Evaluation and application of summary statistic imputation to discover new height-associated loci.

As most of the heritability of complex traits is attributed to common and low frequency genetic variants, imputing them by combining genotyping chips and large sequenced reference panels is the most cost-effective approach to discover the genetic basis of these traits. Association summary statistics from genome-wide meta-analyses are available for hundreds of traits. Updating these to ever-increasing reference panels is very cumbersome as it requires reimputation of the genetic data, rerunning the association scan, and meta-analysing the results. A much more efficient method is to directly impute the summary statistics, termed as summary statistics imputation, which we improved to accommodate variable sample size across SNVs. Its performance relative to genotype imputation and practical utility has not yet been fully investigated. To this end, we compared the two approaches on real (genotyped and imputed) data from 120K samples from the UK Biobank and show that, genotype imputation boasts a 3- to 5-fold lower root-mean-square error, and better distinguishes true associations from null ones: We observed the largest differences in power for variants with low minor allele frequency and low imputation quality. For fixed false positive rates of 0.001, 0.01, 0.05, using summary statistics imputation yielded a decrease in statistical power by 9, 43 and 35%, respectively. To test its capacity to discover novel associations, we applied summary statistics imputation to the GIANT height meta-analysis summary statistics covering HapMap variants, and identified 34 novel loci, 19 of which replicated using data in the UK Biobank. Additionally, we successfully replicated 55 out of the 111 variants published in an exome chip study. Our study demonstrates that summary statistics imputation is a very efficient and cost-effective way to identify and fine-map trait-associated loci. Moreover, the ability to impute summary statistics is important for follow-up analyses, such as Mendelian randomisation or LD-score regression.

An Integrated Systems Genetics and Omics Toolkit to Probe Gene Function.

Identifying genetic and environmental factors that impact complex traits and common diseases is a high biomedical priority. Here, we developed, validated, and implemented a series of multi-layered systems approaches, including (expression-based) phenome-wide association, transcriptome-/proteome-wide association, and (reverse-) mediation analysis, in an open-access web server (systems-genetics.org) to expedite the systems dissection of gene function. We applied these approaches to multi-omics datasets from the BXD mouse genetic reference population, and identified and validated associations between genes and clinical and molecular phenotypes, including previously unreported links between Rpl26 and body weight, and Cpt1a and lipid metabolism. Furthermore, through mediation and reverse-mediation analysis we established regulatory relations between genes, such as the co-regulation of BCKDHA and BCKDHB protein levels, and identified targets of transcription factors E2F6, ZFP277, and ZKSCAN1. Our multifaceted toolkit enabled the identification of gene-gene and gene-phenotype links that are robust and that translate well across populations and species, and can be universally applied to any populations with multi-omics datasets.

Genome-wide meta-analysis associates HLA-DQA1/DRB1 and LPA and lifestyle factors with human longevity.

Genomic analysis of longevity offers the potential to illuminate the biology of human aging. Here, using genome-wide association meta-analysis of 606,059 parents' survival, we discover two regions associated with longevity (HLA-DQA1/DRB1 and LPA). We also validate previous suggestions that APOE, CHRNA3/5, CDKN2A/B, SH2B3 and FOXO3A influence longevity. Next we show that giving up smoking, educational attainment, openness to new experience and high-density lipoprotein (HDL) cholesterol levels are most positively genetically correlated with lifespan while susceptibility to coronary artery disease (CAD), cigarettes smoked per day, lung cancer, insulin resistance and body fat are most negatively correlated. We suggest that the effect of education on lifespan is principally mediated through smoking while the effect of obesity appears to act via CAD. Using instrumental variables, we suggest that an increase of one body mass index unit reduces lifespan by 7 months while 1 year of education adds 11 months to expected lifespan.Variability in human longevity is genetically influenced. Using genetic data of parental lifespan, the authors identify associations at HLA-DQA/DRB1 and LPA and find that genetic variants that increase educational attainment have a positive effect on lifespan whereas increasing BMI negatively affects lifespan.

CNV-association meta-analysis in 191,161 European adults reveals new loci associated with anthropometric traits.

There are few examples of robust associations between rare copy number variants (CNVs) and complex continuous human traits. Here we present a large-scale CNV association meta-analysis on anthropometric traits in up to 191,161 adult samples from 26 cohorts. The study reveals five CNV associations at 1q21.1, 3q29, 7q11.23, 11p14.2, and 18q21.32 and confirms two known loci at 16p11.2 and 22q11.21, implicating at least one anthropometric trait. The discovered CNVs are recurrent and rare (0.01-0.2%), with large effects on height (>2.4 cm), weight (>5 kg), and body mass index (BMI) (>3.5 kg/m2). Burden analysis shows a 0.41 cm decrease in height, a 0.003 increase in waist-to-hip ratio and increase in BMI by 0.14 kg/m2 for each Mb of total deletion burden (P = 2.5 × 10-10, 6.0 × 10-5, and 2.9 × 10-3). Our study provides evidence that the same genes (e.g., MC4R, FIBIN, and FMO5) harbor both common and rare variants affecting body size and that anthropometric traits share genetic loci with developmental and psychiatric disorders.Individual SNPs have small effects on anthropometric traits, yet the impact of CNVs has remained largely unknown. Here, Kutalik and co-workers perform a large-scale genome-wide meta-analysis of structural variation and find rare CNVs associated with height, weight and BMI with large effect sizes.

Bayesian association scan reveals loci associated with human lifespan and linked biomarkers.

The enormous variation in human lifespan is in part due to a myriad of sequence variants, only a few of which have been revealed to date. Since many life-shortening events are related to diseases, we developed a Mendelian randomization-based method combining 58 disease-related GWA studies to derive longevity priors for all HapMap SNPs. A Bayesian association scan, informed by these priors, for parental age of death in the UK Biobank study (n=116,279) revealed 16 independent SNPs with significant Bayes factor at a 5% false discovery rate (FDR). Eleven of them replicate (5% FDR) in five independent longevity studies combined; all but three are depleted of the life-shortening alleles in older Biobank participants. Further analysis revealed that brain expression levels of nearby genes (RBM6, SULT1A1 and CHRNA5) might be causally implicated in longevity. Gene expression and caloric restriction experiments in model organisms confirm the conserved role for RBM6 and SULT1A1 in modulating lifespan.

Risk prediction of developing venous thrombosis in combined oral contraceptive users.

BACKGROUND: Venous thromboembolism (VTE) is a complex multifactorial disease influenced by genetic and environmental risk factors. An example for the latter is the regular use of combined oral contraceptives (CC), which increases the risk to develop VTE by 3 to 7 fold, depending on estrogen dosage and the type of progestin present in the pill. One out of 1'000 women using CC develops thrombosis, often with life-long consequences; a risk assessment is therefore necessary prior to such treatment. Currently known clinical risk factors associated with VTE development in general are routinely checked by medical doctors, however they are far from being sufficient for risk prediction, even when combined with genetic tests for Factor V Leiden and Factor II G20210A variants. Thus, clinical and notably genetic risk factors specific to the development of thrombosis associated with the use of CC in particular should be identified. METHODS AND FINDINGS: Step-wise (logistic) model selection was applied to a population of 1622 women using CC, half of whom (794) had developed a thromboembolic event while using contraceptives. 46 polymorphisms and clinical parameters were tested in the model selection and a specific combination of 4 clinical risk factors and 9 polymorphisms were identified. Among the 9 polymorphisms, there are two novel genetic polymorphisms (rs1799853 and rs4379368) that had not been previously associated with the development of thromboembolic event. This new prediction model outperforms (AUC 0.71, 95% CI 0.69-0.74) previously published models for general thromboembolic events in a cross-validation setting. Further validation in independent populations should be envisaged. CONCLUSION: We identified two new genetic variants associated to VTE development, as well as a robust prediction model to assess the risk of thrombosis for women using combined oral contraceptives. This model outperforms current medical practice as well as previously published models and is the first model specific to CC use.

Quantifying the extent to which index event biases influence large genetic association studies.

As genetic association studies increase in size to 100 000s of individuals, subtle biases may influence conclusions. One possible bias is 'index event bias' (IEB) that appears due to the stratification by, or enrichment for, disease status when testing associations between genetic variants and a disease-associated trait. We aimed to test the extent to which IEB influences some known trait associations in a range of study designs and provide a statistical framework for assessing future associations. Analyzing data from 113 203 non-diabetic UK Biobank participants, we observed three (near TCF7L2, CDKN2AB and CDKAL1) overestimated (body mass index (BMI) decreasing) and one (near MTNR1B) underestimated (BMI increasing) associations among 11 type 2 diabetes risk alleles (at P < 0.05). IEB became even stronger when we tested a type 2 diabetes genetic risk score composed of these 11 variants (-0.010 standard deviations BMI per allele, P = 5 × 10- 4), which was confirmed in four additional independent studies. Similar results emerged when examining the effect of blood pressure increasing alleles on BMI in normotensive UK Biobank samples. Furthermore, we demonstrated that, under realistic scenarios, common disease alleles would become associated at P < 5 × 10- 8 with disease-related traits through IEB alone, if disease prevalence in the sample differs appreciably from the background population prevalence. For example, some hypertension and type 2 diabetes alleles will be associated with BMI in sample sizes of >500 000 if the prevalence of those diseases differs by >10% from the background population. In conclusion, IEB may result in false positive or negative genetic associations in very large studies stratified or strongly enriched for/against disease cases.

Association between parental history and genetic risk scores for coronary heart disease prediction: The population-based CoLaus study.

BACKGROUND AND AIMS: Parental history (PH) and genetic risk scores (GRSs) are separately associated with coronary heart disease (CHD), but evidence regarding their combined effects is lacking. We aimed to evaluate the joint associations and predictive ability of PH and GRSs for incident CHD. METHODS: Data for 4283 Caucasians were obtained from the population-based CoLaus Study, over median follow-up time of 5.6 years. CHD was defined as incident myocardial infarction, angina, percutaneous coronary revascularization or bypass grafting. Single nucleotide polymorphisms for CHD identified by genome-wide association studies were used to construct unweighted and weighted versions of three GRSs, comprising of 38, 53 and 153 SNPs respectively. RESULTS: PH was associated with higher values of all weighted GRSs. After adjustment for age, sex, smoking, diabetes, systolic blood pressure, low and high density lipoprotein cholesterol, PH was significantly associated with CHD [HR 2.61, 95% CI (1.47-4.66)] and further adjustment for GRSs did not change this estimate. Similarly, one standard deviation change of the weighted 153-SNPs GRS was significantly associated with CHD [HR 1.50, 95% CI (1.26-1.80)] and remained so, after further adjustment for PH. The weighted, 153-SNPs GRS, but not PH, modestly improved discrimination [(C-index improvement, 0.016), p = 0.048] and reclassification [(NRI improvement, 8.6%), p = 0.027] beyond cardiovascular risk factors. After including both the GRS and PH, model performance improved further [(C-index improvement, 0.022), p = 0.006]. CONCLUSION: After adjustment for cardiovascular risk factors, PH and a weighted, polygenic GRS were jointly associated with CHD and provided additive information for coronary events prediction.

Seeding for pervasively overlapping communities.

In some social and biological networks, the majority of nodes belong to multiple communities. It has recently been shown that a number of the algorithms specifically designed to detect overlapping communities do not perform well in such highly overlapping settings. Here, we consider one class of these algorithms, those which optimize a local fitness measure, typically by using a greedy heuristic to expand a seed into a community. We perform synthetic benchmarks which indicate that an appropriate seeding strategy becomes more important as the extent of community overlap increases. We find that distinct cliques provide the best seeds. We find further support for this seeding strategy with benchmarks on a Facebook network and the yeast interactome.