Sequencing chromosomal abnormalities reveals neurodevelopmental loci that confer risk across diagnostic boundaries.

Balanced chromosomal abnormalities (BCAs) represent a relatively untapped reservoir of single-gene disruptions in neurodevelopmental disorders (NDDs). We sequenced BCAs in patients with autism or related NDDs, revealing disruption of 33 loci in four general categories: (1) genes previously associated with abnormal neurodevelopment (e.g., AUTS2, FOXP1, and CDKL5), (2) single-gene contributors to microdeletion syndromes (MBD5, SATB2, EHMT1, and SNURF-SNRPN), (3) novel risk loci (e.g., CHD8, KIRREL3, and ZNF507), and (4) genes associated with later-onset psychiatric disorders (e.g., TCF4, ZNF804A, PDE10A, GRIN2B, and ANK3). We also discovered among neurodevelopmental cases a profoundly increased burden of copy-number variants from these 33 loci and a significant enrichment of polygenic risk alleles from genome-wide association studies of autism and schizophrenia. Our findings suggest a polygenic risk model of autism and reveal that some neurodevelopmental genes are sensitive to perturbation by multiple mutational mechanisms, leading to variable phenotypic outcomes that manifest at different life stages.

Mapping and characterization of structural variation in 17,795 human genomes.

A key goal of whole-genome sequencing for studies of human genetics is to interrogate all forms of variation, including single-nucleotide variants, small insertion or deletion (indel) variants and structural variants. However, tools and resources for the study of structural variants have lagged behind those for smaller variants. Here we used a scalable pipeline1 to map and characterize structural variants in 17,795 deeply sequenced human genomes. We publicly release site-frequency data to create the largest, to our knowledge, whole-genome-sequencing-based structural variant resource so far. On average, individuals carry 2.9 rare structural variants that alter coding regions; these variants affect the dosage or structure of 4.2 genes and account for 4.0-11.2% of rare high-impact coding alleles. Using a computational model, we estimate that structural variants account for 17.2% of rare alleles genome-wide, with predicted deleterious effects that are equivalent to loss-of-function coding alleles; approximately 90% of such structural variants are noncoding deletions (mean 19.1 per genome). We report 158,991 ultra-rare structural variants and show that 2% of individuals carry ultra-rare megabase-scale structural variants, nearly half of which are balanced or complex rearrangements. Finally, we infer the dosage sensitivity of genes and noncoding elements, and reveal trends that relate to element class and conservation. This work will help to guide the analysis and interpretation of structural variants in the era of whole-genome sequencing.

Exome sequencing of Finnish isolates enhances rare-variant association power.

Exome-sequencing studies have generally been underpowered to identify deleterious alleles with a large effect on complex traits as such alleles are mostly rare. Because the population of northern and eastern Finland has expanded considerably and in isolation following a series of bottlenecks, individuals of these populations have numerous deleterious alleles at a relatively high frequency. Here, using exome sequencing of nearly 20,000 individuals from these regions, we investigate the role of rare coding variants in clinically relevant quantitative cardiometabolic traits. Exome-wide association studies for 64 quantitative traits identified 26 newly associated deleterious alleles. Of these 26 alleles, 19 are either unique to or more than 20 times more frequent in Finnish individuals than in other Europeans and show geographical clustering comparable to Mendelian disease mutations that are characteristic of the Finnish population. We estimate that sequencing studies of populations without this unique history would require hundreds of thousands to millions of participants to achieve comparable association power.

Author Correction: Exome sequencing of Finnish isolates enhances rare-variant association power.

An Amendment to this paper has been published and can be accessed via a link at the top of the paper.

Association of structural variation with cardiometabolic traits in Finns.

The contribution of genome structural variation (SV) to quantitative traits associated with cardiometabolic diseases remains largely unknown. Here, we present the results of a study examining genetic association between SVs and cardiometabolic traits in the Finnish population. We used sensitive methods to identify and genotype 129,166 high-confidence SVs from deep whole-genome sequencing (WGS) data of 4,848 individuals. We tested the 64,572 common and low-frequency SVs for association with 116 quantitative traits and tested candidate associations using exome sequencing and array genotype data from an additional 15,205 individuals. We discovered 31 genome-wide significant associations at 15 loci, including 2 loci at which SVs have strong phenotypic effects: (1) a deletion of the ALB promoter that is greatly enriched in the Finnish population and causes decreased serum albumin level in carriers (p = 1.47 × 10-54) and is also associated with increased levels of total cholesterol (p = 1.22 × 10-28) and 14 additional cholesterol-related traits, and (2) a multi-allelic copy number variant (CNV) at PDPR that is strongly associated with pyruvate (p = 4.81 × 10-21) and alanine (p = 6.14 × 10-12) levels and resides within a structurally complex genomic region that has accumulated many rearrangements over evolutionary time. We also confirmed six previously reported associations, including five led by stronger signals in single nucleotide variants (SNVs) and one linking recurrent HP gene deletion and cholesterol levels (p = 6.24 × 10-10), which was also found to be strongly associated with increased glycoprotein level (p = 3.53 × 10-35). Our study confirms that integrating SVs in trait-mapping studies will expand our knowledge of genetic factors underlying disease risk.

Structural variants are a major source of gene expression differences in humans and often affect multiple nearby genes.

Structural variants (SVs) are an important source of human genome diversity, but their functional effects are poorly understood. We mapped 61,668 SVs in 613 individuals from the GTEx project and measured their effects on gene expression. We estimate that common SVs are causal at 2.66% of eQTLs, a 10.5-fold enrichment relative to their abundance in the genome. Duplications and deletions were the most impactful variant types, whereas the contribution of mobile element insertions was small (0.12% of eQTLs, 1.9-fold enriched). Multitissue analysis of eQTLs revealed that gene-altering SVs show more constitutive effects than other variant types, with 62.09% of coding SV-eQTLs active in all tissues with eQTL activity compared with 23.08% of coding SNV- and indel-eQTLs. Noncoding SVs, SNVs and indels show broadly similar patterns. We also identified 539 rare SVs associated with nearby gene expression outliers. Of these, 62.34% are noncoding SVs that affect gene expression but have modest enrichment at regulatory elements, showing that rare noncoding SVs are a major source of gene expression differences but remain difficult to predict from current annotations. Both common and rare SVs often affect the expression of multiple genes: SV-eQTLs affect an average of 1.82 nearby genes, whereas SNV- and indel-eQTLs affect an average of 1.09 genes, and 21.34% of rare expression-altering SVs show effects on two to nine different genes. We also observe significant effects on rare gene expression changes extending 1 Mb from the SV. This provides a mechanism by which individual SVs may have strong or pleiotropic effects on phenotypic variation.

The impact of rare variation on gene expression across tissues.

Rare genetic variants are abundant in humans and are expected to contribute to individual disease risk. While genetic association studies have successfully identified common genetic variants associated with susceptibility, these studies are not practical for identifying rare variants. Efforts to distinguish pathogenic variants from benign rare variants have leveraged the genetic code to identify deleterious protein-coding alleles, but no analogous code exists for non-coding variants. Therefore, ascertaining which rare variants have phenotypic effects remains a major challenge. Rare non-coding variants have been associated with extreme gene expression in studies using single tissues, but their effects across tissues are unknown. Here we identify gene expression outliers, or individuals showing extreme expression levels for a particular gene, across 44 human tissues by using combined analyses of whole genomes and multi-tissue RNA-sequencing data from the Genotype-Tissue Expression (GTEx) project v6p release. We find that 58% of underexpression and 28% of overexpression outliers have nearby conserved rare variants compared to 8% of non-outliers. Additionally, we developed RIVER (RNA-informed variant effect on regulation), a Bayesian statistical model that incorporates expression data to predict a regulatory effect for rare variants with higher accuracy than models using genomic annotations alone. Overall, we demonstrate that rare variants contribute to large gene expression changes across tissues and provide an integrative method for interpretation of rare variants in individual genomes.

Mutations in DCHS1 cause mitral valve prolapse.

Mitral valve prolapse (MVP) is a common cardiac valve disease that affects nearly 1 in 40 individuals. It can manifest as mitral regurgitation and is the leading indication for mitral valve surgery. Despite a clear heritable component, the genetic aetiology leading to non-syndromic MVP has remained elusive. Four affected individuals from a large multigenerational family segregating non-syndromic MVP underwent capture sequencing of the linked interval on chromosome 11. We report a missense mutation in the DCHS1 gene, the human homologue of the Drosophila cell polarity gene dachsous (ds), that segregates with MVP in the family. Morpholino knockdown of the zebrafish homologue dachsous1b resulted in a cardiac atrioventricular canal defect that could be rescued by wild-type human DCHS1, but not by DCHS1 messenger RNA with the familial mutation. Further genetic studies identified two additional families in which a second deleterious DCHS1 mutation segregates with MVP. Both DCHS1 mutations reduce protein stability as demonstrated in zebrafish, cultured cells and, notably, in mitral valve interstitial cells (MVICs) obtained during mitral valve repair surgery of a proband. Dchs1(+/-) mice had prolapse of thickened mitral leaflets, which could be traced back to developmental errors in valve morphogenesis. DCHS1 deficiency in MVP patient MVICs, as well as in Dchs1(+/-) mouse MVICs, result in altered migration and cellular patterning, supporting these processes as aetiological underpinnings for the disease. Understanding the role of DCHS1 in mitral valve development and MVP pathogenesis holds potential for therapeutic insights for this very common disease.

svtools: population-scale analysis of structural variation.

SUMMARY: Large-scale human genetics studies are now employing whole genome sequencing with the goal of conducting comprehensive trait mapping analyses of all forms of genome variation. However, methods for structural variation (SV) analysis have lagged far behind those for smaller scale variants, and there is an urgent need to develop more efficient tools that scale to the size of human populations. Here, we present a fast and highly scalable software toolkit (svtools) and cloud-based pipeline for assembling high quality SV maps-including deletions, duplications, mobile element insertions, inversions and other rearrangements-in many thousands of human genomes. We show that this pipeline achieves similar variant detection performance to established per-sample methods (e.g. LUMPY), while providing fast and affordable joint analysis at the scale of ≥100 000 genomes. These tools will help enable the next generation of human genetics studies. AVAILABILITY AND IMPLEMENTATION: svtools is implemented in Python and freely available (MIT) from https://github.com/hall-lab/svtools. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The genome of the vervet (Chlorocebus aethiops sabaeus).

We describe a genome reference of the African green monkey or vervet (Chlorocebus aethiops). This member of the Old World monkey (OWM) superfamily is uniquely valuable for genetic investigations of simian immunodeficiency virus (SIV), for which it is the most abundant natural host species, and of a wide range of health-related phenotypes assessed in Caribbean vervets (C. a. sabaeus), whose numbers have expanded dramatically since Europeans introduced small numbers of their ancestors from West Africa during the colonial era. We use the reference to characterize the genomic relationship between vervets and other primates, the intra-generic phylogeny of vervet subspecies, and genome-wide structural variations of a pedigreed C. a. sabaeus population. Through comparative analyses with human and rhesus macaque, we characterize at high resolution the unique chromosomal fission events that differentiate the vervets and their close relatives from most other catarrhine primates, in whom karyotype is highly conserved. We also provide a summary of transposable elements and contrast these with the rhesus macaque and human. Analysis of sequenced genomes representing each of the main vervet subspecies supports previously hypothesized relationships between these populations, which range across most of sub-Saharan Africa, while uncovering high levels of genetic diversity within each. Sequence-based analyses of major histocompatibility complex (MHC) polymorphisms reveal extremely low diversity in Caribbean C. a. sabaeus vervets, compared to vervets from putatively ancestral West African regions. In the C. a. sabaeus research population, we discover the first structural variations that are, in some cases, predicted to have a deleterious effect; future studies will determine the phenotypic impact of these variations.

SpeedSeq: ultra-fast personal genome analysis and interpretation.

SpeedSeq is an open-source genome analysis platform that accomplishes alignment, variant detection and functional annotation of a 50× human genome in 13 h on a low-cost server and alleviates a bioinformatics bottleneck that typically demands weeks of computation with extensive hands-on expert involvement. SpeedSeq offers performance competitive with or superior to current methods for detecting germline and somatic single-nucleotide variants, structural variants, insertions and deletions, and it includes novel functionality for streamlined interpretation.

Implication of LRRC4C and DPP6 in neurodevelopmental disorders.

We performed whole-genome sequencing on an individual from a family with variable psychiatric phenotypes that had a sensory processing disorder, apraxia, and autism. The proband harbored a maternally inherited balanced translocation (46,XY,t(11;14)(p12;p12)mat) that disrupted LRRC4C, a member of the highly specialized netrin G family of axon guidance molecules. The proband also inherited a paternally derived chromosomal inversion that disrupted DPP6, a potassium channel interacting protein. Copy Number (CN) analysis in 14,077 cases with neurodevelopmental disorders and 8,960 control subjects revealed that 60% of cases with exonic deletions in LRRC4C had a second clinically recognizable syndrome associated with variable clinical phenotypes, including 16p11.2, 1q44, and 2q33.1 CN syndromes, suggesting LRRC4C deletion variants may be modifiers of neurodevelopmental disorders. In vitro, functional assessments modeling patient deletions in LRRC4C suggest a negative regulatory role of these exons found in the untranslated region of LRRC4C, which has a single, terminal coding exon. These data suggest that the proband's autism may be due to the inheritance of disruptions in both DPP6 and LRRC4C, and may highlight the importance of the netrin G family and potassium channel interacting molecules in neurodevelopmental disorders. © 2016 Wiley Periodicals, Inc.

Molecular analysis of a deletion hotspot in the NRXN1 region reveals the involvement of short inverted repeats in deletion CNVs.

NRXN1 microdeletions occur at a relatively high frequency and confer increased risk for neurodevelopmental and neurobehavioral abnormalities. The mechanism that makes NRXN1 a deletion hotspot is unknown. Here, we identified deletions of the NRXN1 region in affected cohorts, confirming a strong association with the autism spectrum and other neurodevelopmental disorders. Interestingly, deletions in both affected and control individuals were clustered in the 5' portion of NRXN1 and its immediate upstream region. To explore the mechanism of deletion, we mapped and analyzed the breakpoints of 32 deletions. At the deletion breakpoints, frequent microhomology (68.8%, 2-19 bp) suggested predominant mechanisms of DNA replication error and/or microhomology-mediated end-joining. Long terminal repeat (LTR) elements, unique non-B-DNA structures, and MEME-defined sequence motifs were significantly enriched, but Alu and LINE sequences were not. Importantly, small-size inverted repeats (minus self chains, minus sequence motifs, and partial complementary sequences) were significantly overrepresented in the vicinity of NRXN1 region deletion breakpoints, suggesting that, although they are not interrupted by the deletion process, such inverted repeats can predispose a region to genomic instability by mediating single-strand DNA looping via the annealing of partially reverse complementary strands and the promoting of DNA replication fork stalling and DNA replication error. Our observations highlight the potential importance of inverted repeats of variable sizes in generating a rearrangement hotspot in which individual breakpoints are not recurrent. Mechanisms that involve short inverted repeats in initiating deletion may also apply to other deletion hotspots in the human genome.

Exonic deletions in AUTS2 cause a syndromic form of intellectual disability and suggest a critical role for the C terminus.

Genomic rearrangements involving AUTS2 (7q11.22) are associated with autism and intellectual disability (ID), although evidence for causality is limited. By combining the results of diagnostic testing of 49,684 individuals, we identified 24 microdeletions that affect at least one exon of AUTS2, as well as one translocation and one inversion each with a breakpoint within the AUTS2 locus. Comparison of 17 well-characterized individuals enabled identification of a variable syndromic phenotype including ID, autism, short stature, microcephaly, cerebral palsy, and facial dysmorphisms. The dysmorphic features were more pronounced in persons with 3'AUTS2 deletions. This part of the gene is shown to encode a C-terminal isoform (with an alternative transcription start site) expressed in the human brain. Consistent with our genetic data, suppression of auts2 in zebrafish embryos caused microcephaly that could be rescued by either the full-length or the C-terminal isoform of AUTS2. Our observations demonstrate a causal role of AUTS2 in neurocognitive disorders, establish a hitherto unappreciated syndromic phenotype at this locus, and show how transcriptional complexity can underpin human pathology. The zebrafish model provides a valuable tool for investigating the etiology of AUTS2 syndrome and facilitating gene-function analysis in the future.

Disruption of a large intergenic noncoding RNA in subjects with neurodevelopmental disabilities.

Large intergenic noncoding (linc) RNAs represent a newly described class of ribonucleic acid whose importance in human disease remains undefined. We identified a severely developmentally delayed 16-year-old female with karyotype 46,XX,t(2;11)(p25.1;p15.1)dn in the absence of clinically significant copy number variants (CNVs). DNA capture followed by next-generation sequencing of the translocation breakpoints revealed disruption of a single noncoding gene on chromosome 2, LINC00299, whose RNA product is expressed in all tissues measured, but most abundantly in brain. Among a series of additional, unrelated subjects referred for clinical diagnostic testing who showed CNV affecting this locus, we identified four with exon-crossing deletions in association with neurodevelopmental abnormalities. No disruption of the LINC00299 coding sequence was seen in almost 14,000 control subjects. Together, these subjects with disruption of LINC00299 implicate this particular noncoding RNA in brain development and raise the possibility that, as a class, abnormalities of lincRNAs may play a significant role in human developmental disorders.

Next-generation sequencing strategies enable routine detection of balanced chromosome rearrangements for clinical diagnostics and genetic research.

The contribution of balanced chromosomal rearrangements to complex disorders remains unclear because they are not detected routinely by genome-wide microarrays and clinical localization is imprecise. Failure to consider these events bypasses a potentially powerful complement to single nucleotide polymorphism and copy-number association approaches to complex disorders, where much of the heritability remains unexplained. To capitalize on this genetic resource, we have applied optimized sequencing and analysis strategies to test whether these potentially high-impact variants can be mapped at reasonable cost and throughput. By using a whole-genome multiplexing strategy, rearrangement breakpoints could be delineated at a fraction of the cost of standard sequencing. For rearrangements already mapped regionally by karyotyping and fluorescence in situ hybridization, a targeted approach enabled capture and sequencing of multiple breakpoints simultaneously. Importantly, this strategy permitted capture and unique alignment of up to 97% of repeat-masked sequences in the targeted regions. Genome-wide analyses estimate that only 3.7% of bases should be routinely omitted from genomic DNA capture experiments. Illustrating the power of these approaches, the rearrangement breakpoints were rapidly defined to base pair resolution and revealed unexpected sequence complexity, such as co-occurrence of inversion and translocation as an underlying feature of karyotypically balanced alterations. These findings have implications ranging from genome annotation to de novo assemblies and could enable sequencing screens for structural variations at a cost comparable to that of microarrays in standard clinical practice.

Assessment of 2q23.1 microdeletion syndrome implicates MBD5 as a single causal locus of intellectual disability, epilepsy, and autism spectrum disorder.

Persons with neurodevelopmental disorders or autism spectrum disorder (ASD) often harbor chromosomal microdeletions, yet the individual genetic contributors within these regions have not been systematically evaluated. We established a consortium of clinical diagnostic and research laboratories to accumulate a large cohort with genetic alterations of chromosomal region 2q23.1 and acquired 65 subjects with microdeletion or translocation. We sequenced translocation breakpoints; aligned microdeletions to determine the critical region; assessed effects on mRNA expression; and examined medical records, photos, and clinical evaluations. We identified a single gene, methyl-CpG-binding domain 5 (MBD5), as the only locus that defined the critical region. Partial or complete deletion of MBD5 was associated with haploinsufficiency of mRNA expression, intellectual disability, epilepsy, and autistic features. Fourteen alterations, including partial deletions of noncoding regions not typically captured or considered pathogenic by current diagnostic screening, disrupted MBD5 alone. Expression profiles and clinical characteristics were largely indistinguishable between MBD5-specific alteration and deletion of the entire 2q23.1 interval. No copy-number alterations of MBD5 were observed in 7878 controls, suggesting MBD5 alterations are highly penetrant. We surveyed MBD5 coding variations among 747 ASD subjects compared to 2043 non-ASD subjects analyzed by whole-exome sequencing and detected an association with a highly conserved methyl-CpG-binding domain missense variant, p.79Gly>Glu (c.236G>A) (p = 0.012). These results suggest that genetic alterations of MBD5 cause features of 2q23.1 microdeletion syndrome and that this epigenetic regulator significantly contributes to ASD risk, warranting further consideration in research and clinical diagnostic screening and highlighting the importance of chromatin remodeling in the etiology of these complex disorders.

Haploinsufficiency of KDM6A is associated with severe psychomotor retardation, global growth restriction, seizures and cleft palate.

We describe a female subject (DGAP100) with a 46,X,t(X;5)(p11.3;q35.3)inv(5)(q35.3q35.1)dn, severe psychomotor retardation with hypotonia, global postnatal growth restriction, microcephaly, globally reduced cerebral volume, seizures, facial dysmorphia and cleft palate. Fluorescence in situ hybridization and whole-genome sequencing demonstrated that the X chromosome breakpoint disrupts KDM6A in the second intron. No genes were directly disrupted on chromosome 5. KDM6A is a histone 3 lysine 27 demethylase and a histone 3 lysine 4 methyltransferase. Expression of KDM6A is significantly reduced in DGAP100 lymphoblastoid cells compared to control samples. We identified nine additional cases with neurodevelopmental delay and various other features consistent with the DGAP100 phenotype with copy number variation encompassing KDM6A from microarray databases. We evaluated haploinsufficiency of kdm6a in a zebrafish model. kdm6a is expressed in the pharyngeal arches and ethmoid plate of the developing zebrafish, while a kdm6a morpholino knockdown exhibited craniofacial defects. We conclude KDM6A dosage regulation is associated with severe and diverse structural defects and developmental abnormalities.

Pervasive correlations between causal disease effects of proximal SNPs vary with functional annotations and implicate stabilizing selection.

The genetic architecture of human diseases and complex traits has been extensively studied, but little is known about the relationship of causal disease effect sizes between proximal SNPs, which have largely been assumed to be independent. We introduce a new method, LD SNP-pair effect correlation regression (LDSPEC), to estimate the correlation of causal disease effect sizes of derived alleles between proximal SNPs, depending on their allele frequencies, LD, and functional annotations; LDSPEC produced robust estimates in simulations across various genetic architectures. We applied LDSPEC to 70 diseases and complex traits from the UK Biobank (average N=306K), meta-analyzing results across diseases/traits. We detected significantly nonzero effect correlations for proximal SNP pairs (e.g., -0.37±0.09 for low-frequency positive-LD 0-100bp SNP pairs) that decayed with distance (e.g., -0.07±0.01 for low-frequency positive-LD 1-10kb), varied with allele frequency (e.g., -0.15±0.04 for common positive-LD 0-100bp), and varied with LD between SNPs (e.g., +0.12±0.05 for common negative-LD 0-100bp) (because we consider derived alleles, positive-LD and negative-LD SNP pairs may yield very different results). We further determined that SNP pairs with shared functions had stronger effect correlations that spanned longer genomic distances, e.g., -0.37±0.08 for low-frequency positive-LD same-gene promoter SNP pairs (average genomic distance of 47kb (due to alternative splicing)) and -0.32±0.04 for low-frequency positive-LD H3K27ac 0-1kb SNP pairs. Consequently, SNP-heritability estimates were substantially smaller than estimates of the sum of causal effect size variances across all SNPs (ratio of 0.87±0.02 across diseases/traits), particularly for certain functional annotations (e.g., 0.78±0.01 for common Super enhancer SNPs)-even though these quantities are widely assumed to be equal. We recapitulated our findings via forward simulations with an evolutionary model involving stabilizing selection, implicating the action of linkage masking, whereby haplotypes containing linked SNPs with opposite effects on disease have reduced effects on fitness and escape negative selection.

Pervasive correlations between causal disease effects of proximal SNPs vary with functional annotations and implicate stabilizing selection.

The genetic architecture of human diseases and complex traits has been extensively studied, but little is known about the relationship of causal disease effect sizes between proximal SNPs, which have largely been assumed to be independent. We introduce a new method, LD SNP-pair effect correlation regression (LDSPEC), to estimate the correlation of causal disease effect sizes of derived alleles between proximal SNPs, depending on their allele frequencies, LD, and functional annotations; LDSPEC produced robust estimates in simulations across various genetic architectures. We applied LDSPEC to 70 diseases and complex traits from the UK Biobank (average N=306K), meta-analyzing results across diseases/traits. We detected significantly nonzero effect correlations for proximal SNP pairs (e.g., -0.37±0.09 for low-frequency positive-LD 0-100bp SNP pairs) that decayed with distance (e.g., -0.07±0.01 for low-frequency positive-LD 1-10kb), varied with allele frequency (e.g., -0.15±0.04 for common positive-LD 0-100bp), and varied with LD between SNPs (e.g., +0.12±0.05 for common negative-LD 0-100bp) (because we consider derived alleles, positive-LD and negative-LD SNP pairs may yield very different results). We further determined that SNP pairs with shared functions had stronger effect correlations that spanned longer genomic distances, e.g., -0.37±0.08 for low-frequency positive-LD same-gene promoter SNP pairs (average genomic distance of 47kb (due to alternative splicing)) and -0.32±0.04 for low-frequency positive-LD H3K27ac 0-1kb SNP pairs. Consequently, SNP-heritability estimates were substantially smaller than estimates of the sum of causal effect size variances across all SNPs (ratio of 0.87±0.02 across diseases/traits), particularly for certain functional annotations (e.g., 0.78±0.01 for common Super enhancer SNPs)-even though these quantities are widely assumed to be equal. We recapitulated our findings via forward simulations with an evolutionary model involving stabilizing selection, implicating the action of linkage masking, whereby haplotypes containing linked SNPs with opposite effects on disease have reduced effects on fitness and escape negative selection.