 SNPs distributed across nine loci that exceeded the threshold of genome-wide significance, five of which are novel. Overall, the common variants associated with serum uric acid levels fall in the following nine regions: SLC2A9 (p = 5.2x10(-201)), ABCG2 (p = 3.1x10(-26)), SLC17A1 (p = 3.0x10(-14)), SLC22A11 (p = 6.7x10(-14)), SLC22A12 (p = 2.0x10(-9)), SLC16A9 (p = 1.1x10(-8)), GCKR (p = 1.4x10(-9)), LRRC16A (p = 8.5x10(-9)), and near PDZK1 (p = 2.7x10(-9)). Identified variants were analyzed for gender differences. We found that the minor allele for rs734553 in SLC2A9 has greater influence in lowering uric acid levels in women and the minor allele of rs2231142 in ABCG2 elevates uric acid levels more strongly in men compared to women. To further characterize the identified variants, we analyzed their association with a panel of metabolites. rs12356193 within SLC16A9 was associated with DL-carnitine (p = 4.0x10(-26)) and propionyl-L-carnitine (p = 5.0x10(-8)) concentrations, which in turn were associated with serum UA levels (p = 1.4x10(-57) and p = 8.1x10(-54), respectively), forming a triangle between SNP, metabolites, and UA levels. Taken together, these associations highlight additional pathways that are important in the regulation of serum uric acid levels and point toward novel potential targets for pharmacological intervention to prevent or treat hyperuricemia. In addition, these findings strongly support the hypothesis that transport proteins are key in regulating serum uric acid levels.

Calcium is vital to the normal functioning of multiple organ systems and its serum concentration is tightly regulated. Apart from CASR, the genes associated with serum calcium are largely unknown. We conducted a genome-wide association meta-analysis of 39,400 individuals from 17 population-based cohorts and investigated the 14 most strongly associated loci in ≤ 21,679 additional individuals. Seven loci (six new regions) in association with serum calcium were identified and replicated. Rs1570669 near CYP24A1 (P = 9.1E-12), rs10491003 upstream of GATA3 (P = 4.8E-09) and rs7481584 in CARS (P = 1.2E-10) implicate regions involved in Mendelian calcemic disorders: Rs1550532 in DGKD (P = 8.2E-11), also associated with bone density, and rs7336933 near DGKH/KIAA0564 (P = 9.1E-10) are near genes that encode distinct isoforms of diacylglycerol kinase. Rs780094 is in GCKR. We characterized the expression of these genes in gut, kidney, and bone, and demonstrate modulation of gene expression in bone in response to dietary calcium in mice. Our results shed new light on the genetics of calcium homeostasis.

The length of female reproductive lifespan is associated with multiple adverse outcomes, including breast cancer, cardiovascular disease and infertility. The biological processes that govern the timing of the beginning and end of reproductive life are not well understood. Genetic variants are known to contribute to ∼50% of the variation in both age at menarche and menopause, but to date the known genes explain <15% of the genetic component. We have used genome-wide association in a bivariate meta-analysis of both traits to identify genes involved in determining reproductive lifespan. We observed significant genetic correlation between the two traits using genome-wide complex trait analysis. However, we found no robust statistical evidence for individual variants with an effect on both traits. A novel association with age at menopause was detected for a variant rs1800932 in the mismatch repair gene MSH6 (P = 1.9 × 10(-9)), which was also associated with altered expression levels of MSH6 mRNA in multiple tissues. This study contributes to the growing evidence that DNA repair processes play a key role in ovarian ageing and could be an important therapeutic target for infertility.

The analysis of less common variants in genome-wide association studies promises to elucidate complex trait genetics but is hampered by low power to reliably detect association. We show that addition of population-specific exome sequence data to global reference data allows more accurate imputation, particularly of less common SNPs (minor allele frequency 1-10%) in two very different European populations. The imputation improvement corresponds to an increase in effective sample size of 28-38%, for SNPs with a minor allele frequency in the range 1-3%.

Most common human traits and diseases have a polygenic pattern of inheritance: DNA sequence variants at many genetic loci influence the phenotype. Genome-wide association (GWA) studies have identified more than 600 variants associated with human traits, but these typically explain small fractions of phenotypic variation, raising questions about the use of further studies. Here, using 183,727 individuals, we show that hundreds of genetic variants, in at least 180 loci, influence adult height, a highly heritable and classic polygenic trait. The large number of loci reveals patterns with important implications for genetic studies of common human diseases and traits. First, the 180 loci are not random, but instead are enriched for genes that are connected in biological pathways (P = 0.016) and that underlie skeletal growth defects (P < 0.001). Second, the likely causal gene is often located near the most strongly associated variant: in 13 of 21 loci containing a known skeletal growth gene, that gene was closest to the associated variant. Third, at least 19 loci have multiple independently associated variants, suggesting that allelic heterogeneity is a frequent feature of polygenic traits, that comprehensive explorations of already-discovered loci should discover additional variants and that an appreciable fraction of associated loci may have been identified. Fourth, associated variants are enriched for likely functional effects on genes, being over-represented among variants that alter amino-acid structure of proteins and expression levels of nearby genes. Our data explain approximately 10% of the phenotypic variation in height, and we estimate that unidentified common variants of similar effect sizes would increase this figure to approximately 16% of phenotypic variation (approximately 20% of heritable variation). Although additional approaches are needed to dissect the genetic architecture of polygenic human traits fully, our findings indicate that GWA studies can identify large numbers of loci that implicate biologically relevant genes and pathways.

The QRS interval, from the beginning of the Q wave to the end of the S wave on an electrocardiogram, reflects ventricular depolarization and conduction time and is a risk factor for mortality, sudden death and heart failure. We performed a genome-wide association meta-analysis in 40,407 individuals of European descent from 14 studies, with further genotyping in 7,170 additional Europeans, and we identified 22 loci associated with QRS duration (P < 5 × 10(-8)). These loci map in or near genes in pathways with established roles in ventricular conduction such as sodium channels, transcription factors and calcium-handling proteins, but also point to previously unidentified biologic processes, such as kinase inhibitors and genes related to tumorigenesis. We demonstrate that SCN10A, a candidate gene at the most significantly associated locus in this study, is expressed in the mouse ventricular conduction system, and treatment with a selective SCN10A blocker prolongs QRS duration. These findings extend our current knowledge of ventricular depolarization and conduction.