Key variants via the Alzheimer's Disease Sequencing Project whole genome sequence data.

INTRODUCTION: Genome-wide association studies (GWAS) have identified loci associated with Alzheimer's disease (AD) but did not identify specific causal genes or variants within those loci. Analysis of whole genome sequence (WGS) data, which interrogates the entire genome and captures rare variations, may identify causal variants within GWAS loci. METHODS: We performed single common variant association analysis and rare variant aggregate analyses in the pooled population (N cases = 2184, N controls = 2383) and targeted analyses in subpopulations using WGS data from the Alzheimer's Disease Sequencing Project (ADSP). The analyses were restricted to variants within 100 kb of 83 previously identified GWAS lead variants. RESULTS: Seventeen variants were significantly associated with AD within five genomic regions implicating the genes OARD1/NFYA/TREML1, JAZF1, FERMT2, and SLC24A4. KAT8 was implicated by both single variant and rare variant aggregate analyses. DISCUSSION: This study demonstrates the utility of leveraging WGS to gain insights into AD loci identified via GWAS.

Frequency of Variants in Mendelian Alzheimer's Disease Genes within the Alzheimer's Disease Sequencing Project (ADSP).

Background: Prior studies using the ADSP data examined variants within presenilin-2 ( PSEN2 ), presenilin-1 ( PSEN1 ), and amyloid precursor protein ( APP ) genes. However, previously-reported clinically-relevant variants and other predicted damaging missense (DM) variants have not been characterized in a newer release of the Alzheimer's Disease Sequencing Project (ADSP). Objective: To characterize previously-reported clinically-relevant variants and DM variants in PSEN2, PSEN1, APP within the participants from the ADSP. Methods: We identified rare variants (MAF <1%) previously-reported in PSEN2 , PSEN1, and APP in the available ADSP sample of 14,641 individuals with whole genome sequencing and 16,849 individuals with whole exome sequencing available for research-use (N total = 31,490). We additionally curated variants in these three genes from ClinVar, OMIM, and Alzforum and report carriers of variants in clinical databases as well as predicted DM variants in these genes. Results: We detected 31 previously-reported clinically-relevant variants with alternate alleles observed within the ADSP: 4 variants in PSEN2 , 25 in PSEN1 , and 2 in APP . The overall variant carrier rate for the 31 clinically-relevant variants in the ADSP was 0.3%. We observed that 79.5% of the variant carriers were cases compared to 3.9% were controls. In those with AD, the mean age of onset of AD among carriers of these clinically-relevant variants was 19.6 ± 1.4 years earlier compared with non-carriers (p-value=7.8×10 -57 ). Conclusion: A small proportion of individuals in the ADSP are carriers of a previously-reported clinically-relevant variant allele for AD and these participants have significantly earlier age of AD onset compared to non-carriers.

Bone mineral density and the risk of incident dementia: A meta-analysis.

BACKGROUND: It is not known whether bone mineral density (BMD) measured at baseline or as the rate of decline prior to baseline (prior bone loss) is a stronger predictor of incident dementia or Alzheimer's disease (AD). METHODS: We performed a meta-analysis of three longitudinal studies, the Framingham Heart Study (FHS), the Rotterdam Study (RS), and the Rush Memory and Aging Project (MAP), modeling the time to diagnosis of dementia as a function of BMD measures accounting for covariates. We included individuals with one or two BMD assessments, aged ≥60 years, and free of dementia at baseline with follow-up available. BMD was measured at the hip femoral neck using dual-energy X-ray absorptiometry (DXA), or at the heel calcaneus using quantitative ultrasound to calculate estimated BMD (eBMD). BMD at study baseline ("baseline BMD") and annualized percentage change in BMD prior to baseline ("prior bone loss") were included as continuous measures. The primary outcome was incident dementia diagnosis within 10 years of baseline, and incident AD was a secondary outcome. Baseline covariates included age, sex, body mass index, ApoE4 genotype, and education. RESULTS: The combined sample size across all three studies was 4431 with 606 incident dementia diagnoses, 498 of which were AD. A meta-analysis of baseline BMD across three studies showed higher BMD to have a significant protective association with incident dementia with a hazard ratio of 0.47 (95% CI: 0.23-0.96; p = 0.038) per increase in g/cm2 , or 0.91 (95% CI: 0.84-0.995) per standard deviation increase. We observed a significant association between prior bone loss and incident dementia with a hazard ratio of 1.30 (95% CI: 1.12-1.51; p < 0.001) per percent increase in prior bone loss only in the FHS cohort. CONCLUSIONS: Baseline BMD but not prior bone loss was associated with incident dementia in a meta-analysis across three studies.

Key variants via Alzheimer's Disease Sequencing Project whole genome sequence data.

INTRODUCTION: Genome-wide association studies (GWAS) have identified loci associated with Alzheimer's disease (AD) but did not identify specific causal genes or variants within those loci. Analysis of whole genome sequence (WGS) data, which interrogates the entire genome and captures rare variations, may identify causal variants within GWAS loci. METHODS: We performed single common variant association analysis and rare variant aggregate analyses in the pooled population (N cases=2,184, N controls=2,383) and targeted analyses in sub-populations using WGS data from the Alzheimer's Disease Sequencing Project (ADSP). The analyses were restricted to variants within 100 kb of 83 previously identified GWAS lead variants. RESULTS: Seventeen variants were significantly associated with AD within five genomic regions implicating the genes OARD1/NFYA/TREML1, JAZF1, FERMT2, and SLC24A4. KAT8 was implicated by both single variant and rare variant aggregate analyses. DISCUSSION: This study demonstrates the utility of leveraging WGS to gain insights into AD loci identified via GWAS.

Association of Common and Rare Variants with Alzheimer's Disease in over 13,000 Diverse Individuals with Whole-Genome Sequencing from the Alzheimer's Disease Sequencing Project.

Alzheimer's Disease (AD) is a common disorder of the elderly that is both highly heritable and genetically heterogeneous. Here, we investigated the association between AD and both common variants and aggregates of rare coding and noncoding variants in 13,371 individuals of diverse ancestry with whole genome sequence (WGS) data. Pooled-population analyses identified genetic variants in or near APOE, BIN1, and LINC00320 significantly associated with AD (p < 5×10-8). Population-specific analyses identified a haplotype on chromosome 14 including PSEN1 associated with AD in Hispanics, further supported by aggregate testing of rare coding and noncoding variants in this region. Finally, we observed suggestive associations (p < 5×10-5) of aggregates of rare coding rare variants in ABCA7 among non-Hispanic Whites (p=5.4×10-6), and rare noncoding variants in the promoter of TOMM40 distinct of APOE in pooled-population analyses (p=7.2×10-8). Complementary pooled-population and population-specific analyses offered unique insights into the genetic architecture of AD.

Exome sequencing identifies rare damaging variants in ATP8B4 and ABCA1 as risk factors for Alzheimer's disease.

Alzheimer's disease (AD), the leading cause of dementia, has an estimated heritability of approximately 70%1. The genetic component of AD has been mainly assessed using genome-wide association studies, which do not capture the risk contributed by rare variants2. Here, we compared the gene-based burden of rare damaging variants in exome sequencing data from 32,558 individuals-16,036 AD cases and 16,522 controls. Next to variants in TREM2, SORL1 and ABCA7, we observed a significant association of rare, predicted damaging variants in ATP8B4 and ABCA1 with AD risk, and a suggestive signal in ADAM10. Additionally, the rare-variant burden in RIN3, CLU, ZCWPW1 and ACE highlighted these genes as potential drivers of respective AD-genome-wide association study loci. Variants associated with the strongest effect on AD risk, in particular loss-of-function variants, are enriched in early-onset AD cases. Our results provide additional evidence for a major role for amyloid-ß precursor protein processing, amyloid-ß aggregation, lipid metabolism and microglial function in AD.

Meta-analysis of genome-wide association studies identifies ancestry-specific associations underlying circulating total tau levels.

Circulating total-tau levels can be used as an endophenotype to identify genetic risk factors for tauopathies and related neurological disorders. Here, we confirmed and better characterized the association of the 17q21 MAPT locus with circulating total-tau in 14,721 European participants and identified three novel loci in 953 African American participants (4q31, 5p13, and 6q25) at P < 5 × 10-8. We additionally detected 14 novel loci at P < 5 × 10-7, specific to either Europeans or African Americans. Using whole-exome sequence data in 2,279 European participants, we identified ten genes associated with circulating total-tau when aggregating rare variants. Our genetic study sheds light on genes reported to be associated with neurological diseases including stroke, Alzheimer's, and Parkinson's (F5, MAP1B, and BCAS3), with Alzheimer's pathological hallmarks (ADAMTS12, IL15, and FHIT), or with an important function in the brain (PARD3, ELFN2, UBASH3B, SLIT3, and NSD3), and suggests that the genetic architecture of circulating total-tau may differ according to ancestry.

Family history aggregation unit-based tests to detect rare genetic variant associations with application to the Framingham Heart Study.

A challenge in standard genetic studies is maintaining good power to detect associations, especially for low prevalent diseases and rare variants. The traditional methods are most powerful when evaluating the association between variants in balanced study designs. Without accounting for family correlation and unbalanced case-control ratio, these analyses could result in inflated type I error. One cost-effective solution to increase statistical power is exploitation of available family history (FH) that contains valuable information about disease heritability. Here, we develop methods to address the aforementioned type I error issues while providing optimal power to analyze aggregates of rare variants by incorporating additional information from FH. With enhanced power in these methods exploiting FH and accounting for relatedness and unbalanced designs, we successfully detect genes with suggestive associations with Alzheimer disease, dementia, and type 2 diabetes by using the exome chip data from the Framingham Heart Study.

Exploiting family history in aggregation unit-based genetic association tests.

The development of sequencing technology calls for new powerful methods to detect disease associations and lower the cost of sequencing studies. Family history (FH) contains information on disease status of relatives, adding valuable information about the probands' health problems and risk of diseases. Incorporating data from FH is a cost-effective way to improve statistical evidence in genetic studies, and moreover, overcomes limitations in study designs with insufficient cases or missing genotype information for association analysis. We proposed family history aggregation unit-based test (FHAT) and optimal FHAT (FHAT-O) to exploit available FH for rare variant association analysis. Moreover, we extended liability threshold model of case-control status and FH (LT-FH) method in aggregated unit-based methods and compared that with FHAT and FHAT-O. The computational efficiency and flexibility of the FHAT and FHAT-O were demonstrated through both simulations and applications. We showed that FHAT, FHAT-O, and LT-FH methods offer reasonable control of the type I error unless case/control ratio is unbalanced, in which case they result in smaller inflation than that observed with conventional methods excluding FH. We also demonstrated that FHAT and FHAT-O are more powerful than LT-FH and conventional methods in many scenarios. By applying FHAT and FHAT-O to the analysis of all cause dementia and hypertension using the exome sequencing data from the UK Biobank, we showed that our methods can improve significance for known regions. Furthermore, we replicated the previous associations in all cause dementia and hypertension and detected novel regions through the exome-wide analysis.

Large-scale sequencing studies expand the known genetic architecture of Alzheimer's disease.

INTRODUCTION: Genes implicated by genome-wide association studies and family-based studies of Alzheimer's disease (AD) are largely discordant. We hypothesized that genes identified by sequencing studies like the Alzheimer's Disease Sequencing Project (ADSP) may bridge this gap and highlight shared biological mechanisms. METHODS: We performed structured literature review of genes prioritized by ADSP studies, genes underlying familial dementias, and genes nominated by genome-wide association studies. Gene set enrichment analyses of each list identified enriched pathways. RESULTS: The genes prioritized by the ADSP, familial dementia studies, and genome-wide association studies minimally overlapped. Each gene set identified dozens of enriched pathways, several of which were shared (e.g., regulation of amyloid beta clearance). DISCUSSION: Alternative study designs provide unique insights into AD genetics. Shared pathways enriched by different genes highlight their relevance to AD pathogenesis, while the patterns of pathway enrichment unique to each gene set provide additional targets for functional studies.

Evaluation of population stratification adjustment using genome-wide or exonic variants.

Population stratification may cause an inflated type-I error and spurious association when assessing the association between genetic variations with an outcome. Many genetic association studies are now using exonic variants, which captures only 1% of the genome, however, population stratification adjustments have not been evaluated in the context of exonic variants. We compare the performance of two established approaches: principal components analysis (PCA) and mixed-effects models and assess the utility of genome-wide (GW) and exonic variants, by simulation and using a data set from the Framingham Heart Study. Our results illustrate that although the PCs and genetic relationship matrices computed by GW and exonic markers are different, the type-I error rate of association tests for common variants with additive effect appear to be properly controlled in the presence of population stratification. In addition, by considering single nucleotide variants (SNVs) that have different levels of confounding by population stratification, we also compare the power across multiple association approaches to account for population stratification such as PC-based corrections and mixed-effects models. We find that while these two methods achieve a similar power for SNVs that have a low or medium level of confounding by population stratification, mixed-effects model can reach a higher power for SNVs highly confounded by population stratification.

Cardiovascular health, genetic risk, and risk of dementia in the Framingham Heart Study.

OBJECTIVE: To determine the joint role of ideal cardiovascular health (CVH) and genetic risk on risk of dementia. METHODS: We categorized CVH on the basis of the American Heart Association Ideal CVH Index and genetic risk through a genetic risk score (GRS) of common genetic variants and the APOE ε4 genotype in 1,211 Framingham Heart Study (FHS) offspring cohort participants. We used multivariable Cox proportional hazards regression models to examine the association between CVH, genetic risk, and incident all-cause dementia with up to 10 years of follow-up (mean 8.4 years, 96 incident dementia cases), adjusting for age, sex, and education. RESULTS: We observed that a high GRS (>80th percentile) was associated with a 2.6-fold risk of dementia (95% confidence interval [CI] of hazard ratio [HR] 1.23-5.29; p = 0.012) compared with having a low GRS (<20th percentile); carrying at least 1 APOE ε4 allele was associated with a 2.3-fold risk of dementia compared with not carrying an APOE ε4 allele (95% CI of HR 1.49-3.53; p = 0.0002), and having a favorable CVH showed a 0.45-fold lower risk of dementia (95% CI of HR 0.20-1.01; p = 0.0527) compared to having an unfavorable CVH when all 3 components were included in the model. We did not observe an interaction between CVH and GRS (p = 0.99) or APOE ε4 (p = 0.16). CONCLUSIONS: We observed that both genetic risk and CVH contribute additively to dementia risk.

Analysis of Whole-Exome Sequencing Data for Alzheimer Disease Stratified by APOE Genotype.

IMPORTANCE: Previous genome-wide association studies of common variants identified associations for Alzheimer disease (AD) loci evident only among individuals with particular APOE alleles. OBJECTIVE: To identify APOE genotype-dependent associations with infrequent and rare variants using whole-exome sequencing. DESIGN, SETTING, AND PARTICIPANTS: The discovery stage included 10 441 non-Hispanic white participants in the Alzheimer Disease Sequencing Project. Replication was sought in 2 independent, whole-exome sequencing data sets (1766 patients with AD, 2906 without AD [controls]) and a chip-based genotype imputation data set (8728 patients with AD, 9808 controls). Bioinformatics and functional analyses were conducted using clinical, cognitive, neuropathologic, whole-exome sequencing, and gene expression data obtained from a longitudinal cohort sample including 402 patients with AD and 647 controls. Data were analyzed between March 2017 and September 2018. MAIN OUTCOMES AND MEASURES: Score, Firth, and sequence kernel association tests were used to test the association of AD risk with individual variants and genes in subgroups of APOE ε4 carriers and noncarriers. Results with P ≤ 1 × 10-5 were further evaluated in the replication data sets and combined by meta-analysis. RESULTS: Among 3145 patients with AD and 4213 controls lacking ε4 (mean [SD] age, 83.4 [7.6] years; 4363 [59.3.%] women), novel genome-wide significant associations were obtained in the discovery sample with rs536940594 in AC099552 (odds ratio [OR], 88.0; 95% CI, 9.08-852.0; P = 2.22 × 10-7) and rs138412600 in GPAA1 (OR, 1.78; 95% CI, 1.44-2.2; meta-P = 7.81 × 10-8). GPAA1 was also associated with expression in the brain of GPAA1 (ß = -0.08; P = .03) and its repressive transcription factor, FOXG1 (ß = 0.13; P = .003), and global cognition function (ß = -0.53; P = .009). Significant gene-wide associations (threshold P ≤ 6.35 × 10-7) were observed for OR8G5 (P = 4.67 × 10-7), IGHV3-7 (P = 9.75 × 10-16), and SLC24A3 (P = 2.67 × 10-12) in 2377 patients with AD and 706 controls with ε4 (mean [SD] age, 75.2 [9.6] years; 1668 [54.1%] women). CONCLUSIONS AND RELEVANCE: The study identified multiple possible novel associations for AD with individual and aggregated rare variants in groups of individuals with and without APOE ε4 alleles that reinforce known and suggest additional pathways leading to AD.

Publisher Correction: Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes.

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

Quality control and integration of genotypes from two calling pipelines for whole genome sequence data in the Alzheimer's disease sequencing project.

The Alzheimer's Disease Sequencing Project (ADSP) performed whole genome sequencing (WGS) of 584 subjects from 111 multiplex families at three sequencing centers. Genotype calling of single nucleotide variants (SNVs) and insertion-deletion variants (indels) was performed centrally using GATK-HaplotypeCaller and Atlas V2. The ADSP Quality Control (QC) Working Group applied QC protocols to project-level variant call format files (VCFs) from each pipeline, and developed and implemented a novel protocol, termed "consensus calling," to combine genotype calls from both pipelines into a single high-quality set. QC was applied to autosomal bi-allelic SNVs and indels, and included pipeline-recommended QC filters, variant-level QC, and sample-level QC. Low-quality variants or genotypes were excluded, and sample outliers were noted. Quality was assessed by examining Mendelian inconsistencies (MIs) among 67 parent-offspring pairs, and MIs were used to establish additional genotype-specific filters for GATK calls. After QC, 578 subjects remained. Pipeline-specific QC excluded ~12.0% of GATK and 14.5% of Atlas SNVs. Between pipelines, ~91% of SNV genotypes across all QCed variants were concordant; 4.23% and 4.56% of genotypes were exclusive to Atlas or GATK, respectively; the remaining ~0.01% of discordant genotypes were excluded. For indels, variant-level QC excluded ~36.8% of GATK and 35.3% of Atlas indels. Between pipelines, ~55.6% of indel genotypes were concordant; while 10.3% and 28.3% were exclusive to Atlas or GATK, respectively; and ~0.29% of discordant genotypes were. The final WGS consensus dataset contains 27,896,774 SNVs and 3,133,926 indels and is publicly available.

Genetic Interaction with Plasma Lipids on Alzheimer's Disease in the Framingham Heart Study.

Epidemiological and genetic studies have pointed to the role of cholesterol in Alzheimer's disease (AD). We explored the interaction of a genetic risk score (GRS) of AD risk alleles with mid-life plasma lipid levels (LDL-C, HDL-C, and triglycerides) on risk for AD in the Framingham Heart Study (FHS). Mid-life (between the ages of 40-60 years old) lipid levels were obtained from individuals in the FHS Original and Offspring cohorts (157 cases and 2,882 controls) with genetic data and AD status available. Cox proportional hazards regression was performed to test the interaction between mid-life lipid levels and an AD GRS, as well as the individual contributing SNPs, on risk of incident AD adjusting for age, sex, and cohort. We found a significant interaction between a GRS of AD loci and log triglyceride levels on risk of clinical AD (p = 0.006), but no interaction of the GRS with HDL-C (p = 0.458) or LDL-C (p = 0.366). We then tested the interaction between the individual SNPs contributing to the GRS and log triglycerides. We found two SNPs that had interactions with triglycerides on AD risk that reached a p-value < 0.05 (rs11218343 and APOEɛ4). The association between some AD SNPs and risk of AD may be modified by triglyceride levels. Furthermore, sequential testing of a GRS with a set of traits on disease followed by testing individual SNPs for interaction provides a framework for narrowing the associations that need to be tested for interaction analyses. Replication is needed to confirm these findings.

Genetically elevated high-density lipoprotein cholesterol through the cholesteryl ester transfer protein gene does not associate with risk of Alzheimer's disease.

INTRODUCTION: There is conflicting evidence whether high-density lipoprotein cholesterol (HDL-C) is a risk factor for Alzheimer's disease (AD) and dementia. Genetic variation in the cholesteryl ester transfer protein (CETP) locus is associated with altered HDL-C. We aimed to assess AD risk by genetically predicted HDL-C. METHODS: Ten single nucleotide polymorphisms within the CETP locus predicting HDL-C were applied to the International Genomics of Alzheimer's Project (IGAP) exome chip stage 1 results in up 16,097 late onset AD cases and 18,077 cognitively normal elderly controls. We performed instrumental variables analysis using inverse variance weighting, weighted median, and MR-Egger. RESULTS: Based on 10 single nucleotide polymorphisms distinctly predicting HDL-C in the CETP locus, we found that HDL-C was not associated with risk of AD (P > .7). DISCUSSION: Our study does not support the role of HDL-C on risk of AD through HDL-C altered by CETP. This study does not rule out other mechanisms by which HDL-C affects risk of AD.

Network analysis of drug effect on triglyceride-associated DNA methylation.

BACKGROUND: DNA methylation, an epigenetic modification, can be affected by environmental factors and thus regulate gene expression levels that can lead to alterations of certain phenotypes. Network analysis has been used successfully to discover gene sets that are expressed differently across multiple disease states and suggest possible pathways of disease progression. We applied this framework to compare DNA methylation levels before and after lipid-lowering medication and to identify modules that differ topologically between the two time points, revealing the association between lipid medication and these triglyceride-related methylation sites. METHODS: We performed quality control using beta-mixture quantile normalization on 463,995 cytosine-phosphate-guanine (CpG) sites and deleted problematic sites, resulting in 423,004 probes. We identified 14,850 probes that were nominally associated with triglycerides prior to treatment and performed weighted gene correlation network analysis (WGCNA) to construct pre- and posttreatment methylation networks of these probes. We then applied both WGCNA module preservation and generalized Hamming distance (GHD) to identify modules with topological differences between the pre- and posttreatment. For modules with structural changes between 2 time points, we performed pathway-enrichment analysis to gain further insight into the biological function of the genes from these modules. RESULTS: Six triglyceride-associated modules were identified using pretreatment methylation probes. The same 3 modules were not preserved in posttreatment data using both the module-preservation and the GHD methods. Top-enriched pathways for the 3 differentially methylated modules are sphingolipid signaling pathway, proteoglycans in cancer, and metabolic pathways (p values < 0.005). One module in particular included an enrichment of lipid-related pathways among the top results. CONCLUSIONS: The same 3 modules, which were differentially methylated between pre- and posttreatment, were identified using both WGCNA module-preservation and GHD methods. Pathway analysis revealed that triglyceride-associated modules contain groups of genes that are involved in lipid signaling and metabolism. These 3 modules may provide insight into the effect of fenofibrate on changes in triglyceride levels and these methylation sites.

Integrative methylation score to identify epigenetic modifications associated with lipid changes resulting from fenofibrate treatment in families.

Epigenome-wide association studies (EWAS) have traditionally focused on the association test of single epigenetic markers with complex traits. However, it is possible that multiple cytosine-phosphate-guanine (CpG) sites at the same locus could jointly exert their effects on human traits. Therefore, a region-based test that combines multiple markers could be more powerful. We used 2 different region-based tests to investigate the association between changes in DNA methylation and drug response, including the median methylation level test (MMLT) and sequence kernel association test (SKAT). No genes were found to be significantly associated with the drug response (for triglycerides, the false discovery rate ranged from 0.855 to 0.999; for high-density lipoprotein cholesterol, and the false discovery rate ranged from 0.584 to 0.915). Further evidence is needed to explore potential application of gene-level methylation association analysis.

Exome Chip Analysis Identifies Low-Frequency and Rare Variants in MRPL38 for White Matter Hyperintensities on Brain Magnetic Resonance Imaging.

Background and Purpose- White matter hyperintensities (WMH) on brain magnetic resonance imaging are typical signs of cerebral small vessel disease and may indicate various preclinical, age-related neurological disorders, such as stroke. Though WMH are highly heritable, known common variants explain a small proportion of the WMH variance. The contribution of low-frequency/rare coding variants to WMH burden has not been explored. Methods- In the discovery sample we recruited 20 719 stroke/dementia-free adults from 13 population-based cohort studies within the Cohorts for Heart and Aging Research in Genomic Epidemiology consortium, among which 17 790 were of European ancestry and 2929 of African ancestry. We genotyped these participants at ≈250 000 mostly exonic variants with Illumina HumanExome BeadChip arrays. We performed ethnicity-specific linear regression on rank-normalized WMH in each study separately, which were then combined in meta-analyses to test for association with single variants and genes aggregating the effects of putatively functional low-frequency/rare variants. We then sought replication of the top findings in 1192 adults (European ancestry) with whole exome/genome sequencing data from 2 independent studies. Results- At 17q25, we confirmed the association of multiple common variants in TRIM65, FBF1, and ACOX1 ( P<6×10-7). We also identified a novel association with 2 low-frequency nonsynonymous variants in MRPL38 (lead, rs34136221; PEA=4.5×10-8) partially independent of known common signal ( PEA(conditional)=1.4×10-3). We further identified a locus at 2q33 containing common variants in NBEAL1, CARF, and WDR12 (lead, rs2351524; Pall=1.9×10-10). Although our novel findings were not replicated because of limited power and possible differences in study design, meta-analysis of the discovery and replication samples yielded stronger association for the 2 low-frequency MRPL38 variants ( Prs34136221=2.8×10-8). Conclusions- Both common and low-frequency/rare functional variants influence WMH. Larger replication and experimental follow-up are essential to confirm our findings and uncover the biological causal mechanisms of age-related WMH.