86As yet undiscovered rare variants are hypothesized to substantially influence an 87 individual's risk for common diseases and traits, but sequencing studies aiming to 88 identify such variants have generally been underpowered. In isolated populations that 89 have expanded rapidly after a population bottleneck, deleterious alleles that passed 90 through the bottleneck may be maintained at much higher frequencies than in other 91 populations. In an exome sequencing study of nearly 20,000 cohort participants from 92 northern and eastern Finnish populations that exemplify this phenomenon, most novel 93 trait-associated deleterious variants are seen only in Finland or display frequencies more 94 than 20 times higher than in other European populations. These enriched alleles underlie 95 34 novel associations with 21 disease-related quantitative traits and demonstrate a 96 geographical clustering equivalent to that of Mendelian disease mutations characteristic 97 of the Finnish population. Sequencing studies in populations without this unique history 98 would require hundreds of thousands to millions of participants for comparable power for 99 these variants. 100 101 (defined here as MAF≤1%) which are not well-tagged by the single-nucleotide 109 polymorphisms (SNPs) on genome-wide genotyping arrays are probably responsible for 110 much of the heritability that remains unexplained 2 . Additionally, because purifying 111 selection acts to remove deleterious alleles from the population, most variants that exert a 112 sizable effect on complex traits, and that likely offer the best prospect for revealing 113 biological mechanisms, should be particularly rare. 114 115 Rare variants are unevenly distributed between populations and difficult to represent 116 effectively on commercial genotyping arrays, as evidenced by relatively sparse 117 association findings even from large array-based studies of coding variants 3-6 . 118Discovering rare variant associations will therefore almost certainly require exome or 119 genome sequencing of very large numbers of individuals. However, the sample size 120 required to reliably identify rare-variant associations remains uncertain; most sequencing 121 studies to date have identified few novel associations, and theoretical analyses confirm 122 that they have been underpowered to do so 7 . These analyses also suggest that power to 123 detect rare variant associations varies enormously between populations that have 124 expanded in isolation from recent bottlenecks compared to those that have not. 125 126In isolated populations that expand rapidly following a bottleneck, alleles that pass 127 through the bottleneck often rise to a much higher frequency than in other populations [8][9][10] . 128If the bottleneck was recent, even deleterious alleles under negative selection may remain 129 relatively frequent in these populations, resulting in increased power to detect association 130 with disease-related traits. The Finnish population exemplifies this type of history. It 131 5 grew from bottle...

The prevalence of microalbuminuria (MAU) in African populations has not been reported, nor has the relationship between MAU and hypertension been reported for these populations. We collected spot urine samples from 370 women, 25 years and older as a part of a population-based, cross-sectional blood pressure survey in an urban community in Zimbabwe and analysed the samples for albumin and ␤ 2 -microglobulin. The ageadjusted prevalence of hypertension was 30% for women 25 years and older in this community. After excluding the samples with hematuria (11%), the prevalence of MAU (3.0 р albumin-to-creatinine ratio (ACR, mg/mmol) Ͻ25.0) in the study population was 9%. When age-adjusted to the population in the community, the prevalence was 8% among women 25 years and older. The prevalence of MAU was substantially higher in

297

A test of association between the phenotype and a set of genes within a biological pathway can be complementary to single variant or single gene association analysis and provide further insights into the genetic architecture of complex phenotypes. Although multiple methods exist to perform such a gene-set analysis, most have low statistical power when only a small fraction of the genes are associated with the phenotype. Further, since existing methods cannot identify possible genes driving association signals, interpreting results of such association in terms of the underlying genetic mechanism is challenging. Here, we introduce Gene-set analysis Association Using Sparse Signals (GAUSS), a method for gene-set association analysis with GWAS summary statistics. In addition to providing a p-value for association, GAUSS identifies the subset of genes that have the maximal evidence of association and appears to drive the association. Using pre-computed correlation structure among test statistics from a reference panel, the p-value calculation is substantially faster compared to other permutation or simulationbased approaches. Our numerical experiments show that GAUSS can increase power over several existing methods while controlling type-I error under a variety of association models.Through the analysis of summary statistics from the UK Biobank data for 1,403 phenotypes, we show that GAUSS is scalable and can identify associations across many phenotypes and genesets.