Articles https://doi.org/10.1038/s41588-022-01024-z New insights into the genetic etiology of Alzheimer’s disease and related dementias Characterization of the genetic landscape of Alzheimer’s disease (AD) and related dementias (ADD) provides a unique opportunity for a better understanding of the associated pathophysiological processes. We performed a two-stage genome-wide association study totaling 111,326 clinically diagnosed/‘proxy’ AD cases and 677,663 controls. We found 75 risk loci, of which 42 were new at the time of analysis. Pathway enrichment analyses confirmed the involvement of amyloid/tau pathways and highlighted microglia implication. Gene prioritization in the new loci identified 31 genes that were suggestive of new genetically associated processes, including the tumor necrosis factor alpha pathway through the linear ubiquitin chain assembly complex. We also built a new genetic risk score associated with the risk of future AD/dementia or progression from mild cognitive impairment to AD/dementia. The improvement in prediction led to a 1.6- to 1.9-fold increase in AD risk from the lowest to the highest decile, in addition to effects of age and the APOE ε4 allele. A D is the most common form of dementia. The heritability P ≤ 5 × 10−8 in the stage I and stage II meta-analysis. Furthermore, is high, estimated to be between 60% and 80%1. This strong we applied a PLINK clumping procedure4 to define potential inde- genetic component provides an opportunity to determine pendent hits within the stage I results (Methods). After validation the pathophysiological processes in AD and to identify new bio- by conditional analyses (Supplementary Note and Supplementary logical features, new prognostic/diagnostic markers and new thera- Tables 3 and 4), this approach enabled us to define 39 signals in peutic targets through translational genomics. Characterizing the 33 loci already known to be associated with the risk of developing genetic risk factors in AD is therefore a major objective; with the ADD3,5–10 and identify 42 loci defined as new at the time of analysis advent of high-throughput genomic techniques, a large number of (Tables 1 and 2, Supplementary Table 5 and Supplementary Figs. putative AD-associated loci/genes have been reported2. However, 2–29). Of the 42 new loci, 17 had P ≤ 5 × 10−8 in stage I and 25 were much of the underlying heritability remains unexplained. Hence, associated with P ≤ 5 × 10−8 after follow-up (stage I and stage II increasing the sample size of genome-wide association studies meta-analysis, including the ADGC, CHARGE and FinnGen data). (GWASs) is an obvious solution that has already been used to char- We also identified 6 loci with P ≤ 5 × 10−8 in the stage I and stage acterize new genetic risk factors in other common, complex dis- II analysis but with P > 0.05 in stage II (Supplementary Table 6). It eases (e.g., diabetes). is noteworthy that the magnitude of the associations in stage I did not change substantially if we restricted the analysis to clinically GWAS meta-analysis diagnosed AD cases (Supplementary Table 7 and Supplementary The European Alzheimer & Dementia Biobank (EADB) consor- Fig. 30). Similarly, none of the signals observed appeared to be tium brings together the various European GWAS consortia already especially driven by the UKBB data (Supplementary Table 7 and working on AD. A new dataset of 20,464 clinically diagnosed AD Supplementary Figs. 2–29). Nine of these loci (APP, CCDC6, GRN, cases and 22,244 controls has been collated from 15 European coun- LILRB2, NCK2, TNIP1, TMEM106B, TSPAN14 and SHARPIN) tries. The EADB GWAS results were meta-analyzed with a proxy-AD were recently reported in three articles using part of the GWAS GWASs of the UK Biobank (UKBB) dataset. The UKBB’s proxy-AD data included in our study11–13. We also generated a detailed anal- designation is based on questionnaire data in which individuals are ysis of the human leukocyte antigen (HLA) locus on the basis of asked whether their parents had dementia. This method has been the clinically diagnosed AD cases (Supplementary Tables 8 and 9, used successfully in the past3 but is less specific than a clinical or Supplementary Figs. 31 and 32 and Supplementary Note). pathological diagnosis of AD; hence, we will refer to these cases as proxy AD and related dementia (proxy-ADD). EADB stage I (GWAS Genetic overlap with other neurodegenerative diseases meta-analysis) was based on 39,106 clinically diagnosed AD cases, We tested the association of the lead variants within our new loci 46,828 proxy-ADD cases (as defined in the Supplementary Note), with the risk of developing other neurodegenerative diseases or 401,577 controls (Supplementary Tables 1 and 2) and 21,101,114 AD-related disorders (Supplementary Fig. 33 and Supplementary variants that passed our quality control (Fig. 1; see Supplementary Tables 10–12). We also performed more precise colocalization Fig. 1 for the quantile–quantile plot and genomic inflation factors). analyses (using Coloc R package, https://cran.r-project.org/web/ We selected all variants with a P value below 1 × 10−5 in stage I. We packages/coloc/index.html) for five loci known to be associated defined nonoverlapping regions around these variants, excluded the with Parkinson’s disease (IDUA and CTSB), types of frontotem- region corresponding to APOE and examined the remaining vari- poral dementia (TMEM106B and GRN) and amyotrophic lateral ants in a large follow-up sample that included AD cases and controls sclerosis (TNIP1) (Supplementary Tables 13 and 14). The IDUA sig- from the ADGC, FinnGen and CHARGE consortia (stage II; 25,392 nal for Parkinson’s disease was independent of the signal in ADD AD cases and 276,086 controls). A signal was considered as signifi- (coloc posterior probability (PP)3 = 99.9%), but we were not able to cant on the genome-wide level if it (1) was nominally associated determine whether the CTSB signals colocalized. The TMEM106B (P ≤ 0.05) in stage II, (2) had the same direction of association in the and GRN signals in frontotemporal lobar degeneration with TAR stage I and II analyses and (3) was associated with the ADD risk with DNA-binding protein (TDP-43) inclusions (frontotemporal lobar A full list of author and affiliations appears at the end of the paper. 412 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles APOE PICALM BIN1 36 CLU CR1 32 ABCA7 TREM2 28 MS4A 24 APH1B ZCWPW1/NYAP1 SCIMP/RABEP1 CASS4 20 −log10(P ) SLC24A4/RIN3 INPP5D HLA ABI3 PLCG2 EPHA1 ACE CLNK/HS3ST1 TSPAN14 PTK2B TMEM106B SLC2A4RG SORL1 16 COX7C GRN CELF1/SPI1 RASGEF1C CD2AP SHARPIN APP PRDM7 ECHDC3 IGH gene cluster DOC2A ADAM10 MYO15A WDR81 JAZF1 NME8 12 NCK2 ICA1 LILRB2 FERMT2 ADAMTS1 IDUA MAF FOXF1 SEC61G CTSH KAT8 MAPT KLF16 CTSB SIGLEC11 ABCA1 TSPOAP1 ANK3 PLEKHA1 HS3ST5 ADAM17 MME SPPL2A WDR12 UMAD1 SORT1 PRKD3 TPCN1 RBCK1 RHOH TNIP1 ANKH BLNK SNX1 8 IL34 4 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Chromosome Fig. 1 | Manhattan plot of the stage I results. P values are two-sided raw P values derived from a fixed-effect meta-analysis. Variants with a P value below 1 × 10−36 are not shown. Loci with a genome-wide significant signal are annotated (known loci in black and new loci in red). Variants in new loci are highlighted in red. The red dotted line represents the genome-wide significance level (P = 5 × 10−8), and the black dotted line represents the suggestive significance level (P = 1 × 10−5). degeneration TDP) probably share causal variants with ADD (coloc was observed using a mouse single-cell dataset14 (Supplementary PP4 = 99.8% and coloc PP4 = 80.1%, respectively). Lastly, we were Table 17 and Supplementary Note). not able to determine whether the TNIP1 signals colocalized for Lastly, we looked at whether the relationship between an elevated ADD and amyotrophic lateral sclerosis. microglia Av. Exp. and a genetic association with the ADD risk was specific to particular biological processes (Supplementary Table 18) Pathway analyses by analyzing the interaction between microglia Av. Exp. and pathway Next, we sought to perform a pathway enrichment analysis on the membership in MAGMA15. Of the five most significant interaction stage I association results to gain better biological understanding signals (q ≤ 10−3), two were directly associated with endocytosis pro- of this newly expanded genetic landscape for ADD. Ninety-three cesses (GO:0006898 and GO:0031623); this suggested a functional gene sets were still statistically significant after correction for mul- relationship between microglia and endocytosis, which is known to tiple testing (q ≤ 0.05; Methods and Supplementary Table 15). As be involved in phagocytosis (Supplementary Table 18). It is notewor- described previously, the most significant gene sets are related to thy that we also detected an interaction between GO:1902991 (regu- amyloid and tau5; other significant gene sets are related to lipids, lation of amyloid precursor protein (APP) catabolic process) and endocytosis and immunity (including macrophage and microglial the gene expression level in microglia (q = 1.4 × 10−3; Supplementary cell activation). When restricting this analysis to the meta-analysis Table 18). Even though these data suggest a functional relationship based on the clinically diagnosed AD cases, 54 gene sets were sig- between microglia and APP/amyloid beta (Aβ) peptide pathways, nificant (q ≤ 0.05). Of these 54 gene sets, 33 reached q ≤ 0.05 in this observation reinforces the likely involvement of microglial endo- the stage I analysis and all reached P ≤ 0.05. This indicates that cytosis in AD, a mechanism that is also strongly involved in APP the inclusion of proxy-ADD cases does not cause disease-relevant metabolism16. Of note, there are overall similarities in the interaction biological information to be missed and underlines the additional effects of human and mouse microglia expression with genes in bio- power of this type of analysis. logical pathways of relevance to the AD genetic risk (Supplementary We next performed a single-cell expression enrichment analysis Table 18 and Supplementary Note). by using the average gene expression per nucleus (Av. Exp.) data in the human Allen Brain Atlas (49,495 nuclei from 8 human brains). Gene prioritization Only the microglial expression reached a high level of significance We next attempted to identify the genes most likely to be respon- (P = 1.7 × 10−8; Supplementary Table 16); greater expression corre- sible for the association signal with ADD at each new locus. To sponded to a more significant association with ADD. After adjust- this end, we studied the downstream effects of ADD-associated ing for microglial Av. Exp., the remaining associations became variants on molecular phenotypes (i.e., expression, splicing, pro- nonsignificant; this indicates that microglial Av. Exp. drives all the tein expression, methylation and histone acetylation) in various other cell-type associations. These results were observed whatever cis-quantitative trait locus (cis-QTL) catalogues from AD-relevant the brain region studied (Supplementary Table 16). A similar result tissues, cell types and brain regions. We investigated the genetic Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 413 Articles NATure GeneTiCs Table 1 | Summary of association results in the stage I and stage II analysis for known loci with a genome-wide significant signal Varianta Chromosome Positionb Genec Known locus Minor/major allele MAFd ORe 95% CI P value rs679515 1 207577223 CR1 CR1 T/C 0.188 1.13 1.11–1.15 7.2 × 10−46 rs6733839 2 127135234 BIN1 BIN1 T/C 0.389 1.17 1.16–1.19 6.1 × 10−118 rs10933431 2 233117202 INPP5D INPP5D G/C 0.234 0.93 0.92–0.95 3.6 × 10−18 rs6846529 4 11023507 CLNK CLNK/HS3ST1 C/T 0.283 1.07 1.05–1.08 2.2 × 10−17 rs6605556 6 32615322 HLA-DQA1 HLA G/A 0.161 0.91 0.90–0.93 7.1 × 10−20 rs10947943 6 41036354 UNC5CL TREM2 A/G 0.142 0.94 0.93–0.96 1.1 × 10−9 rs143332484 6 41161469 TREM2 TREM2 T/C 0.013 1.41 1.32–1.50 2.8 × 10−25 rs75932628 6 41161514 TREM2 TREM2 T/C 0.003 2.39 2.09–2.73 2.5 × 10−37 rs60755019 6 41181270 TREML2 TREM2 G/A 0.004 1.55 1.33–1.80 2.1 × 10−8 rs7767350 6 47517390 CD2AP CD2AP T/C 0.271 1.08 1.06–1.09 7.9 × 10−22 rs6966331 7 37844191 EPDR1 NME8 T/C 0.349 0.96 0.94–0.97 4.6 × 10−10 rs7384878 7 100334426 SPDYE3 ZCWPW1/NYAP1 C/T 0.31 0.92 0.91–0.94 1.1 × 10−26 rs11771145 7 143413669 EPHA1 EPHA1 A/G 0.348 0.95 0.93–0.96 3.3 × 10−14 rs73223431 8 27362470 PTK2B PTK2B T/C 0.369 1.07 1.06–1.08 4.0 × 10−22 rs11787077 8 27607795 CLU CLU T/C 0.392 0.91 0.90–0.92 1.7 × 10−44 rs7912495 10 11676714 USP6NL ECHDC3 G/A 0.462 1.06 1.05–1.08 9.7 × 10−19 rs10437655 11 47370397 SPI1 CELF1/SPI1 A/G 0.399 1.06 1.04–1.07 5.3 × 10−14 rs1582763 11 60254475 MS4A4A MS4A A/G 0.371 0.91 0.90–0.92 3.7 × 10−42 rs3851179 11 86157598 EED PICALM T/C 0.358 0.9 0.89–0.92 3.0 × 10−48 rs74685827 11 121482368 SORL1 SORL1 G/T 0.019 1.19 1.13–1.25 2.8 × 10−11 rs11218343 11 121564878 SORL1 SORL1 C/T 0.039 0.84 0.81–0.87 1.4 × 10−21 rs17125924 14 52924962 FERMT2 FERMT2 G/A 0.089 1.1 1.07–1.12 8.3 × 10−16 rs7401792 14 92464917 SLC24A4 SLC24A4/RIN3 G/A 0.371 1.04 1.02–1.05 4.8 × 10−8 rs12590654 14 92472511 SLC24A4 SLC24A4/RIN3 A/G 0.328 0.93 0.92–0.95 4.2 × 10−21 rs8025980 15 50701814 SPPL2A SPPL2A G/A 0.345 0.96 0.94–0.97 1.3 × 10−8 rs602602 15 58764824 MINDY2 ADAM10 A/T 0.28 0.94 0.93–0.96 2.1 × 10−15 rs117618017 15 63277703 APH1B APH1B T/C 0.144 1.11 1.09–1.13 2.2 × 10−25 rs889555 16 31111250 BCKDK KAT8 T/C 0.281 0.95 0.94–0.97 2.0 × 10−11 rs4985556 16 70660097 IL34 IL34 A/C 0.115 1.07 1.05–1.09 6.0 × 10−10 rs12446759 16 81739398 PLCG2 PLCG2 G/A 0.403 0.95 0.94–0.96 1.2 × 10−13 rs72824905 16 81908423 PLCG2 PLCG2 G/C 0.008 0.74 0.68–0.81 8.5 × 10−12 rs7225151 17 5233752 SCIMP SCIMP/RABEP1 A/G 0.124 1.08 1.05–1.10 4.1 × 10−13 rs199515 17 46779275 WNT3 MAPT G/C 0.219 0.94 0.93–0.96 9.3 × 10−13 rs616338 17 49219935 ABI3 ABI3 T/C 0.012 1.32 1.23–1.42 2.8 × 10−14 rs2526377 17 58332680 TSPOAP1 TSPOAP1 G/A 0.445 0.95 0.94–0.97 1.6 × 10−12 rs4277405 17 63471557 ACE ACE C/T 0.384 0.94 0.93–0.95 8.8 × 10−20 rs12151021 19 1050875 ABCA7 ABCA7 A/G 0.336 1.1 1.09–1.12 1.6 × 10−37 rs6014724 20 56423488 CASS4 CASS4 G/A 0.09 0.89 0.87–0.91 4.1 × 10−21 rs2830489 21 26775872 ADAMTS1 ADAMTS1 T/C 0.281 0.95 0.94–0.97 1.7 × 10−10 P values are two-sided raw P values derived from a fixed-effect meta-analysis.CI, confidence interval; OR, odds ratio; MAF, minor allele frequency. Reference single-nucleotide polymorphism (SNP) a (rs) number, according to dbSNP build 153. bGRCh38 assembly. cNearest protein-coding gene according to GENCODE release 33. dWeighted average MAF across all discovery studies. eApproximate OR calculated with respect to the minor allele. colocalization between association signals for the ADD risk and that yielded a total weighted score of between 0 and 100 for each those for the molecular phenotypes and the association between gene (Supplementary Fig. 34 and Supplementary Note). This score the ADD risk and these phenotypes by integrating cis-QTL infor- was used to compare and prioritize genes in the new loci within mation into our ADD GWAS. Moreover, we considered the lead 1 Mb upstream and 1 Mb downstream of the lead variants. Genes variant annotation (the allele frequency, protein-altering effects either were ranked as tier 1 (greater likelihood of being the causal and nearest protein-coding gene) and a genome-wide, high-content risk gene responsible for the ADD signal) or tier 2 (lower likeli- short interfering RNA screen for APP metabolism17. Based on this hood and the absence of a minimum level of evidence as a causal evidence, we developed a systematic gene prioritization strategy risk gene) or were not ranked. 414 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Table 2 | Summary of association results in the stage I and stage II analysis for new loci at the time of analysis with a genome-wide significant signal Locus Varianta Chromosome Positionb Genec Minor/major MAFd ORe 95% CI P value number allele 1 rs141749679 1 109345810 SORT1 C/T 0.004 1.38 1.24–1.54 7.5 × 10−9 2 rs72777026 2 9558882 ADAM17 G/A 0.144 1.06 1.04–1.08 2.7 × 10−8 3 rs17020490 2 37304796 PRKD3 C/T 0.145 1.06 1.04–1.08 3.3 × 10−9 4 rs143080277 2 105749599 NCK2 C/T 0.005 1.47 1.33–1.63 2.1 × 10−13 5 rs139643391 2 202878716 WDR12 T/TC 0.131 0.94 0.92–0.96 1.1 × 10−8 6 rs16824536 3 155069722 MME A/G 0.054 0.92 0.89–0.95 3.6 × 10−8 6 rs61762319 3 155084189 MME G/A 0.026 1.16 1.11–1.21 2.2 × 10−11 7 rs3822030 4 993555 IDUA G/T 0.429 0.95 0.94–0.96 8.3 × 10−12 8 rs2245466 4 40197226 RHOH G/C 0.343 1.05 1.03–1.06 1.2 × 10−9 9 rs112403360 5 14724304 ANKH A/T 0.073 1.09 1.06–1.12 2.3 × 10−9 10 rs62374257 5 86927378 COX7C C/T 0.23 1.07 1.05–1.09 1.4 × 10−15 11 rs871269 5 151052827 TNIP1 T/C 0.326 0.96 0.95–0.97 8.7 × 10−9 12 rs113706587 5 180201150 RASGEF1C A/G 0.11 1.09 1.07–1.12 2.2 × 10−16 13 rs785129 6 114291731 HS3ST5 T/C 0.35 1.04 1.03–1.06 2.4 × 10−9 14 rs6943429 7 7817263 UMAD1 T/C 0.42 1.05 1.03–1.06 1.0 × 10−10 15 rs10952097 7 8204382 ICA1 T/C 0.114 1.07 1.05–1.10 6.8 × 10−9 16 rs13237518 7 12229967 TMEM106B A/C 0.411 0.96 0.94–0.97 4.9 × 10−11 17 rs1160871 7 28129126 JAZF1 G/GTCTT 0.222 0.95 0.93–0.97 9.8 × 10−9 18 rs76928645 7 54873635 SEC61G T/C 0.103 0.93 0.91–0.95 1.6 × 10−10 19 rs1065712 8 11844613 CTSB C/G 0.053 1.09 1.06–1.12 1.9 × 10−9 20 rs34173062 8 144103704 SHARPIN A/G 0.081 1.13 1.09–1.16 1.7 × 10−16 21 rs1800978 9 104903697 ABCA1 G/C 0.13 1.06 1.04–1.08 1.6 × 10−9 22 rs7068231 10 60025170 ANK3 T/G 0.403 0.95 0.94–0.96 3.3 × 10−13 23 rs6586028 10 80494228 TSPAN14 C/T 0.196 0.93 0.91–0.94 2.0 × 10−19 24 rs6584063 10 96266650 BLNK G/A 0.043 0.89 0.86–0.92 6.7 × 10−11 25 rs7908662 10 122413396 PLEKHA1 G/A 0.467 0.96 0.95–0.97 2.6 × 10−9 26 rs6489896 12 113281983 TPCN1 C/T 0.076 1.08 1.05–1.10 1.8 × 10−9 27 rs7157106 14 105761758 IGH gene cluster A/G 0.36 1.05 1.03–1.07 2.0 × 10−8 27 rs10131280 14 106665591 IGH gene cluster A/G 0.133 0.94 0.92–0.96 4.3 × 10−10 28 rs3848143 15 64131307 SNX1 G/A 0.22 1.05 1.04–1.07 8.4 × 10−11 29 rs12592898 15 78936857 CTSH A/G 0.133 0.94 0.92–0.96 4.2 × 10−9 30 rs1140239 16 30010081 DOC2A T/C 0.379 0.94 0.93–0.96 2.6 × 10−13 31 rs450674 16 79574511 MAF C/T 0.373 0.96 0.95–0.98 3.2 × 10−8 32 rs16941239 16 86420604 FOXF1 A/T 0.029 1.13 1.08–1.17 1.3 × 10−8 33 rs56407236 16 90103687 PRDM7 A/G 0.069 1.11 1.08–1.14 6.5 × 10−15 34 rs35048651 17 1728046 WDR81 T/TGAG 0.214 1.06 1.04–1.08 7.7 × 10−11 35 rs2242595 17 18156140 MYO15A A/G 0.112 0.94 0.92–0.96 1.1 × 10−9 36 rs5848 17 44352876 GRN T/C 0.289 1.07 1.06–1.09 2.4 × 10−20 37 rs149080927 19 1854254 KLF16 G/GC 0.48 1.05 1.04–1.07 5.1 × 10−10 38 rs9304690 19 49950060 SIGLEC11 T/C 0.24 1.05 1.03–1.07 4.7 × 10−9 39 rs587709 19 54267597 LILRB2 C/T 0.325 1.05 1.04–1.07 3.6 × 10−11 40 rs1358782 20 413334 RBCK1 A/G 0.246 0.95 0.94–0.97 1.6 × 10−8 41 rs6742 20 63743088 SLC2A4RG T/C 0.221 0.95 0.93–0.97 2.6 × 10−9 42 rs2154481 21 26101558 APP C/T 0.476 0.95 0.94–0.97 1.0 × 10−12 P values are two-sided raw P values derived from a fixed-effect meta-analysis. rs number, according to dbSNP build 153. GRCh38 assembly. Nearest protein-coding gene according to GENCODE release a b c 33. dWeighted average MAF across all discovery studies. eApproximate OR calculated with respect to the minor allele. Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 415 Articles NATure GeneTiCs From all newly identified loci, this gene prioritization yielded 31 Lastly, we used STRING v11 (ref. 19) to analyze protein–pro- tier 1 genes and 24 tier 2. The 55 prioritized genes, the details of tein interaction for (1) previously known AD genes from GWASs, the analyses and the supporting evidence are summarized in Fig. (2) our prioritized new genes (tier 1 in Fig. 2a and Supplementary 2a and the Supplementary Note (Supplementary Tables 19–30 and Table 20) and (3) a combination of the two (Supplementary Note). Supplementary Figs. 35–45). Among the 31 tier 1 genes, we observed The largest networks contained 14, 8 and 30 proteins, respectively that 25 of these genes were the only prioritized gene in their respec- (Supplementary Fig. 46). These networks were larger than would tive locus. For the remaining 6 tier 1 genes, we also found tier 2 genes be expected by chance (respectively, P < 2 × 10−5, P = 2.8 × 10−3 and in their respective locus. We also identified five loci containing sev- P < 2 × 10−5 based on comparison with 50,000 randomly simulated eral tier 2 prioritized genes. In one of these loci, locus 39 (L39), the protein lists matched for the number of proteins and the total num- tier 2 prioritized gene LILRB2 had strong additional support from ber of interactions for each protein). Notably, the number of inter- published literature (Supplementary Note). In five loci, our prioriti- actions between our prioritized genes and previously known genes zation score did not identify sufficient molecular evidence to priori- is also significantly greater than would be expected (P < 1 × 10−4), tize genes with exception of being the nearest gene (L10, L12, L13, indicating that the newly prioritized genes are biologically relevant L14 and L32). Finally, we excluded the complex IGH cluster (L27) in AD. No such enrichment (P = 0.88) was observed for the remain- from gene prioritization analyses due to genomic complexity of the ing genes in the new loci, again highlighting the value of our priori- telomeric locus as a consequence of known fusion events18. tization approach. We highlight two examples, L18 and L23. In L18, the lead vari- We next performed a pathway enrichment analysis of the tier ant, rs76928645 (MAF = 10%), is intergenic and is located more than 1 genes using STRING. We found that several gene sets linked to 100 kb downstream or upstream of the two nearest protein-coding the immune system remained statistically significant after correc- genes (SEC61G and EGFR, respectively). Our gene prioritization anal- tion for multiple testing (Fig. 2b and Supplementary Table 31), yses suggested that EGFR was the only risk gene (Fig. 3). We found that especially regulation of the tumor necrosis factor (TNF)-mediated both the lead variant (rs76928645) and the other nearby variants in signaling pathway (GO:0010803). We report the potential genetic linkage disequilibrium (LD) are significant expression QTLs (eQTLs) implication of the linear ubiquitin chain assembly complex for regulating EGFR expression downstream. The eQTL signals in (LUBAC), which is a major regulator of the aforementioned sig- brain strongly colocalized with the GWAS signal (with eQTL coloc naling pathway20. Two of the LUBAC’s three complements are PP4s of 98.3% in the temporal cortex (TCX) and 99.5% in the dor- encoded by the new tier 1 prioritized genes SHARPIN and RBCK1, solateral prefrontal cortex (DLPFC)). Accordingly, the fine-mapped and the complex’s function is directly regulated by OTULIN (also a expression transcriptome-wide association study (eTWAS) associa- new tier 1 prioritized gene). tions (Fine-mapping Of CaUsal gene Sets (FOCUS) posterior inclu- sion probability (PIP) = 1; eTWAS P = 6.9 × 10−9, eTWAS Z = + 5.8 GRS in the TCX; eTWAS P = 3.1 × 10−11, eTWAS Z = + 6.6 in the DLPFC) We next looked at whether the genetic ADD burden (as measured indicated that genetic downregulation of EGFR expression is associ- by a genetic risk score (GRS)) generated from our genome-wide sig- ated with a lower ADD risk (Fig. 3; Supplementary Tables 22, 24 and nificant variants (n = 83, excluding APOE; Supplementary Table 32) 26; and Supplementary Figs. 36a, 39 and 41). might influence the rate of conversion to AD in (1) individuals from In L23, we observed numerous eQTL-GWAS and methylation several prospective, population-based cohorts and (2) patients with QTL (mQTL)-GWAS hits for TSPAN14 that support the hypoth- mild cognitive impairment (MCI) in prospective memory clinic esis that increased brain expression of TSPAN14 is associated with studies (Supplementary Table 33). We used Cox regression models increased ADD risk. We also identified several splice junctions in to assess the association after adjustment for age at baseline, sex, TSPAN14 whose genetic regulation signals in lymphoblastoid cell the number of APOE-ε4 and APOE-ε2 alleles, and genetic principal lines (LCLs) and brain colocalized with the ADD association signal. components (PCs). These splice junctions were also associated with ADD risk (Fig. 4, In population-based cohorts with clinically diagnosed AD cases, Supplementary Tables 22–28 and Supplementary Figs. 36–42 and the GRS was significantly associated with conversion to AD; this was 44c). As three of these splice junctions were related to new complex shown in a fixed-effect meta-analysis (hazard ratio (HR) (95%CI) cryptic splicing events that were predicted to result in two cryptic per average risk allele = 1.076 (1.064–1.088), P = 9.2 × 10−40; Fig. 5 exons not previously described in known TSPAN14 transcripts and Supplementary Table 34). Likewise, the GRS was significantly (based on GENCODE v38), we designed a long-read single-molecule associated with AD conversion in patients with MCI (HR = 1.056 (Nanopore) sequencing experiment (Supplementary Note) to (1.040–1.072), P = 2.8 × 10−12; Fig. 5 and Supplementary Table 35). validate these cryptic exons on a total of 93 complementary DNA Furthermore, we found that the GRS association increased sig- (cDNA) samples derived from LCLs, frontal cortex and hippo- nificantly when the new variants discovered in the present study campus and consequently validated those cryptic exons (Fig. 4). were added to the previously described variants (Supplementary All three of the validated cryptic splicing events occur within the Table 36) for both population-based studies (HR = 1.052 (1.037– ADAM10-interacting domain of TSPAN14. Cryptic exon 1 is at 1.068), P = 1.5 × 10−11) and MCI cohorts (HR = 1.034 (1.013–1.055), least 45 bp long, and cryptic exon 2 is 118 bp long. P = 1.4 × 10−3). Fig. 2 | Gene prioritization. a, Summary of weighted scores for each evidence category for the prioritized genes in the 42 new genome-wide-significant loci. Using our gene prioritization method, we considered the genes within 1 Mb of each new lead variant and prioritized a total of 55 genes in 42 new loci at two different confidence levels (31 tier 1 genes and 24 tier 2 genes). The leftmost squares indicate the new locus index number. The different types of evidence are colored according to the seven different domains to which they belonged. Weighted scores for each evidence category are rescaled to a 0–100 scale, and the proportions of mean human brain cell-type-specific expression for each gene are also rescaled to a 0–100 scale; darker colors represent higher scores or higher expression proportions. Tier 1 genes are shown in dark green, and tier 2 genes are shown in light green. Only tier 1 and tier 2 genes are shown for each locus. Supplementary Fig. 35 shows full results. MAFs and CADD (v1.6) PHRED scores for rare and/or protein-altering rare variants are labeled in white within the respective squares. b, Pathway enrichment analysis based on the tier 1 gene list. Only the ten strongest associations (according to STRING software) are presented here. coloc, colocalization; eQTL, expression QTL; eTWAS, expression transcriptome-wide association study; GO, Gene Ontology; haQTL, histone acetylation QTL; Mon. Mac., monocytes and macrophages; sTWAS, splicing transcriptome-wide association study; m/haQTL, methylation/histone acetylation QTL; sQTL, splicing QTL; FDR, false discovery rate. 416 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles a Category AS AS WAS AS ns TL eTW eTW sTW t al. 85 coloc euro s /haQ d sT uron L L L L L L L on s t eQT t eQT t eQT t eQT t sQT t sQT t sQT lism rocyte cells c and c and rgic n and tizati ng L ) nt m n ic ne (Win S & pQT a alteri tabo n n n n n n n c coloc coloc coloc priori elial varia varia varia varia varia varia varia varia cytes L colo L colo L colo mate dend Meth Aerg go e glia me st in th Neare sQTL sQTL Gene sQTL Micro Prote Gluta Astro Oligo Endo PWA Lead Lead Lead Meta Lead Lead Lead Lead Lead Rare GAB APP eQT eQT eQT Variant AD−relevant brain regions Microglia Mon. LCL Single-cell expression Blood APP Tier annotation (AMP-AD and GTEx) (MiGA) Mac. (EADB Belgian and GTEx) (Allen Brain Atlas) SORT1 1 0.5% 22.8 1 7.2 6 7.2 45.9 11 22.7 ITGB1BP1 2 100 40 2 23.2 19.7 11 16.3 8.3 21.4 ADAM17 44.4 100 40 1 17.4 14 10.6 17.7 24.3 15.9 EIF2AK2 66.7 2 11.7 14.1 13.3 7.9 20.5 32.5 CEBPZOS 3 66.7 2 11.2 14.3 13 15 23.5 23 PRKD3 2 11.5 9.5 10.8 7.8 30 30.5 NCK2 4 0.5% 1 11.5 7.6 12.2 3.7 59.3 5.7 ICA1L 5 100 88.9 100 100 81.8 75 40 100 1 34.3 24.1 21.7 9.1 3.9 6.9 MME 6 23.1 100 1 21.7 49.2 7.3 4.8 15.7 1.3 DGKQ 7 100 44.4 50 100 66.7 100 100 40 100 1 15.1 18.5 17.2 8.4 18.4 22.5 IDUA 100 33.3 100 100 100 66.7 2 9.9 7.3 24.8 7.4 21.9 28.6 RHOH 8 66.7 33.3 1 0.8 0.6 1.4 1.7 92.6 2.9 OTULIN 9 66.7 88.9 33.3 66.7 100 1 6.3 7.3 10.3 15.7 28.5 31.9 COX7C 10 2 17.3 22.8 20.4 9 12 18.5 TNIP1 11 66.7 1 7.4 8.3 9.8 12.5 31.5 30.6 RASGEF1C 12 2 2 1.1 0.4 2 94.4 0 HS3ST5 13 2 9.9 20.2 9.2 57.2 3.5 0 UMAD1 14 2 15.4 17.6 14.8 15.2 27.6 9.4 ICA1 15 100 1 28.7 26 1.3 1.4 12.3 30.4 TMEM106B 16 66.7 33.3 100 100 66.7 1 18.5 28 10.6 20 10.4 12.6 JAZF1 17 50 100 1 14.6 13.8 4 12 45.1 10.5 EGFR 18 100 100 1 0.6 22.8 34.8 0.5 0.2 41 CTSB 19 66.7 1 8.4 10.4 4.5 4.6 51.8 20.2 SHARPIN 20 19 44.4 50 40 1 6.3 9.4 18.8 17 35.7 12.8 Prioritized gene ABCA1 21 100 1 3.6 10.3 46 8.1 18 14 CCDC6 22 66.7 75 1 22.9 15.9 19.7 9.5 24.2 7.8 ANK3 66.7 2 21.6 22.2 1 52 2.4 0.7 TSPAN14 23 100 100 100 100 100 100 100 100 100 1 4.9 3.5 6.2 4.2 30.9 50.3 BLNK 24 66.7 88.9 75 66.7 1 0.4 0.4 0.9 0.7 97.6 0.1 PLEKHA1 25 66.7 50 33.3 72.7 100 100 66.7 1 22.4 17.1 4.2 29.9 10.2 16.2 RITA1 26 100 100 100 1 20.9 18.7 14.8 12.8 1.7 31 IQCD 66.7 77.8 2 13.2 12.5 40.4 14.4 18.6 0.9 FAM96A 28 100 50 100 66.7 75 2 9.4 13.3 7.3 18.3 36.7 15 SNX1 66.7 44.4 66.7 40 2 8.3 8 9.9 38.9 22.7 12.3 CTSH 29 100 100 100 100 100 75 66.7 100 66.7 50 40 1 1.7 2.4 34.6 1 40.4 19.9 DOC2A 30 10 66.7 77.8 100 100 100 100 100 100 1 34.3 16 17.8 5.6 16.3 10.1 MAF 31 100 1 0.7 33.2 4.1 5.2 56.1 0.6 FOXF1 32 2 0.1 0.1 1 0 0.3 98.6 CDK10 33 33.3 2 20.1 17.8 27.7 11.6 14.3 8.5 PRDM7 50 2 8.5 11.1 36.2 14 21 9.3 WDR81 34 16.4 33.3 50 33.3 66.7 100 1 6.9 8.8 18 13.3 42.3 10.8 SERPINF2 100 50 33.3 100 2 5.5 7.7 9.4 12.3 63.8 1.3 MYO15A 44.4 50 33.3 1 33.3 28.6 8.5 5.8 6.5 17.3 LLGL1 35 66.7 33.3 66.7 2 5.3 6 23.2 53.6 7.4 4.6 TOP3A 100 33.3 2 16.8 20.8 15.7 12.5 14.8 19.4 GRN 36 100 100 1 2.9 3.9 5.1 8.4 40.3 39.3 ATP8B3 66.7 55.6 50 2 9.5 18.2 16.2 22.8 26.4 6.9 REXO1 37 66.7 44.4 100 66.7 2 9.7 9.1 20.2 16.5 24.6 19.8 KLF16 100 44.4 2 9.2 8.6 8.6 4.6 58 11 SIGLEC11 38 100 44.4 100 75 100 1 0.3 0.2 0.2 0.6 98.6 0.1 MYADM 39 33.3 2 17.9 18.5 3.3 41.4 18.7 0.2 LILRB2 2 0 0 0 0.1 99.7 0.1 RBCK1 40 100 77.8 1 7.4 9.9 28 13.4 16.7 24.7 LIME1 41 66.7 100 1 14 16.1 33.1 24.1 7.8 4.9 APP 42 100 100 1 16.1 17.2 3.2 31.1 2.7 29.6 Variant eQTL-GWAS sQTL-GWAS pQTL-GWAS mQTL-GWAS Per category rescaled score Domain annotation integration integration integration integration or per gene cell-type 0 25 50 75 100 or type haQTL-GWAS APP Prioritized Single-cell expression (%) integration metabolism gene tier expression b Number of Number of GO term Term description tier 1 genes in FDR Matching proteins in the network genes pathway –5 GO:0001775 Cell activation 12 1,024 4.09 × 10 GRN,CTSH, BLNK, NCK2, DGKQ, EGFR, APP, ADAM17, CTSB, RHOH, TSPAN14, MME –4 GO:0002376 Immune system process 15 2,370 2.60 × 10 GRN,CTSH, BLNK, NCK2, OTULIN, APP, LIME1, ADAM17, TNIP1, CTSB, RBCK1, PLEKHA1, RHOH, TSPAN14, MME –4 GO:0009966 Regulation of signal transduction 17 3,033 2.60 × 10 GRN,CTSH, BLNK, NCK2, DGKQ, EGFR, OTULIN, APP, ADAM17, TNIP1, RBCK1, PLEKHA1, ABCA1, RHOH, SHARPIN, TSPAN14, RITA1 –4 GO:0010646 Regulation of cell communication 18 3,327 2.60 × 10 GRN,CTSH, BLNK, NCK2, DGKQ, EGFR, OTULIN, APP, ADAM17, TNIP1, RBCK1, PLEKHA1, ABCA1, RHOH, ICA1, SHARPIN, TSPAN14, RITA1 –4 GO:0023051 Regulation of signaling 18 3,360 2.60 × 10 GRN,CTSH, BLNK, NCK2, DGKQ, EGFR, OTULIN, APP, ADAM17, TNIP1, RBCK1, PLEKHA1, ABCA1, RHOH, ICA1, SHARPIN, TSPAN14, RITA1 –4 GO:0045321 Leukocyte activation 10 894 2.60 × 10 GRN, CTSH, BLNK, NCK2, APP, ADAM17, CTSB, RHOH, TSPAN14, MME –4 GO:0048583 Regulation of response to stimulus 19 3,882 2.60 × 10 GRN,CTSH, BLNK, NCK2, DGKQ, EGFR, OTULIN, APP, LIME1, ADAM17, TNIP1, RBCK1, PLEKHA1, ABCA1, RHOH, SHARPIN, TSPAN14, SIGLEC11, RITA1 –4 GO:0048584 Positive regulation of response to stimulus 14 2,054 2.60 × 10 CTSH, BLNK, NCK2, EGFR, OTULIN, APP, LIME1, ADAM17, TNIP1, RBCK1, PLEKHA1, SHARPIN, TSPAN14 –4 GO:0010803 Regulation of TNF-mediated signaling pathway 4 56 4.30 × 10 OTULIN, ADAM17, RBCK1, SHARPIN –3 GO:0042058 Regulation of EGFR signaling pathway 4 80 1.50 × 10 NCK2, EGFR, APP, ADAM17 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 417 Articles NATure GeneTiCs a SEC61G locus (18) eQTL-GWAS regional plot b EGFR 17.5 r2 EADB GWAS stage I 15.0 PP4: 98.3% MayoRNAseq TCX sQTL –log10(P) 0.8 eTWAS Z: +5.8 15 12.5 0.6 10.0 0.4 0.2 7.5 rs76928645 10 5.0 2.5 0 5 54.7 Mb 54.8 Mb 54.9 Mb 55 Mb Chromosome 7 17.5 0 SEC61G eQTL-MayoRNAseq TCX GWAS or eQTL –log10 (P) 15.0 0 2.5 5.0 7.5 12.5 EADB GWAS –log10(P ) 10.0 c rs76928645 eQTL-EGFR MayoRNAseq TCX 7.5 5.0 rs76928645 eQTL P: 3 × 10–18 Z: +5.8 2.5 eQTL β: –0.39 2 Normalized EGFR expression 0 54.7 Mb 54.8 Mb 54.9 Mb 55 Mb Chromosome 7 17.5 0 EGFR eQTL-MayoRNAseq TCX 15.0 rs76928645 12.5 10.0 7.5 –2 5.0 2.5 0 CC CT TT 54.7 Mb 54.8 Mb 54.9 Mb 55 Mb rs76928645 genotype Chromosome 7 SEC61G EGFR LOC100996654 Fig. 3 | Regulation of EGFR expression by the ADD-risk-associated and colocalized brain eQTLs within the intergenic SEC61G locus. a, The regional plot of the new SEC61G locus (L18) shows the EADB GWAS stage I (n = 487,511) ADD association signal within 200 kb of the intergenic lead variant, rs76928645 (the two closest protein-coding genes, SEC61G and EGFR, are more than 100 kb from the lead variant), together with the eQTLs in the same region identified for SEC61G and EGFR expression separately in the TCX (MayoRNAseq TCX eQTL catalog based on n = 259 individuals). The rs7692864 lead variant is shown in purple, and LD r2 values (calculated for the EADB Trans-Omics for Precision Medicine (TOPMed) dataset (n = 42,140) with respect to the lead variant) are indicated on a color scale. y axis, −log10 for the GWAS or eQTL P value; x axis, hg38 genomic position on chromosome 7. b, Colocalization between the EGFR eQTL signal (MayoRNAseq TCX, n = 259 individuals) and the EADB GWAS stage I (n = 487,511) signal (eQTL coloc PP4 = 98.3%); with the significant eTWAS association (eTWAS P = 6.9 × 10−9 and eTWAS Z = 5.8) and fine-mapped (FOCUS PIP = 1) eTWAS association in the same catalog. y axis, eQTL −log10(P) value; x axis, GWAS −log10(P) value. LD r2 values and color scales are as in a. c, The eQTL violin plot shows a significant association (eQTL P = 3 × 10−18) between the rs76928645 lead variant genotype and EGFR expression in the TCX (MayoRNAseq TCX, n = 259 individuals), where the protective allele T is associated with lower EGFR expression (eQTL β, −0.39). Each data point represents a sample whose normalized EGFR expression value is indicated on the y axis, and the rs76928645 genotype information is indicated on the x axis. The lower and upper hinges of the boxes represent the first and third quantiles, the whiskers extend 1.5 times the interquartile range from the hinges and the line represents the median. Importantly, the results of our meta-analysis suggest that In an MCI cohort setting, this effect of the GRS corresponds the risk of conversion to AD rises with the number of risk to a median AD conversion probability within 3 years of 21.9% alleles from non-APOE risk variants in the GRS by 1.9-fold in in patients with a GRS below the first decile (range, 4.1–34.9%) population-based cohorts (HR = 1.93 (1.75–2.13); Fig. 5) and and 37.5% (range, 10.8–56.2%) in patients with a GRS above the 1.6-fold in MCI cohorts (HR = 1.63 (1.42–1.87); Fig. 6) on top of ninth decile. There was a consistent increase in probability between effects of age and the APOE ε4 allele. These observations result these deciles in all cohorts (median (range), 13.8% (6.6–25.0%); from the comparison of hypothetical individuals with a GRS value Supplementary Table 38). at the first decile of the distribution versus those with a GRS value To better define the GRS discriminative ability regarding AD con- at the ninth decile (Fig. 6). With regard to APOE, carrying an version, we assessed the improvements in three indices of predictive additional APOE-ε4 allele was associated with a slightly higher performance after adding the GRS to a Cox model containing age, increase in the AD risk in population-based cohorts (HR = 2.19 sex, PCs and the number of APOE-ε4 and APOE-ε2 alleles as covari- (2.03–2.37)) and MCI cohorts (HR = 1.90 (1.73–2.07)). There ates (Supplementary Tables 34 and 35). We found a small but con- was no interaction between the GRS and the number of APOE-ε4 sistent increase in the discrimination between AD converters and alleles (Supplementary Table 37). nonconverters, as indicated by the concordance index (C-index) in 418 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles population-based cohorts (Δ5years-C-indexfixed-effects = 0.002 (0.0004– These pathway enrichment analyses also confirmed the involve- 0.004)) and MCI cohorts (Δ3years-C-indexfixed-effects = 0.007 (0.001– ment of innate immunity and microglial activation in ADD 0.012)). This finding was further supported by small-to-moderate (Supplementary Table 15). Our single-cell expression enrich- increases in the continuous NRI (net reclassification improvement) ment analysis also highlighted genes expressed in microglia index in population-based cohorts (NRI5year-fixed-effects = 0.248 (0.159– (Supplementary Tables 16 and 17). Indeed, three of our prioritized 0.336)) and MCI cohorts (NRI3year-fixed-effects = 0.232 (0.140–0.325)); (tier 1) genes (RHOH, BLNK and SIGLEC11) and two of our tier 2 this indicates that the risk assignment is more appropriate to indi- genes (LILRB2 and RASGE1FC) appeared to be mainly expressed in viduals when the GRS is taken into account21. Furthermore, an microglia (>90% relative to the total expression summed across cell increase in the index of prediction accuracy (IPA) was observed in types; Fig. 2a and Supplementary Table 45). Importantly, SIGLEC11 all of the population-based cohorts (average Δ5years-IPAfixed-effects = 0. and LILRB2 have already been linked to Aβ peptides/amyloid 29% (0.23%–0.35%)) and all but one of the MCI cohorts (average plaques25,26. Δ3years-IPAfixed-effects = 1.53% (1.31%–1.76%)), indicating an overall Here, we also provide genetic evidence of the LUBAC’s poten- improvement in predictive performance. As expected, the amount tial implication in ADD. Two of the LUBAC’s three complements of improvement in this index varied greatly from one cohort to are encoded by SHARPIN and RBCK1, and the LUBAC is regulated another, given its dependency on incidence rates. The value of add- by OTULIN; all three genes were found to be high-confidence, pri- ing the new genetic variants was emphasized by the fact that effect oritized risk genes in our study. The LUBAC is the only E3 ligase sizes (as measured by the indices of predictive ability) were lower known to form linear ubiquitin chains de novo through ubiquitin’s when only previously known AD risk variants were included in the N-terminal methionine. The complex has mostly been studied in GRS (Supplementary Table 39). the context of inflammation, innate immunity and defense against The results were similar when we (1) computed indices for other intracellular pathogens. For instance, the LUBAC is reportedly follow-up time points, (2) applied a random effects meta-analysis, essential for NLRP3 inflammasome activation27 and thus acts as a (3) considered conversion to all-cause-dementia as the outcome and key innate immune regulator28. In turn, the NLRP3 inflammasome (4) excluded the Framingham Heart Study (FHS), as it was part of is essential for the development and progression of Aβ pathology in the stage II of the GWAS from which ORs for PRS computation were mice29 and may drive tau pathology through Aβ-induced microglial extracted (Supplementary Tables 34–44 and Supplementary Fig. 47). activation30. The LUBAC is also reportedly involved in autophagy, and linear ubiquitin chain modifications of TDP-43-positive neuro- Discussion nal cytoplasmic inclusions have been described as potential induc- Our meta-analysis combined a large, new case–control study with ers of autophagic clearance31. Lastly, the LUBAC has been studied as previous GWASs. We identified 75 independent loci for ADD; 33 a regulator of TNF-α signaling in particular20. had been reported previously, and 42 correspond to new signals Interestingly, the TNF-α signaling pathway was also flagged by at the time of this analysis. The prioritized genes and their poten- other genetic findings in our study (Supplementary Fig. 48). For tial impact on the pathophysiology of AD are described in the example, ADAM17 (also known as TNF-α-converting enzyme) is Supplementary Note. of pivotal importance in the activation of TNF-α signaling32. For Our pathway enrichment analyses removed ambiguities con- TNIP1, its gene product (TNF-α-induced protein 3-interacting pro- cerning the involvement of tau-binding proteins and APP/Aβ tein 1) is involved in the inhibition of the TNF-α signaling pathway peptide metabolism in late-onset AD processes at a much higher and nuclear factor κB activation/translocation33. Additional signal level than had been described previously5. It is noteworthy that related to TNF-α is the one found at SPPL2A (one of the 33 con- new genetic risk factors are often first evaluated in the context of firmed loci). The protein encoded by SPPL2A is involved in non- known pathways; many new research approaches were developed to canonical shedding of TNF-α34, and PGRN has been described as a systematically characterize putative links among APP metabolism, TNF receptor ligand and an antagonist of TNF-α signaling35. Several tau function and ADD genetic risk factors22,23. This approach can lines of evidence had linked the inhibition of TNF-α signaling with lead to circular reasoning and thus artificial enrichment in specific reduction of both Aβ and tau pathologies in vivo36,37. Although a processes. However, we implicate ADAM17, a gene whose protein potential inflammatory connection has been suggested for TNF-α product is known to carry α-secretase activity as ADAM10 (ref. through the activation of NLRP3 inflammasome38, the TNF-α sig- 24 ). This observation suggests that the nonamyloidogenic pathway naling pathway is also involved in many other brain physiological for APP metabolism might be deregulated in AD. In addition to functions (e.g., synaptic plasticity in neurons) and pathophysiologi- APP, we also identified six highly plausible prioritized (tier 1) genes cal processes (e.g., synapse loss) in the brain39. Furthermore, the (ICA1L, DGKQ, ICA1, DOC2A, WDR81 and LIME1) that are likely involvement of the TNF-α signaling pathway and the LUBAC might to modulate the metabolism of APP. be important in cell types other than microglia in AD. It is important Fig. 4 | Focus on TSPAN14 locus. a, Splicing QTL (sQTL)-GWAS integration results. Known TSPAN14 transcripts (GENCODE v38; green, coding sequences; gray, noncoding) plotted with −log10(P) for (1) EADB GWAS stage I (n = 487,511) signal (black), (2) sQTL signal for chr10:80509471–80510106 junction (supporting cryptic exon 1) in the EADB Belgian LCL sQTL catalog (n = 70 individuals, blue) and (3) sQTL signal for chr10:80512269–80512719 junction in the MayoRNAseq TCX sQTL catalog (n = 259 individuals, red); hg38 genomic position is shown above. LCL and brain-based sQTL coloc and sTWAS analyses associate ADD risk with these junctions that suggest cryptic splicing within ADAM10-interacting domain of TSPAN14 (magenta), which was predicted to result in two cryptic exons. b, Long-read sequencing validation of TSPAN14 cryptic exons. Nanopore sequencing results (Supplementary Note) in the zoomed-in region of chr10:80506973–80516400 (cumulative coverage in log10 scale). Pooled LCL cDNA sample sequenced for cDNA Amplicon2 shown in blue. cDNA Amplicon1 was sequenced on biologically independent hippocampal (HPC; n = 16, red), frontal cortex (FC; n = 18, pink) and LCL (n = 59, orange) cDNA samples. Green, canonical exons (8–12); dotted black lines, canonical splicing; blue, cryptic exon 1 (>45 bp); red, cryptic exon 2 (118 bp). All annotated junctions use canonical splice donor (GT) and acceptor (AG) sites. c,d, sQTL-GWAS colocalization plots for chr10:80509471– 80510106 (supporting cryptic exon 1) in the EADB Belgian LCL sQTL catalog (n = 70 individuals) (c) and chr10:80512269–80512719 (supporting cryptic exon 2) in the MayoRNAseq TCX sQTL catalog (n = 259 individuals) (d). sQTL signals for the two junctions colocalize with ADD signal (PP4s of 98.8% and 97.4%, respectively), and sTWAS associates with increased preference for the cryptic splicing with decreased ADD risk (sTWAS P = 6.28 × 10−12 and 1.6 × 10−13, sTWAS Z = −6.9 and −7.4, respectively). y axis, sQTL −log10(P); x axis, EADB GWAS stage I −log10(P). LD r2 values calculated within EADB-TOPMed dataset (n = 42,140) based on the lead variant rs6586028 (purple) are indicated on a color scale. Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 419 Articles NATure GeneTiCs to note that six of our prioritized (tier 1) genes (ICA1L, EGFR, A better understanding of the etiology of ADD might also result RITA1, MYO15A, LIME1 and APP) are expressed at a low level in from the observation that the risks of developing ADD and fronto- microglia (<10%, relative to the total expression summed across cell temporal dementia are associated with the same causal variants in types; Supplementary Table 45), emphasizing that ADD results from GRN and TMEM106B. This association might be due to the misclas- complex crosstalk between different cell types in the brain23,40. It is sification of clinical diagnosis of AD and the presence of proxy-ADD also noteworthy that the EGFR pathway is known to interact with cases in the UKBB. However, GRN and TMEM106B have also been the TNF-α signaling pathway41, which suggests interplay between linked to brain health and many other neurodegenerative diseases. the two signaling pathways during the ADD development. For instance, GRN and TMEM106B are reportedly potential genetic a Chr10 80,460 80,470 80,480 80,490 80,500 80,510 80,520 80,530 Kb 19 GWAS signal 0 19 –log10(P ) sQTL signal for cryptic exon 1 0 19 sQTL signal for cryptic exon 2 0 Known TSPAN14 transcripts (GENCODE v38) Cryptic splicing (sQTL coloc and sTWAS hits) ADAM10-interacting domain cDNA amplicon2 Nanopore cDNA amplicon1 sequencing amplicons b Chr10 80,508 80,510 80,512 80,514 80,516 Kb 2.5 Cumulative coverage in log10 scale LCL amplicon2 Nanopore 0 sequencing 4.8 HPC amplicon1 Nanopore 0 sequencing 4.8 FC amplicon1 Nanopore 0 sequencing 5.3 LCL amplicon1 Nanopore 0 sequencing Exon 8 Exon 9 Exon 10 Exon 11 Exon 12 Canonical splicing GT AG GT AG GT AG GT AG GT AG Cryptic splicing Cryptic exon 1 Cryptic exon 2 (>45 bp) (118 bp) c Chr10:80509471–80510106 d Chr10:80512269–80512719 PP4: 98.8% 15 PP4: 97.4% EADB Belgian LCL sQTL –log10(P ) MayoRNAseq TCX sQTL –log10(P ) 15 sTWAS Z: –6.9 sTWAS Z: –7.4 10 10 5 5 0 0 0 5 10 0 5 10 EADB GWAS –log10(P ) EADB GWAS –log10(P ) 420 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles a b Cohorts HR (95% CI) Cohorts HR (95% CI) 3C 1.055 (1.033, 1.077) FACE 1.070 (1.046, 1.095) AMC 1.035 (0.993, 1.080) AgeCoDe 1.035 (0.999, 1.072) ADC 1.043 (1.007, 1.080) VITA 1.054 (0.997, 1.115) UAN 1.066 (1.008, 1.127) MAS 1.086 (1.005, 1.173) DCN 1.086 (0.987, 1.196) FHS 1.097 (1.065, 1.129) HAN 1.034 (0.988, 1.081) RS1 1.093 (1.074, 1.112) UHA 1.102 (0.877, 1.386) RS2 1.094 (1.042, 1.148) ZIM 1.030 (0.917, 1.158) FE model: P = 9.23 × 10–40 1.076 (1.064, 1.088) FE model: P = 2.81 × 10–12 1.056 (1.040, 1.072) –4 –5 RE model: P = 2.23 × 10 1.073 (1.050, 1.097) RE model: P = 4.44 × 10 1.056 (1.041, 1.071) I2 = 56.8%, Q(6) = 13.88, P = 0.03 I2 = 0%,Q(7) = 4.27, P = 0.75 0.800 0.913 1.025 1.137 1.250 0.800 0.913 1.025 1.137 1.250 HR HR Fig. 5 | Association between the GRS and the risk of progression to AD. a,b, Meta-analysis results of the association between the GRS and the risk of progression to AD in population-based cohorts (n = 17,545 independent samples) (a) and MCI cohorts (n = 4,114 independent samples) (b). Data are presented as HR together with 95% CIs derived from Cox regression analyses for each individual cohort. HRs indicate the effect of the GRS as the increment in the AD risk associated with each additional average risk allele in the GRS. Null hypothesis testing is based on a meta-analysis of individual cohort effects using fixed effects (FE) and random effects (RE) models. Resulting HRs and 95% CIs and the respective Z test and associated two-sided P value are shown at the bottom of the figure. Heterogeneity between cohorts is indicated by the I2 index together with the respective Cochran’s Q statistic (distributed as χ² statistic), associated degrees of freedom and P value. 3C, Three-City Study; AgeCoDe, German study on aging cognition and dementia; AMC, additional, independent memory clinic cohort from Fundacio ACE; DCN, German Dementia Competence Network study; FACE, Fundacio ACE memory clinic cohort; FHS, Framingham Heart Study; HAN, BALTAZAR multicenter prospective memory clinic study; MAS, Sydney Memory and Ageing Study; RS1, Rotterdam Study first cohort; RS2, Rotterdam Study second cohort; VITA, Vienna Transdanube Aging study; UAN, memory clinic cohort from the Hospital Network Antwerp; UHA, University of Halle memory clinic cohort; ZIM, Heidelberg/Mannheim memory clinic sample. risk factors for differential aging in the cerebral cortex42 and cogni- to a 1.6- to 1.9-fold increase in the AD risk from the lowest to the tive impairment in amyotrophic lateral sclerosis43 and Parkinson’s highest decile, in addition to the effects of age and APOE status. We disease44,45. Lastly, both GRN and TMEM106B have already been also showed that in addition to known risk variants, the new risk associated with neuropathological features of AD46–48. Taken as variants identified in the present study are significantly associated a whole, these data may thus emphasize a potential continuum with progression to AD. The results of future GWASs are expected between neurodegenerative diseases in which common pathologi- to further improve AD-risk prediction. Hence, the GRS will help to cal mechanisms are driven by GRN and TMEM106B. Interestingly, sharpen the threshold that differentiates between people at risk of both GRN and TMEM106B are reported to be involved in defective progressing to dementia and those who are not. endosome/lysosome trafficking/function49,50, a defect that is also A recent study estimated that fewer than 100 causal common observed in AD. variants may explain the entire AD risk52; if that estimate is correct, By applying a GRS derived from all the genome-wide-significant then our study might have already characterized a large proportion variants discovered in this study, we identified an association with of this genetic component due to common variants. However, sev- the risk of incident AD in prospective population-based cohorts eral reasons strongly underscore the need for additional efforts to and with the risk of progression over time from MCI to AD (Fig. 5 fully characterize the still-missing AD genetic component. First, it is and Supplementary Table 33). In patients with MCI, previous asso- probable that additional, yet-unknown loci bear common variants ciations of AD risk with a GRS built on previously known genetic modulating the risk for AD. Second, identification of rare variants AD risk variants has been inconsistent51. It is important to note that with very low frequencies is a major challenge for genetic studies, the GRS has an impact on the AD risk in addition to that of age and because available samples with sequencing data in AD are under- that the GRS’s effect is independent of APOE status. With a view to powered. Notably, almost all the genes with rare variants associated translating genetic findings into preventive measures and personal- with AD risk also present common variants associated with AD ized medicine, we also sought to provide the GRS’s added value for risk (i.e., TREM2, SORL1, ABCA7, ABCA1, PLCγ2 and ADAM10)53. risk prediction by calculating the discriminative capacity through Third, gene–gene and gene–environment interactions have not yet three different indices. Overall, the indices suggested that the effect been studied in detail. Hence, by increasing the GWAS sample size size for the association between the GRS and AD was small but sig- and improving imputation panels, conventional and (above all) nificant. Despite this modest effect, the inclusion of the GRS into more complex analyses will have more statistical power and should the predictive model consistently improved the assignment of the enable the characterization of associations with rare/structural vari- risk of progression, as expressed by the net reclassification improve- ants. Lastly, higher-powered GWASs of multiancestry populations ment (NRI) index21. Importantly, the cumulative improvements in will be particularly welcome for characterizing potential new genetic risk prediction (due to inclusion of the new variants in the GRS) led risk factors, improving fine-mapping approaches and developing Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 421 Articles NATure GeneTiCs Highest decile Lowest decile Highest decile Lowest decile 1.00 1.0 0.9 1 – AD conversion probability 1 – AD conversion probability 0.95 0.8 0.7 0.90 0.6 0.85 0.5 0 2.5 5.0 7.5 10.0 0 1 2 3 4 5 Time (years) Time (years) Fig. 6 | Risk of progression to AD according to the GRS. a,b, Representative plots of the progression to AD over 10 years in the population-based 3C study (a) and the progression from MCI to AD over 5 years in the Fundació ACE cohort (b). The figures show the probabilities of conversion (survival probabilities) to AD (y axes) for a hypothetical participant with average covariates (mean values for age and PCs, and the mode for sex and APOE) and a GRS at the first (lowest) decile (in blue) or a GRS at the ninth (highest) decile (red). The shaded regions correspond to the 95% CI. specific GRSs (because GRSs developed with European-ancestry 10. Jun, G. et al. A novel Alzheimer disease locus located near the gene encoding populations are known to be less effective with other ancestries). tau protein. Mol. Psychiatry 21, 108–117 (2016). 11. Schwartzentruber, J. et al. Genome-wide meta-analysis, fine-mapping and In conclusion, we have validated 33 previous loci, doubled the integrative prioritization implicate new Alzheimer’s disease risk genes. Nat. total number of genetic loci associated with the ADD risk, expanded Genet. 53, 392–402 (2021). our current knowledge of the pathophysiology of ADD, identified 12. de Rojas, I. et al. Common variants in Alzheimer’s disease and risk new opportunities for the development of GRSs and gene-specific stratification by polygenic risk scores. Nat. Commun. 12, 3417 (2021). treatments and opened up a pathway to translational genomics and 13. Wightman, D. P. et al. A genome-wide association study with 1,126,563 personalized medicine. individuals identifies new risk loci for Alzheimer’s disease. Nat. Genet. 53, 1276–1282 (2021). 14. Skene, N. G. et al. Genetic identification of brain cell types underlying Online content schizophrenia. Nat. Genet. 50, 825–833 (2018). Any methods, additional references, Nature Research report- 15. de Leeuw, C. A., Stringer, S., Dekkers, I. A., Heskes, T. & Posthuma, D. ing summaries, source data, extended data, supplementary infor- Conditional and interaction gene-set analysis reveals novel functional mation, acknowledgements, peer review information; details of pathways for blood pressure. Nat. Commun. 9, 3768 (2018). 16. Haass, C., Kaether, C., Thinakaran, G. & Sisodia, S. Trafficking and proteolytic author contributions and competing interests; and statements of processing of APP. Cold Spring Harb. Perspect. Med. 2, a006270 (2012). data and code availability are available at https://doi.org/10.1038/ 17. Chapuis, J. et al. Genome-wide, high-content siRNA screening identifies the s41588-022-01024-z. Alzheimer’s genetic risk factor FERMT2 as a major modulator of APP metabolism. Acta Neuropathol. 133, 955–966 (2017). Received: 16 June 2021; Accepted: 27 January 2022; 18. Cleynen, A. et al. Expressed fusion gene landscape and its impact in multiple Published online: 4 April 2022 myeloma. Nat. Commun. 8, 1893 (2017). 19. Szklarczyk, D. et al. STRING v11: Protein-protein association networks with increased coverage, supporting functional discovery in genome-wide References experimental datasets. Nucleic Acids Res. 47, D607–D613 (2019). 1. Gatz, M. et al. Role of genes and environments for explaining Alzheimer 20. Spit, M., Rieser, E. & Walczak, H. Linear ubiquitination at a glance. J. Cell Sci. disease. Arch. Gen. Psychiatry 63, 168–174 (2006). 132, jcs208512 (2019). 2. Bellenguez, C., Grenier-Boley, B. & Lambert, J. C. Genetics of Alzheimer’s disease: where we are, and where we are going. Curr. Opin. Neurobiol. 61, 21. Pencina, M. J., D’Agostino, R. B., Pencina, K. M., Janssens, A. C. J. W. & 40–48 (2020). Greenland, P. Interpreting incremental value of markers added to risk 3. Jansen, I. E. et al. Genome-wide meta-analysis identifies new loci and prediction models. Am. J. Epidemiol. 176, 473–481 (2012). functional pathways influencing Alzheimer’s disease risk. Nat. Genet. 51, 22. Dourlen, P., Chapuis, J. & Lambert, J.-C. Using high-throughput animal or 404–413 (2019). cell-based models to functionally characterize GWAS signals. Curr. Genet. 4. Chang, C. C. et al. Second-generation PLINK: rising to the challenge of larger Med. Rep. 6, 107–115 (2018). and richer datasets. Gigascience 4, 7 (2015). 23. Dourlen, P., Kilinc, D., Malmanche, N., Chapuis, J. & Lambert, J. C. The new 5. Kunkle, B. W. et al. Genetic meta-analysis of diagnosed Alzheimer’s disease genetic landscape of Alzheimer’s disease: from amyloid cascade to genetically identifies new risk loci and implicates Aβ, tau, immunity and lipid processing. driven synaptic failure hypothesis? Acta Neuropathol. 138, 221–236 (2019). Nat. Genet. 51, 414–430 (2019). 24. Deuss, M., Reiss, K. & Hartmann, D. Part-time α-secretases: the functional 6. Lambert, J. C. et al. Meta-analysis of 74,046 individuals identifies 11 new biology of ADAM 9, 10 and 17. Curr. Alzheimer Res. 5, 187–201 (2008). susceptibility loci for Alzheimer’s disease. Nat. Genet. 45, 1452–1458 (2013). 25. Kim, T. et al. Human LilrB2 is a β-amyloid receptor and its murine homolog 7. Liu, J. Z., Erlich, Y. & Pickrell, J. K. Case-control association mapping by PirB regulates synaptic plasticity in an Alzheimer’s model. Science 341, proxy using family history of disease. Nat. Genet. 49, 325–331 (2017). 1399–1404 (2013). 8. Marioni, R. E. et al. GWAS on family history of Alzheimer’s disease. Transl. 26. Salminen, A. & Kaarniranta, K. Siglec receptors and hiding plaques in Psychiatry 8, 99 (2018). Alzheimer’s disease. J. Mol. Med. 87, 697–701 (2009). 9. Sims, R. et al. Rare coding variants in PLCG2, ABI3, and TREM2 implicate 27. Rodgers, M. A. et al. The linear ubiquitin assembly complex (LUBAC) microglial-mediated innate immunity in Alzheimer’s disease.Nat. Genet. 49, is essential for NLRP3 inflammasome activation. J. Exp. Med. 211, 1373–1384 (2017). 1333–1347 (2014). 422 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles 28. Iwai, K. LUBAC-mediated linear ubiquitination: a crucial regulator of 44. Baizabal-Carvallo, J. F. & Jankovic, J. Parkinsonism, movement disorders and immune signaling. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 97, 120–133 (2021). genetics in frontotemporal dementia. Nat. Rev. Neurol. 12, 175–185 (2016). 29. Venegas, C. et al. Microglia-derived ASC specks crossseed amyloid-β in 45. Tropea, T. F. et al. TMEM106B Effect on cognition in Parkinson disease and Alzheimer’s disease. Nature 552, 355–361 (2017). frontotemporal dementia. Ann. Neurol. 85, 801–811 (2019). 30. Ising, C. et al. NLRP3 inflammasome activation drives tau pathology. Nature 46. Mendsaikhan, A., Tooyama, I. & Walker, D. G. Microglial progranulin: 575, 669–673 (2019). involvement in Alzheimer’s disease and neurodegenerative diseases. Cells 8, 31. Nakayama, Y. et al. Linear polyubiquitin chain modification of TDP- 230 (2019). 43-positive neuronal cytoplasmic inclusions in amyotrophic lateral sclerosis. 47. Li, Z. et al. The TMEM106B FTLD-protective variant, rs1990621, is also J. Neuropathol. Exp. Neurol. 79, 256–265 (2020). associated with increased neuronal proportion. Acta Neuropathol. 139, 32. Black, R. A. et al. A metalloproteinase disintegrin that releases 45–61 (2020). tumour-necrosis factor-∅ from cells. Nature 385, 729–733 (1997). 48. Yang, H. S. et al. Genetics of gene expression in the aging human brain reveal 33. Verstrepen, L., Carpentier, I., Verhelst, K. & Beyaert, R. ABINs: A20 binding TDP-43 proteinopathy pathophysiology. Neuron 107, 496–508.e6 (2020). inhibitors of NF-κB and apoptosis signaling. Biochem. Pharmacol. 78, 49. Paushter, D. H., Du, H., Feng, T. & Hu, F. The lysosomal function of 105–114 (2009). progranulin, a guardian against neurodegeneration. Acta Neuropathol. 136, 34. Spitz, C. et al. Non-canonical Shedding of TNFα by SPPL2a is determined 1–17 (2018). by the conformational flexibility of its transmembrane helix. iScience 23, 50. Feng, T., Lacrampe, A. & Hu, F. Physiological and pathological functions of 101775 (2020). TMEM106B: a gene associated with brain aging and multiple brain disorders. 35. Tang, W. et al. The growth factor progranulin binds to tnf receptors Acta Neuropathol. 141, 327–339 (2021). and is therapeutic against inflammatory arthritis in mice. Science 332, 51. Lacour, A. et al. Genome-wide significant risk factors for Alzheimer’s disease: 478–484 (2011). role in progression to dementia due to Alzheimer’s disease among subjects 36. He, P. et al. Deletion of tumor necrosis factor death receptor inhibits amyloid with mild cognitive impairment. Mol. Psychiatry 22, 153–160 (2017). β generation and prevents learning and memory deficits in Alzheimer’s mice. 52. Zhang, Q. et al. Risk prediction of late-onset Alzheimer’s disease implies an J. Cell Biol. 178, 829–841 (2007). oligogenic architecture. Nat. Commun. 11, 1–11 (2020). 37. Shi, J. Q. et al. Anti-TNF-α reduces amyloid plaques and tau phosphorylation 53. Holstege, H. et al. Exome sequencing identifies novel AD-associated genes. and induces CD11c-positive dendritic-like cell in the APP/PS1 transgenic medRxiv 18, 24 (2020). mouse brains. Brain Res. 1368, 239–247 (2011). 38. Bezbradica, J. S., Coll, R. C. & Schroder, K. Sterile signals generate weaker Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in and delayed macrophage NLRP3 inflammasome responses relative to published maps and institutional affiliations. microbial signals. Cell. Mol. Immunol. 14, 118–126 (2017). 39. Decourt, B., Lahiri, D. K. & Sabbagh, M. N. Targeting tumor necrosis factor Open Access This article is licensed under a Creative Commons alpha for Alzheimer’s disease. Curr. Alzheimer Res. 14, 412–425 (2016). Attribution 4.0 International License, which permits use, sharing, adap- 40. De Strooper, B. & Karran, E. The cellular phase of Alzheimer’s disease. Cell tation, distribution and reproduction in any medium or format, as long 164, 603–615 (2016). as you give appropriate credit to the original author(s) and the source, provide a link to 41. Gong, K. et al. TNF-driven adaptive response mediates resistance to EGFR the Creative Commons license, and indicate if changes were made. The images or other inhibition in lung cancer. J. Clin. Invest. 128, 2500–2518 (2018). third party material in this article are included in the article’s Creative Commons license, 42. Rhinn, H. & Abeliovich, A. Differential aging analysis in human cerebral unless indicated otherwise in a credit line to the material. If material is not included in cortex identifies variants in TMEM106B and GRN that regulate aging the article’s Creative Commons license and your intended use is not permitted by statu- phenotypes. Cell Syst. 4, 404–415 (2017). tory regulation or exceeds the permitted use, you will need to obtain permission directly 43. Vass, R. et al. Risk genotypes at TMEM106B are associated with cognitive from the copyright holder. To view a copy of this license, visit http://creativecommons. impairment in amyotrophic lateral sclerosis. Acta Neuropathol. 121, org/licenses/by/4.0/. 373–380 (2011). © The Author(s) 2022 Céline Bellenguez 1,567 ✉, Fahri Küçükali 2,3,4,567, Iris E. Jansen5,6,567, Luca Kleineidam7,8,9,567, Sonia Moreno-Grau10,11,567, Najaf Amin 12,13,567, Adam C. Naj14,15,567, Rafael Campos-Martin8,567, Benjamin Grenier-Boley1, Victor Andrade7,8, Peter A. Holmans 16, Anne Boland17, Vincent Damotte1, Sven J. van der Lee 5,18, Marcos R. Costa 1,19, Teemu Kuulasmaa 20, Qiong Yang 21,22, Itziar de Rojas 10,11, Joshua C. Bis 23, Amber Yaqub 12, Ivana Prokic 12, Julien Chapuis 1, Shahzad Ahmad 12,24, Vilmantas Giedraitis 25, Dag Aarsland26,27, Pablo Garcia-Gonzalez 10,11, Carla Abdelnour10,11, Emilio Alarcón-Martín10,28, Daniel Alcolea 11,29, Montserrat Alegret10,11, Ignacio Alvarez 30,31, Victoria Álvarez32,33, Nicola J. Armstrong34, Anthoula Tsolaki 35,36, Carmen Antúnez37, Ildebrando Appollonio38,39, Marina Arcaro40, Silvana Archetti41, Alfonso Arias Pastor42,43, Beatrice Arosio44,45, Lavinia Athanasiu46, Henri Bailly47, Nerisa Banaj48, Miquel Baquero 49, Sandra Barral50,51,52, Alexa Beiser20,53, Ana Belén Pastor 54, Jennifer E. Below55, Penelope Benchek56,57, Luisa Benussi58, Claudine Berr 59, Céline Besse17, Valentina Bessi 60,61, Giuliano Binetti58,62, Alessandra Bizarro63, Rafael Blesa11,29, Mercè Boada 10,11, Eric Boerwinkle64,65, Barbara Borroni66, Silvia Boschi67, Paola Bossù 68, Geir Bråthen 69,70, Jan Bressler 64,71, Catherine Bresner 16, Henry Brodaty 34,72, Keeley J. Brookes 73, Luis Ignacio Brusco74,75,76, Dolores Buiza-Rueda11,77, Katharina Bûrger78,79, Vanessa Burholt 80,81, William S. Bush 82, Miguel Calero 10,54,83, Laura B. Cantwell84, Geneviève Chene85,86, Jaeyoon Chung 87, Michael L. Cuccaro88, Ángel Carracedo89,90, Roberta Cecchetti91, Laura Cervera-Carles 11,29, Camille Charbonnier92, Hung-Hsin Chen93, Caterina Chillotti94, Simona Ciccone45, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 423 Articles NATure GeneTiCs Jurgen A. H. R. Claassen95, Christopher Clark96, Elisa Conti38, Anaïs Corma-Gómez97, Emanuele Costantini 98, Carlo Custodero99, Delphine Daian17, Maria Carolina Dalmasso 8, Antonio Daniele98, Efthimios Dardiotis100, Jean-François Dartigues101, Peter Paul de Deyn102, Katia de Paiva Lopes 103,104,105,106, Lot D. de Witte 106, Stéphanie Debette 101, Jürgen Deckert107, Teodoro del Ser 54, Nicola Denning108, Anita DeStefano20,21,109, Martin Dichgans 78,110, Janine Diehl-Schmid 111, Mónica Diez-Fairen 30,31, Paolo Dionigi Rossi45, Srdjan Djurovic 46, Emmanuelle Duron47, Emrah Düzel112,113, Carole Dufouil 85,86, Gudny Eiriksdottir114, Sebastiaan Engelborghs115,116,117,118, Valentina Escott-Price 15,108, Ana Espinosa10,11, Michael Ewers 78,79, Kelley M. Faber119, Tagliavini Fabrizio 120, Sune Fallgaard Nielsen121, David W. Fardo122, Lucia Farotti123, Chiara Fenoglio124, Marta Fernández-Fuertes97, Raffaele Ferrari125,126, Catarina B. Ferreira127, Evelyn Ferri45, Bertrand Fin17, Peter Fischer128, Tormod Fladby 129, Klaus Fließbach8,9, Bernard Fongang130, Myriam Fornage 70,71, Juan Fortea11,29, Tatiana M. Foroud119, Silvia Fostinelli58, Nick C. Fox 131, Emlio Franco-Macías132, María J. Bullido 11,133,134, Ana Frank-García11,133,135, Lutz Froelich 136, Brian Fulton-Howard137, Daniela Galimberti 40,124, Jose Maria García-Alberca 11,138, Pablo García-González10, Sebastian Garcia-Madrona139, Guillermo Garcia-Ribas139, Roberta Ghidoni58, Ina Giegling140, Giaccone Giorgio120, Alison M. Goate 137, Oliver Goldhardt111, Duber Gomez-Fonseca141, Antonio González-Pérez 142, Caroline Graff143,144, Giulia Grande145, Emma Green 146, Timo Grimmer111, Edna Grünblatt 147,148,149, Michelle Grunin57, Vilmundur Gudnason 150, Tamar Guetta-Baranes151, Annakaisa Haapasalo 152, Georgios Hadjigeorgiou153, Jonathan L. Haines82, Kara L. Hamilton-Nelson154, Harald Hampel155, Olivier Hanon47, John Hardy126, Annette M. Hartmann140, Lucrezia Hausner136, Janet Harwood 16, Stefanie Heilmann-Heimbach 156, Seppo Helisalmi157,158, Michael T. Heneka7,9, Isabel Hernández10,11, Martin J. Herrmann 107, Per Hoffmann156, Clive Holmes159, Henne Holstege 5,18, Raquel Huerto Vilas42,43, Marc Hulsman5,18, Jack Humphrey103,104,105,160, Geert Jan Biessels 161, Xueqiu Jian 130, Charlotte Johansson 143, Gyungah R. Jun 87, Yuriko Kastumata 162, John Kauwe163, Patrick G. Kehoe 164, Lena Kilander21, Anne Kinhult Ståhlbom143, Miia Kivipelto165,166,167,168, Anne Koivisto157,169,170, Johannes Kornhuber 171, Mary H. Kosmidis 172, Walter A. Kukull173, Pavel P. Kuksa15, Brian W. Kunkle153, Amanda B. Kuzma84, Carmen Lage 11,174, Erika J. Laukka145,175, Lenore Launer 176,177, Alessandra Lauria63, Chien-Yueh Lee 15, Jenni Lehtisalo157,178, Ondrej Lerch179,180, Alberto Lleó 11,29, William Longstreth Jr181, Oscar Lopez 22, Adolfo Lopez de Munain 11,182, Seth Love164, Malin Löwemark21, Lauren Luckcuck16, Kathryn L. Lunetta 20, Yiyi Ma 18,183, Juan Macías97, Catherine A. MacLeod 184, Wolfgang Maier7,9, Francesca Mangialasche165, Marco Spallazzi51, Marta Marquié 10,11, Rachel Marshall16, Eden R. Martin154, Angel Martín Montes 11,133,135, Carmen Martínez Rodríguez33, Carlo Masullo185, Richard Mayeux50,186, Simon Mead 187, Patrizia Mecocci91, Miguel Medina 11,54, Alun Meggy108, Shima Mehrabian52, Silvia Mendoza138, Manuel Menéndez-González 33, Pablo Mir11,188, Susanne Moebus189, Merel Mol 77, Laura Molina-Porcel190,191, Laura Montrreal10, Laura Morelli192, Fermin Moreno11,182, Kevin Morgan 193, Thomas Mosley194, Markus M. Nöthen 156, Carolina Muchnik 74,195, Shubhabrata Mukherjee196, Benedetta Nacmias 60,197, Tiia Ngandu178, Gael Nicolas 92, Børge G. Nordestgaard 121,198, Robert Olaso17, Adelina Orellana10,11, Michela Orsini98, Gemma Ortega10,11, Alessandro Padovani65, Caffarra Paolo199, Goran Papenberg145, Lucilla Parnetti123, Florence Pasquier 200, Pau Pastor 30,31, Gina Peloso 20,53, Alba Pérez-Cordón 10, Jordi Pérez-Tur 11,201,202, Pierre Pericard 203, Oliver Peters204,205, Yolande A. L. Pijnenburg5, Juan A. Pineda 97, Gerard Piñol-Ripoll42,43, 424 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Claudia Pisanu 206, Thomas Polak107, Julius Popp207,208,209, Danielle Posthuma 6, Josef Priller205,210, Raquel Puerta 10, Olivier Quenez 92, Inés Quintela89, Jesper Qvist Thomassen 211, Alberto Rábano11,54, Innocenzo Rainero66, Farid Rajabli154, Inez Ramakers212, Luis M. Real 97,213, Marcel J. T. Reinders 214, Christiane Reitz 186,214,215, Dolly Reyes-Dumeyer183,215, Perry Ridge216, Steffi Riedel-Heller 217, Peter Riederer218, Natalia Roberto10, Eloy Rodriguez-Rodriguez 11,174, Arvid Rongve 219,220, Irene Rosas Allende32,33, Maitée Rosende-Roca10,11, Jose Luis Royo221, Elisa Rubino222, Dan Rujescu140, María Eugenia Sáez 142, Paraskevi Sakka223, Ingvild Saltvedt 69,224, Ángela Sanabria10,11, María Bernal Sánchez-Arjona132, Florentino Sanchez-Garcia225, Pascual Sánchez Juan 11,174, Raquel Sánchez-Valle226, Sigrid B. Sando68,69, Chloé Sarnowski 64, Claudia L. Satizabal 21,109,130, Michela Scamosci91, Nikolaos Scarmeas50,227, Elio Scarpini40,124, Philip Scheltens5, Norbert Scherbaum228, Martin Scherer229, Matthias Schmid9,230, Anja Schneider7,9, Jonathan M. Schott 131, Geir Selbæk 129,231, Davide Seripa232, Manuel Serrano233, Jin Sha14, Alexey A. Shadrin 46, Olivia Skrobot164, Susan Slifer154, Gijsje J. L. Snijders106, Hilkka Soininen 157, Vincenzo Solfrizzi 99, Alina Solomon157,165, Yeunjoo Song 57, Sandro Sorbi60,197, Oscar Sotolongo-Grau 10, Gianfranco Spalletta 48, Annika Spottke9,234, Alessio Squassina 235, Eystein Stordal 236, Juan Pablo Tartan10, Lluís Tárraga10,11, Niccolo Tesí5,18, Anbupalam Thalamuthu34, Tegos Thomas35,36, Giuseppe Tosto 50,183, Latchezar Traykov52, Lucio Tremolizzo38,39, Anne Tybjærg-Hansen198,211, Andre Uitterlinden 237, Abbe Ullgren143, Ingun Ulstein231, Sergi Valero10,11, Otto Valladares 15, Christine Van Broeckhoven 2,3,238, Jeffery Vance88, Badri N. Vardarajan50, Aad van der Lugt239, Jasper Van Dongen2,3,4, Jeroen van Rooij77,239, John van Swieten77, Rik Vandenberghe240,241, Frans Verhey212, Jean-Sébastien Vidal 47, Jonathan Vogelgsang 242,243, Martin Vyhnalek179,180, Michael Wagner 7,9, David Wallon 244, Li-San Wang15, Ruiqi Wang20,21, Leonie Weinhold230, Jens Wiltfang 242,245,246, Gill Windle184, Bob Woods 184, Mary Yannakoulia 247, Habil Zare 130, Yi Zhao15, Xiaoling Zhang 248, Congcong Zhu248, Miren Zulaica11,249, EADB, GR@ACE, DEGESCO, EADI, GERAD, Demgene, FinnGen, ADGC, CHARGE, Lindsay A. Farrer 20,87,109, Bruce M. Psaty 22,84,250, Mohsen Ghanbari 12, Towfique Raj 103,104,105,160, Perminder Sachdev 34, Karen Mather 34, Frank Jessen7,9, M. Arfan Ikram 12, Alexandre de Mendonça127, Jakub Hort177,179, Magda Tsolaki 35,36, Margaret A. Pericak-Vance 152, Philippe Amouyel 1, Julie Williams 16,108, Ruth Frikke-Schmidt 198,211, Jordi Clarimon11,29, Jean-François Deleuze17, Giacomina Rossi120, Sudha Seshadri 21,109,130, Ole A. Andreassen 46, Martin Ingelsson25, Mikko Hiltunen19,568, Kristel Sleegers 2,3,4,568, Gerard D. Schellenberg15,568, Cornelia M. van Duijn 12,13,568, Rebecca Sims 16,568, Wiesje M. van der Flier 5,568, Agustín Ruiz10,11,568, Alfredo Ramirez 7,8,9,130,251,568 and Jean-Charles Lambert 1,568 ✉ 1 Université de Lille, INSERM, CHU Lille, Institut Pasteur Lille, U1167-RID-AGE, Facteurs de risque et déterminants moléculaires des maladies liées au vieillissement, Lille, France. 2Complex Genetics of Alzheimer’s Disease Group, VIB Center for Molecular Neurology, VIB, Antwerp, Belgium. 3Laboratory of Neurogenetics, Institute Born - Bunge, Antwerp, Belgium. 4Department of Biomedical Sciences, University of Antwerp, Antwerp, Belgium. 5Alzheimer Center Amsterdam, Department of Neurology, Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, the Netherlands. 6 Department of Complex Trait Genetics, Center for Neurogenomics and Cognitive Research, Amsterdam Neuroscience, Vrije University, Amsterdam, the Netherlands. 7Department of Neurodegenerative Diseases and Geriatric Psychiatry, University Hospital Bonn, Bonn, Germany. 8Division of Neurogenetics and Molecular Psychiatry, Department of Psychiatry and Psychotherapy, University of Cologne, Medical Faculty, Cologne, Germany. 9German Center for Neurodegenerative Diseases (DZNE Bonn), Bonn, Germany. 10Research Center and Memory Clinic Fundació ACE, Institut Català de Neurociències Aplicades, Universitat Internacional de Catalunya, Barcelona, Spain. 11CIBERNED, Network Center for Biomedical Research in Neurodegenerative Diseases, National Institute of Health Carlos III, Madrid, Spain. 12Department of Epidemiology, Erasmus MC, Rotterdam, the Netherlands. 13Nuffield Department of Population Health, Oxford University, Oxford, UK. 14Department of Biostatistics, Epidemiology, and Informatics, Penn Neurodegeneration Genomics Center, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. 15Department of Pathology and Laboratory Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. 16MRC Centre for Neuropsychiatric Genetics and Genomics, Division of Psychological Medicine and Clinical Neuroscience, School of Medicine, Cardiff University, Cardiff, UK. 17CEA, Centre National de Recherche en Génomique Humaine, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 425 Articles NATure GeneTiCs Université Paris-Saclay, Evry, France. 18Section Genomics of Neurodegenerative Diseases and Aging, Department of Human Genetics Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, the Netherlands. 19Brain Institute, Federal University of Rio Grande do Norte, Natal, Brazil. 20 Institute of Biomedicine, University of Eastern Finland, Kuopio, Finland. 21Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA. 22Framingham Heart Study, Framingham, MA, USA. 23Cardiovascular Health Research Unit, Department of Medicine, University of Washington, Seattle, WA, USA. 24LACDR, Leiden, the Netherlands. 25Department of Public Health and Carins Sciences/Geriatrics, Uppsala University, Uppsala, Sweden. 26 Centre of Age-Related Medicine, Stavanger University Hospital, Stavanger, Norway. 27Institute of Psychiatry, Psychology & Neuroscience, London, UK. 28 Department of Surgery, Biochemistry and Molecular Biology, School of Medicine, University of Málaga, Málaga, Spain. 29Department of Neurology, II B Sant Pau, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain. 30Fundació Docència i Recerca MútuaTerrassa and Movement Disorders Unit, Department of Neurology, University Hospital MútuaTerrassa, Terrassa, Spain. 31Memory Disorders Unit, Department of Neurology, Hospital Universitari Mutua de Terrassa, Terrassa, Spain. 32Laboratorio de Genética, Hospital Universitario Central de Asturias, Oviedo, Spain. 33 Servicio de Neurología, Hospital Universitario Central de Asturias- Oviedo and Instituto de Investigación Biosanitaria del Principado de Asturias, Oviedo, Spain. 34Centre for Healthy Brain Ageing, School of Psychiatry, Faculty of Medicine, University of New South Wales, Sydney, New South Wales, Australia. 35 First Department of Neurology, Medical School, Aristotle University of Thessaloniki, Thessaloniki, Greece. 36Alzheimer Hellas, Thessaloniki, Greece. 37 Unidad de Demencias, Hospital Clínico Universitario Virgen de la Arrixaca, Murcia, Spain. 38School of Medicine and Surgery, University of Milano-Bicocca, Milano, Italy. 39Neurology Unit, San Gerardo Hospital, Monza, Italy. 40Fondazione IRCCS Ca’Granda, Ospedale Policlinico, Milan, Italy. 41 Department of Laboratory Diagnostics, III Laboratory of Analysis, Brescia Hospital, Brescia, Italy. 42Unitat Trastorns Cognitius, Hospital Universitari Santa Maria de Lleida, Lleida, Spain. 43Institut de Recerca Biomedica de Lleida (IRBLLeida), Lleida, Spain. 44Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy. 45Geriatic Unit, Fondazione Cà Granda, IRCCS Ospedale Maggiore Policlinico, Milan, Italy. 46NORMENT Centre, University of Oslo, Oslo, Norway. 47EA 4468, Université de Paris, APHP, Hôpital Broca, Paris, France. 48Laboratory of Neuropsychiatry, Department of Clinical and Behavioral Neurology, IRCCS Santa Lucia Foundation, Rome, Italy. 49Servei de Neurologia, Hospital Universitari i Politècnic La Fe, Valencia, Spain. 50Taub Institute on Alzheimer’s Disease and the Aging Brain, Department of Neurology, Columbia University, New York, NY, USA. 51Unit of Neurology, University of Parma and AOU, Parma, Italy. 52Clinic of Neurology, UH ‘Alexandrovska’, Medical University - Sofia, Sofia, Bulgaria. 53Boston University and the NHLBI’s Framingham Heart Study, Boston, MA, USA. 54CIEN Foundation/Queen Sofia Foundation Alzheimer Center, Madrid, Spain. 55 Vanderbilt Brain Institute, Vanderbilt University, Nashville, TN, USA. 56Cleveland Institute for Computational Biology, Case Western Reserve University, Cleveland, OH, USA. 57Department of Population and Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH, USA. 58Molecular Markers Laboratory, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy. 59Neuropsychiatry: Epidemiological and Clinical Research, PSNREC, Université de Montpellier, INSERM U1061, Montpellier, France. 60Department of Neuroscience, Psychology, Drug Research and Child Health, University of Florence, Florence, Italy. 61Azienda Ospedaliero-Universitaria Careggi, Florence, Italy. 62MAC - Memory Clinic, IRCCS Istituto Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy. 63Geriatrics Unit, Fondazione Policlinico A. Gemelli IRCCS, Rome, Italy. 64Human Genetics Center, School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA. 65Human Genome Sequencing Center, Baylor College of Medicine, Houston, TX, USA. 66Centre for Neurodegenerative Disorders, Department of Clinical and Experimental Sciences, University of Brescia, Brescia, Italy. 67 Department of Neuroscience “Rita Levi Montalcini”, University of Torino, Torino, Italy. 68Experimental Neuro-psychobiology Laboratory, Department of Clinical and Behavioral Neurology, IRCCS Santa Lucia Foundation, Rome, Italy. 69Department of Neurology and Clinical Neurophysiology, University Hospital of Trondheim, Trondheim, Norway. 70Department of Neuromedicine and Movement Science, Norwegian University of Science and Technology, Trondheim, Norway. 71School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA. 72Dementia Centre for Research Collaboration, School of Psychiatry, University of New South Wales, Sydney, New South Wales, Australia. 73Biosciences, School of Science and Technology, Nottingham Trent University, Nottingham, UK. 74Centro de Neuropsiquiatría y Neurología de la Conducta (CENECON), Facultad de Medicina, Universidad de Buenos Aires (UBA), C.A.B.A., Buenos Aires, Argentina. 75Departamento Ciencias Fisiológicas UAII, Facultad de Medicina, UBA, C.A.B.A., Buenos Aires, Argentina. 76Hospital Interzonal General de Agudos Eva Perón, San Martín, Buenos Aires, Argentina. 77Department of Neurology, Erasmus MC, Rotterdam, the Netherlands. 78Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig Maximilians Universität (LMU), Munich, Germany. 79German Center for Neurodegenerative Diseases (DZNE, Munich), Munich, Germany. 80Faculty of Medical & Health Sciences, University of Auckland, Auckland, New Zealand. 81Wales Centre for Ageing & Dementia Research, Swansea University, Wales, New Zealand. 82Department of Population & Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH, USA. 83UFIEC, Instituto de Salud Carlos III, Madrid, Spain. 84 Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA, USA. 85INSERM, Bordeaux Population Health Research Center, UMR 1219, ISPED, CIC 1401-EC, Université de Bordeaux, Bordeaux, France. 86Pole Santé Publique, CHU de Bordeaux, Bordeaux, France. 87Medicine Biomedical Genetics Boston University School of Medicine, Boston, MA, USA. 88Dr. John T. Macdonald Foundation Department of Human Genetics, University of Miami, Miami, FL, USA. 89Grupo de Medicina Xenómica, Centro Nacional de Genotipado (CEGEN-PRB3-ISCIII), Universidade de Santiago de Compostela, Santiago de Compostela, Spain. 90Fundación Pública Galega de Medicina Xenómica- CIBERER-IDIS, University of Santiago de Compostela, Santiago de Compostela, Spain. 91Institute of Gerontology and Geriatrics, Department of Medicine and Surgery, University of Perugia, Perugia, Italy. 92 Department of Genetics and CNR-MAJ, Normandie University, UNIROUEN, INSERM U1245, CHU Rouen, Rouen, France. 93Division of Genetic Medicine, Vanderbilt University, Nashville, TN, USA. 94Unit of Clinical Pharmacology, University Hospital of Cagliari, Cagliari, Italy. 95Radboudumc Alzheimer Center, Department of Geriatrics, Radboud University Medical Center, Nijmegen, the Netherlands. 96Institute for Regenerative Medicine, University of Zürich, Schlieren, Switzerland. 97Unidad Clínica de Enfermedades Infecciosas y Microbiología, Hospital Universitario de Valme, Sevilla, Spain. 98Department of Neuroscience, Catholic University of Sacred Heart, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy. 99University of Bari, “A. Moro”, Bary, Italy. 100School of Medicine, University of Thessaly, Larissa, Greece. 101Bordeaux Population Health Research Center, University Bordeaux, INSERM, Bordeaux, France. 102Department of Neurology, University Medical Center Groningen, Groningen, the Netherlands. 103Ronald M. Loeb Center for Alzheimer’s Disease, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 104Department of Genetics and Genomic Sciences & Icahn Institute for Data Science and Genomic Technology, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 105Estelle and Daniel Maggin Department of Neurology, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 106Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 107Department of Psychiatry, Psychosomatics and Psychotherapy, Center of Mental Health, University Hospital, Wuerzburg, Germany. 108UKDRI@ Cardiff, School of Medicine, Cardiff University, Cardiff, UK. 109Department of Neurology, Boston University School of Medicine, Boston, MA, USA. 110Munich Cluster for Systems Neurology (SyNergy), Munich, Germany. 111Klinikum rechts der Isar, Department of Psychiatry and Psychotherapy, Technical University of Munich, School of Medicine, Munich, Germany. 112Institute of Cognitive Neurology and Dementia Research (IKND), Otto-Von-Guericke University, Magdeburg, Germany. 113German Center for Neurodegenerative Diseases (DZNE), Magdeburg, Germany. 114Icelandic Heart Association, Kopovagur, Iceland. 115Center for Neurosciences, Vrije Universiteit Brussel (VUB), Brussels, Belgium. 116Reference Center for Biological Markers of Dementia (BIODEM), Institute Born-Bunge, University of Antwerp, Antwerp, Belgium. 117Institute Born-Bunge, University of Antwerp, Antwerp, Belgium. 118Department of Neurology, UZ Brussel, Brussels, Belgium. 119Department of Medical and Molecular Genetics, Indiana University, Indianapolis, IN, USA. 120Fondazione IRCCS, Istituto Neurologico Carlo Besta, Milan, Italy. 121Department of Clinical Biochemistry, Herlev and Gentofte Hospital, Herlev, Denmark. 122 Sanders-Brown Center on Aging, Department of Biostatistics, University of Kentucky, Lexington, KY, USA. 123Centre for Memory Disturbances, Lab of 426 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Clinical Neurochemistry, Section of Neurology, University of Perugia, Perugia, Italy. 124University of Milan, Milan, Italy. 125Laboratory of Neurogenetics, Department of Internal Medicine, Texas Tech University Health Science Center, Lubbock, TX, USA. 126Reta Lila Weston Research Laboratories, Department of Molecular Neuroscience, UCL Institute of Neurology, London, UK. 127Faculty of Medicine, University of Lisbon, Lisbon, Portugal. 128Department of Psychiatry, Social Medicine Center East- Donauspital, Vienna, Austria. 129Institute of Clinical Medicine, University of Oslo, Oslo, Norway. 130Glenn Biggs Institute for Alzheimer’s & Neurodegenerative Diseases, University of Texas Health Sciences Center, San Antonio, TX, USA. 131Dementia Research Centre, UCL Queen Square Institute of Neurology, London, UK. 132Unidad de Demencias, Servicio de Neurología y Neurofisiología. Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocío/CSIC/Universidad de Sevilla, Seville, Spain. 133Instituto de Investigacion Sanitaria ‘Hospital la Paz’ (IdIPaz), Madrid, Spain. 134Centro de Biología Molecular Severo Ochoa (UAM-CSIC), Madrid, Spain. 135Hospital Universitario la Paz, Madrid, Spain. 136 Department of Geriatric Psychiatry, Central Institute for Mental Health, Mannheim, University of Heidelberg, Heidelberg, Germany. 137Department of Genetics and Genomic Sciences, Ronald M. Loeb Center for Alzheimer’s Disease Icahn School of Medicine at Mount Sinai, New York, NY, USA. 138 Alzheimer Research Center & Memory Clinic, Andalusian Institute for Neuroscience, Málaga, Spain. 139Hospital Universitario Ramon y Cajal, IRYCIS, Madrid, Spain. 140Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria. 141Department of Biostatistics, Epidemiology, and Informatics Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 142CAEBI, Centro Andaluz de Estudios Bioinformáticos, Sevilla, Spain. 143Center for Alzheimer Research, Department NVS, Division of Neurogeriatrics, Karolinska Institutet, Stockholm, Sweden. 144Unit for Hereditary Dementias, Karolinska University Hospital-Solna, Stockholm, Sweden. 145Aging Research Center, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet and Stockholm University, Stockholm, Sweden. 146Institute of Public Health, University of Cambridge, Cambridge, UK. 147 Department of Child and Adolescent Psychiatry and Psychotherapy, University Hospital of Psychiatry Zurich, University of Zurich, Zurich, Switzerland. 148 Neuroscience Center Zurich, University of Zurich and ETH Zurich, Zurich, Switzerland. 149Zurich Center for Integrative Human Physiology, University of Zurich, Zurich, Switzerland. 150Icelandic Heart Association, Faculty of Medicine, University of Iceland, Reykjavik, Iceland. 151Human Genetics, School of Life Sciences, Life Sciences Building, University Park, University of Nottingham, Nottingham, UK. 152AI Virtanen Institute for Molecular Sciences, University of Eastern Finland, Kuopio, Finland. 153Department of Neurology, Medical School, University of Cyprus, Nicosia, Cyprus. 154The John P. Hussman Institute for Human Genomics, University of Miami, Miami, FL, USA. 155GRC 21, Alzheimer Precision Medicine Initiative (APMI), Sorbonne University, AP-HP, Pitié-Salpêtrière Hospital, Paris, France. 156Institute of Human Genetics, University of Bonn, School of Medicine & University Hospital Bonn, Bonn, Germany. 157 Institute of Clinical Medicine, Neurology, University of Eastern, Kuopio, Finland. 158Institute of Clinical Medicine, Internal Medicine, University of Eastern Finland, Kuopio, Finland. 159Clinical and Experimental Science, Faculty of Medicine, University of Southampton, Southampton, UK. 160Nash Family Department of Neuroscience & Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 161Department of Neurology, UMC Utrecht Brain Center, Utrecht, the Netherlands. 162Biostatistics, University of Kentucky College of Public Health, Lexington, KY, USA. 163Department of Biology, Brigham Young University, Provo, UT, USA. 164Translational Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK. 165Division of Clinical Geriatrics, Center for Alzheimer Research, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm, Sweden. 166Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland. 167Neuroepidemiology and Ageing Research Unit, School of Public Health, Imperial College London, London, UK. 168Research & Development, UnitStockholms Sjukhem, Stockholm, Sweden. 169Department of Neurology, Kuopio University Hospital, Kuopio, Finland. 170Department of Neurosciences, University of Helsinki and Department of Geriatrics, Helsinki University Hospital, Helsinki, Finland. 171Department of Psychiatry and Psychotherapy, Universitätsklinikum Erlangen, and Friedrich-Alexander Universität Erlangen-Nürnberg, Erlangen, Germany. 172Laboratory of Cognitive Neuroscience, School of Psychology, Aristotle University of Thessaloniki, Thessaloniki, Greece. 173Department of Epidemiology, University of Washington, Seattle, WA, USA. 174Neurology Service, Marqués de Valdecilla University Hospital (University of Cantabria and IDIVAL), Santander, Spain. 175Stockholm Gerontology Research Center, Stockholm, Sweden. 176Laboratory of Epidemiology, Demography, and Biometry, National Institute of Aging, The National Institutes of Health, Bethesda, MD, USA. 177Intramural Research Program/National Institute on Aging/National Institutes of Health, Bethesda, MD, USA. 178Public Health Promotion Unit, Finnish Institute for Health and Welfare, Helsinki, Finland. 179Memory Clinic, Department of Neurology, Charles University, 2nd Faculty of Medicine and Motol University Hospital, Praha, Czechia. 180International Clinical Research Center, St. Anne’s University Hospital Brno, Brno, Czechia. 181Departments of Neurology and Epidemiology, University of Washington, Seattle, WA, USA. 182 Department of Neurology, Hospital Universitario Donostia, OSAKIDETZA-Servicio Vasco de Salud, San Sebastian, Spain. 183Department of Neurology, Columbia University, New York, NY, USA. 184School of Health Sciences, Bangor University, Bangor, UK. 185Institute of Neurology, Catholic University of the Sacred Heart, Rome, Italy. 186Gertrude H. Sergievsky Center, Columbia University, New York, NY, USA. 187MRC Prion Unit at UCL, UCL Institute of Prion Diseases, London, UK. 188Unidad de Trastornos del Movimiento, Servicio de Neurología y Neurofisiología. Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocío/CSIC/Universidad de Sevilla, Seville, Spain. 189Institute for Urban Public Health, University Hospital of University Duisburg-Essen, Essen, Germany. 190Neurological Tissue Bank of the Biobanc-Hospital Clinic-IDIBAPS, Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain. 191Alzheimer’s Disease and Other Cognitive Disorders Unit, Neurology Department, Hospital Clinic, Barcelona, Spain. 192 Laboratory of Brain Aging and Neurodegeneration, FIL-CONICET, Buenos Aires, Argentina. 193Human Genetics, School of Life Sciences, University of Nottingham, Nottingham, UK. 194Memory Impairment and Neurodegenerative Dementia (MIND) Center, University of Mississippi Medical Center, Jackson, MS, USA. 195Laboratorio de Bioquímica Molecular, Facultad de Medicina, Instituto de Investigaciones Médicas A. Lanari, UBA, C.A.B.A, Buenos Aires, Argentina. 196Department of Medicine, University of Washington, Seattle, WA, USA. 197IRCCS Fondazione Don Carlo Gnocchi, Florence, Italy. 198 Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark. 199DIMEC, University of Parma, Parma, Italy. 200Resources and Research Memory Center (MRRC) of Distalz, LicendUniversity of Lille, INSERM, CHU Lille, UMR1172, Lille, France. 201Institut de Biomedicina de València-CSIC CIBERNED, València, Spain. 202Unitat Mixta de de Neurología y Genética, Institut d’Investigació Sanitària La Fe, València, Spain. 203US 41-UMS 2014-PLBS, bilille, Université de Lille, CNRS, INSERM, CHU Lille, Institut Pasteur de Lille, Lille, France. 204Institute of Psychiatry and Psychotherapy, Charité-Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität Zu Berlin, and Berlin Institute of Health, Berlin, Germany. 205German Center for Neurodegenerative Diseases (DZNE), Berlin, Germany. 206Department of Biomedical Sciences, University of Cagliari, Cagliari, Italy. 207CHUV, Old Age Psychiatry, Department of Psychiatry, Lausanne, Switzerland. 208Old Age Psychiatry, Department of Psychiatry, Lausanne University Hospital, Lausanne, Switzerland. 209Department of Geriatric Psychiatry, University Hospital of Psychiatry Zürich, Zürich, Switzerland. 210Department of Neuropsychiatry and Laboratory of Molecular Psychiatry, Charité, Charitéplatz 1, Berlin, Germany. 211Department of Clinical Biochemistry, Rigshospitalet, Copenhagen, Denmark. 212Department of Psychiatry & Neuropsychologie, Maastricht University, Alzheimer Center Limburg, Maastricht, the Netherlands. 213 Depatamento de Especialidades Quirúrgicas Bioquímica e Inmunología, Facultad de Medicina, Universidad de Málaga, Málaga, Spain. 214Delft Bioinformatics Lab, Delft University of Technology, Delft, the Netherlands. 215Taub Institute, Columbia University, New York, NY, USA. 216Bioinformatics, College of Life Sciences, Brigham Young University, Provo, UT, USA. 217Institute of Social Medicine, Occupational Health and Public Health, University of Leipzig, Leipzig, Germany. 218Center of Mental Health, Clinic and Policlinic of Psychiatry, Psychosomatics and Psychotherapy, University Hospital of Würzburg, Wuerzburg, Germany. 219Department of Research and Innovation, Helse Fonna, Haugesund Hospital, Haugesund, Norway. 220Institute of Clinical Medicine (K1), The University of Bergen, Bergen, Norway. 221Departamento de Especialidades Quirúrgicas, Bioquímicas e Inmunología, School of Medicine, University of Málaga, Málaga, Spain. 222Department of Neuroscience and Mental Health, AOU Città della Salute e della Scienza di Torino, Torino, Italy. 223 Athens Association of Alzheimer’s Disease and Related Disorders, Athens, Greece. 224Department of Geriatrics, St. Olav’s Hospital, Trondheim Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 427 Articles NATure GeneTiCs University Hospital, Trondheim, Norway. 225Department of Immunology, Hospital Universitario Doctor Negrín, Las Palmas de Gran Canaria, Las Palmas, Spain. 226Neurology Department-Hospital Clínic, IDIBAPS, Universitat de Barcelona, Barcelona, Spain. 227First Department of Neurology, Aiginition Hospital, National and Kapodistrian University of Athens, Medical School, Athens, Greece. 228LVR-Hospital Essen, Department of Psychiatry and Psychotherapy, Medical Faculty, University of Duisburg-Essen, Essen, Germany. 229Department of Primary Medical Care, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany. 230Institute of Medical Biometry, Informatics and Epidemiology, University Hospital of Bonn, Bonn, Germany. 231Department of Geriatric Medicine, Oslo University Hospital, Oslo, Norway. 232Laboratory for Advanced Hematological Diagnostics, Department of Hematology and Stem Cell Transplant, Vito Fazzi Hospital, Lecce, Italy. 233Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas, CIBERDEM, Hospital Clínico San Carlos, Madrid, Spain. 234Department of Neurology, University of Bonn, Bonn, Germany. 235Department of Biomedical Sciences, Section of Neuroscience and Clinical Pharmacology, University of Cagliari, Cagliari, Italy. 236Department of Psychiatry, Namsos Hospital, Namsos, Norway. 237Department of Internal Medicine and Biostatistics, Erasmus MC, Rotterdam, the Netherlands. 238Neurodegenerative Brain Diseases Group, VIB Center for Molecular Neurology, VIB, Antwerp, Belgium. 239Department of Neurology, ErasmusMC, Rotterdam, the Netherlands. 240Laboratory for Cognitive Neurology, Department of Neurosciences, University of Leuven, Leuven, Belgium. 241Neurology Department, University Hospitals Leuven, Leuven, Belgium. 242Department of Psychiatry and Psychotherapy, University Medical Center Goettingen, Goettingen, Germany. 243Department of Psychiatry, Harvard Medical School, McLean Hospital, Belmont, MA, USA. 244Department of Neurology and CNR-MAJ, F 76000, Normandy Center for Genomic and Personalized Medicine, Normandie University, UNIROUEN, INSERM U1245, CHU Rouen, Rouen, France. 245German Center for Neurodegenerative Diseases (DZNE), Goettingen, Germany. 246Medical Science Department, iBiMED, Aveiro, Portugal. 247Department of Nutrition and Diatetics, Harokopio University, Athens, Greece. 248Department of Medicine (Biomedical Genetics), Boston University School of Medicine, Boston, MA, USA. 249Neurosciences Area, Instituto Biodonostia, San Sebastian, Spain. 250Department of Health Service, University of Washington, Seattle, WA, USA. 251Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne, Cologne, Germany. 567These authors contributed equally: Céline Bellenguez, Fhari Kuçukali, Iris Jansen, Luca Kleineidam, Sonia Moreno-Grau, Najaf Amin, Adam Naj and Rafael Campos-Martin. 568These authors jointly supervised this work: Mikko Hiltunen, Kristel Sleegers, Gerard Schellenberg, Cornelia van Duijn, Rebecca Sims, Wiesje van der Flier, Agustin Ruiz, Alfredo Ramirez and Jean-Charles Lambert. ✉e-mail: celine.bellenguez@pasteur-lille.fr; Jean-Charles.Lambert@pasteur-lille.fr EADB Jan Laczo179,180, Vaclav Matoska252, Maria Serpente124, Francesca Assogna48, Fabrizio Piras48, Federica Piras48, Valentina Ciullo48, Jacob Shofany48, Carlo Ferrarese38,39, Simona Andreoni38, Gessica Sala38, Chiara Paola Zoia38, Maria Del Zompo235, Alberto Benussi66, Patrizia Bastiani253, Mari Takalo254, Teemu Natunen254, Tiina Laatikainen166,178, Jaakko Tuomilehto166,178, Riitta Antikainen255,256, Timo Strandberg255,257, Jaana Lindström178, Markku Peltonen178, Richard Abraham258, Ammar Al-Chalabi259, Nicholas J. Bass260, Carol Brayne261, Kristelle S. Brown262, John Collinge187, David Craig263, Pangiotis Deloukas264, Nick Fox265, Amy Gerrish265, Michael Gill266, Rhian Gwilliam264, Denise Harold267, Paul Hollingworth258, Jarret A. Johnston268, Lesley Jones258, Brian Lawlor266, Gill Livingston260, Simon Lovestone269, Michelle Lupton270,271, Aoibhinn Lynch266, David Mann272, Bernadette McGuinness268, Andrew McQuillin260, Michael C. O’Donovan258, Michael J. Owen258, Peter Passmore268, John F. Powell270,271, Petra Proitsi270,271, Martin Rossor265, Christopher E. Shaw259, A. David Smith273, Hugh Gurling274, Stephen Todd275, Catherine Mummery276, Nathalie Ryan276, Giordano Lacidogna98, Ad Adarmes-Gómez11,77, Ana Mauleón10, Ana Pancho10, Anna Gailhajenet10, Asunción Lafuente10, D. Macias-García11,77, Elvira Martín10, Esther Pelejà10, F. Carrillo11,77, Isabel Sastre Merlín11,134, L. Garrote-Espina11,77, Liliana Vargas10, M. Carrion-Claro11,77, M. Marín96, Ma Labrador11,77, Mar Buendia10, María Dolores Alonso277, Marina Guitart10, Mariona Moreno10, Marta Ibarria10, Mt Periñán11,77, Nuria Aguilera10, P. Gómez-Garre11,77, Pilar Cañabate10, R. Escuela11,77, R. Pineda-Sánchez1,77, R. Vigo-Ortega11,77, S. Jesús11,77, Silvia Preckler10, Silvia Rodrigo-Herrero132, Susana Diego10, Alessandro Vacca67, Fausto Roveta67, Nicola Salvadori112, Elena Chipi112, Henning Boecker9,278, Christoph Laske279,280, Robert Perneczky81,281, Costas Anastasiou247, Daniel Janowitz78, Rainer Malik78, Anna Anastasiou35, Kayenat Parveen7, Carmen Lage282, Sara López-García282, Anna Antonell226, Kalina Yonkova Mihova283, Diyana Belezhanska52, Heike Weber284, Silvia Kochen285, Patricia Solis285, Nancy Medel285, Julieta Lisso285, Zulma Sevillano285, Daniel G. Politis285,286, Valeria Cores285,286, Carolina Cuesta285,286, Cecilia Ortiz287, Juan Ignacio Bacha287, Mario Rios288, Aldo Saenz288, Mariana Sanchez Abalos289, Eduardo Kohler290, Dana Lis Palacio291, Ignacio Etchepareborda291, Matias Kohler291, Gisela Novack292, Federico Ariel Prestia292, Pablo Galeano292, Eduardo M. Castaño292, Sandra Germani293, 428 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Carlos Reyes Toso293, Matias Rojo293, Carlos Ingino293, Carlos Mangone293, David C. Rubinsztein294, Stefan Teipel295, Nathalie Fievet1, Vincent Deramerourt200, Charlotte Forsell143,144, Håkan Thonberg143,144, Maria Bjerke89, Ellen De Roeck89, María Teresa Martínez-Larrad296 and Natividad Olivar293 252 Department of Clinical Biochemistry, Hematology and Immunology, Na Homolce Hospital, Prague, Czechia. 253Institute of Gerontology and Geriatrics, Department of Medicine, University of Perugia, Perugia, Italy. 254Insitute of Biomedicine, University of Eastern Finland, Kuopio, Finland. 255Center for Life Course Health Research, University of Oulu, Oulu, Finland. 256Medical Research Center Oulu, Oulu University Hospital, Oulu, Finland. 257University of Helsinki and Helsinki University Hospital, Helsinki, Finland. 258Division of Psychological Medicine and Clinial Neurosciences, MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff, UK. 259Institute of Psychiatry, Psychology and Neuroscience, Kings College London, London, UK. 260Division of Psychiatry, University College London, London, UK. 261Institute of Public Health, University of Cambridge, Cambridge, UK. 262 Institute of Genetics, Queens Medical Centre, University of Nottingham, Nottingham, UK. 263Ageing Group, Centre for Public Health, School of Medicine, Dentistry and Biomedical Sciences, Queen’s University Belfast, Belfast, UK. 264The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, UK. 265Dementia Research Centre, Department of Neurodegenerative Disease, UCL Institute of Neurology, London, UK. 266Mercer’s Institute for Research on Ageing, St James’ Hospital, Dublin, Ireland. 267School of Biotechnology, Dublin City University, Dublin, Ireland. 268Centre for Public Health, School of Medicine, Dentistry and Biomedical Sciences, Queens University, Belfast, UK. 269Department of Psychiatry, University of Oxford, Oxford, UK. 270Genetic Epidemiology, QIMR Berghofer Medical Research Institute, Herston, Queensland, Australia. 271Department of Basic and Clinical Neuroscience, Institute of Psychiatry, Psychology and Neuroscience, Kings College London, London, UK. 272Division of Neuroscience and Experimental Psychology, School of Biological Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK. 273Oxford Project to Investigate Memory and Ageing (OPTIMA), University of Oxford, Level 4, John Radcliffe Hospital, Oxford, UK. 274Department of Mental Health Sciences, University College London, London, UK. 275Ageing Group, Centre for Public Health, School of Medicine, Dentistry and Biomedical Sciences, Queen’s University Belfast, Belfast, UK. 276Dementia Research Centre, UCL, London, UK. 277Servei de Neurologia Hospital Clínic, Universitari de València, Valencia, Spain. 278Department of Radiology, University Hospital Bonn, Bonn, Germany. 279German Center for Neurodegenerative Diseases (DZNE), Tübingen, Germany. 280Section for Dementia Research, Department of Psychiatry, Hertie Institute for Clinical Brain Research, Tübingen, Germany. 281Department of Psychiatry and Psychotherapy, University Hospital, LMU Munich, Munich, Germany. 282Service of Neurology, University Hospital Marqués de Valdecilla, IDIVAL, University of Cantabria, Santander, Spain. 283Molecular Medicine Center, Department of Medical chemistry and biochemistry, Medical University of Sofia, Sofia, Bulgaria. 284Department of Psychiatry, Psychosomatics and Psychotherapy, Center of Mental Health, University Hospital of Würzburg, Würzburg, Germany. 285ENYS (Estudio en Neurociencias y Sistemas Complejos) CONICET- Hospital El Cruce “Nestor Kirchner”- UNAJ, Buenos Aires, Argentina. 286HIGA Eva Perón, Buenes Aires, Argentina. 287Neurología Clinica, Buenes Aires, Argentina. 288 Dirección de Atención de Adultos Mayores del Min. Salud Desarrollo Social y Deportes de la Pcia. de Mendoza, Mendoza, Argentina. 289Laboratorio de Genética Forense del Ministerio Público de la Pcia de La Pampa, La Pampa, Argentina. 290Fundacion Sinapsis, Santa Rosa, Argentina. 291Hospital Dr. Lucio Molas, Santa Rosa; Fundacion Ayuda Enfermo Renal y Alta Complejidad (FERNAC), Santa Rosa, Argentina. 292Laboratory of Brain Aging and Neurodegeneration (FIL), Buneos Aires, Argentina. 293Centro de Neuropsiquiatría y Neurología de la Conducta (CENECON), Facultad de Medicina, Universidad de Buenos Aires (UBA), C.A.B.A., Buenos Aires, Argentina. 294Cambridge Institute for Medical Research and UK Dementia Research Institute, University of Cambridge, Cambridge, UK. 295German Center for Neurodegenerative Diseases (DZNE), Rostock, Germany. 296Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas Asociadas, CIBERDEM, Hospital Clínico San Carlos, Madrid, Spain. GR@ACE Nuria Aguilera10,11, Mar Buendia10, Amanda Cano10, Pilar Cañabate10,11, Susana Diego10, Anna Gailhajenet10, Marina Guitart10, Marta Ibarria10, Asunción Lafuente10, Juan Macias97, Olalla Maroñas297, Elvira Martín10, Mariona Moreno10, Raúl Nuñez-Llaves10, Clàudia Olivé10, Ana Pancho10, Ester Pelejá10, Silvia Preckler10 and Liliana Vargas10 Grupo de Medicina Xen´omica, Centro Nacional de Genotipado (CEGEN-PRB3-ISCIII), Universidad de Santiago de Compostela, Santiago de 297 Compostela, Spain. DEGESCO Astrid D. Adarmes-Gómez11,188, María Dolores Alonso298, Guillermo Amer-Ferrer299, Martirio Antequera37, Juan Andrés Burguera49, Fátima Carrillo11,188, Mario Carrión-Claro11,188, María José Casajeros139, Marian Martinez de Pancorbo300, Rocío Escuela11,188, Lorena Garrote-Espina11,188, Pilar Gómez-Garre11,188, Saray Hevilla138, Silvia Jesús11,188, Miguel Angel Labrador Espinosa11,188, Agustina Legaz37, Sara López-García11,174, Daniel Macias-García11,188, Salvadora Manzanares299,301, Marta Marín132, Juan Marín-Muñoz37, Tamara Marín138, Begoña Martínez37, Victoriana Martínez37, Pablo Martínez-Lage Álvarez302, Maite Mendioroz Iriarte303, María Teresa Periñán-Tocino11,188, Rocío Pineda-Sánchez11,188, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 429 Articles NATure GeneTiCs Diego Real de Asúa304, Silvia Rodrigo132, Isabel Sastre11,134, Maria Pilar Vicente37, Rosario Vigo-Ortega11,188 and Liliana Vivancos37 Servei de Neurologia, Hospital Clínic Universitari de València, Valencia, Spain. 299Department of Neurology, Hospital Universitario Son Espases, Palma, 298 Spain. 300BIOMICs, País Vasco; Centro de Investigación Lascaray, Universidad del País Vasco UPV/EHU, Vitoria-Gasteiz, Spain. 301Fundación para la Formación e Investigación Sanitarias de la Región de Murcia, Palma, Spain. 302Centro de Investigación y Terapias Avanzadas, Fundación CITA-Alzheimer, San Sebastian, Spain. 303Navarrabiomed, Pamplona, Spain. 304Hospital Universitario La Princesa, Madrid, Spain. EADI Jacques Epelbaum305, Didier Hannequin244, Dominique campion92, Vincent Deramecourt200, Nathalie Fievet1, Christophe Tzourio101, Alexis Brice306 and Bruno Dubois307 UMR 7179 CNRS/MNHN, Brunoy, France. 306Sorbonne Université, Paris Brain Institute, APHP, INSERM, CNRS, Paris, France. 307Department of Neurology, 305 Institute of Memory and Alzheimer’s Disease (IM2A), Pitié-Salpêtrière Hospital, AP-HP, Boulevard de l’Hôpital, Paris, France. GERAD Denise Harold267, Paul Hollingworth258, Amy Gerrish265, Amy Williams16, Charlene Thomas16, Chloe Davies16, William Nash16, Kimberley Dowzell16, Atahualpa Castillo Morales16,108, Mateus Bernardo-Harrington16,108, James Turton308, Jenny Lord308, Kristelle Brown262, Emma Vardy309, Elizabeth Fisher310, Jason D. Warren310, Martin Rossor29, Natalie S. Ryan276, Rita Guerreiro310, James Uphill187, John Collinge187, Michelle Lupton270,271, Ammar Al-Chalabi259, Christopher E. Shaw259, Nick Bass260, Richard Abraham258, Reinhard Heun311, Heike Kölsch312, Britta Schürmann312, André Lacour9, Christine Herold9, Simon Lovestone269, Bernadette McGuinness268, David Craig263, Janet A. Johnston268, Michael Gill266, Peter Passmore263, Stephen Todd275, John Powell270,271, Petra Proitsi270,271, Yogen Patel313, Angela Hodges311, Tim Becker9,314, A. David Smith273, Donald Warden273, Gordon Wilcock273, Robert Clarke310, Aoibhinn Lynch266, Brian Lawlor266, Andrew McQuillin260, Gill Livingston260, David C. Rubinsztein294, Carol Brayne261, Rhian Gwilliam264, Panagiotis Deloukas264, Yoav Ben-Shlomo315, David Mann272, Nigel M. Hooper272, Stuart Pickering-Brown272, Rebecca Sussams316, Nick Warner317, Anthony Bayer318, Isabella Heuser319, Dmitriy Drichel320, Norman Klopp321, Manuel Mayhaus322, Matthias Riemenschneider322, Sabrina Pinchler322, Thomas Feulner322, Wei Gu322, Hendrik van den Bussche229, Michael Hüll323, Lutz Frölich324, H-Erich Wichmann321, Karl-Heinz Jöckel325, Michael O’Donovan258, Lesley Jones258 and Michael Owen258 308 Institute of Genetics, Queen’s Medical Centre, University of Nottingham, Nottingham, UK. 309Institute for Ageing and Health, Newcastle University, Campus for Ageing and Vitality, Newcastle upon Tyne, UK. 310Department of Neurodegenerative Disease, UCL Institute of Neurology, London, UK. 311 Department of Old Age Psychiatry, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK. 312Department of Psychiatry and Psychotherapy, University of Bonn, Bonn, Germany. 313Department of Basic and Clinical Neuroscience, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, London, UK. 314Institute for Medical Biometry, Informatics and Epidemiology, University of Bonn, Bonn, Germany. 315Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK. 316Division of Clinical Neurosciences, School of Medicine, University of Southampton, Southampton, UK. 317Somerset Partnership NHS Trust, Somerset, UK. 318Institute of Primary Care and Public Health, Cardiff University, University Hospital of Wales, Cardiff, UK. 319Department of Psychiatry and Psychotherapy, Charité University Medicine, Berlin, Germany. 320 Cologne Center for Genomics, University of Cologne, Cologne, Germany. 321Institute of Epidemiology, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Munich, Germany. 322Department of Psychiatry and Psychotherapy, University Hospital, Saarland, Germany. 323Department of Psychiatry, University of Freiburg, Freiburg, Germany. 324Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Heidelberg, Germany. 325Institute for Medical Informatics, Biometry and Epidemiology, University Hospital of Essen, University Duisburg-Essen, Essen, Germany. Demgene Shahram Bahrami46,326, Ingunn Bosnes327,328, Per Selnes329 and Sverre Bergh330 430 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Division of Mental Health and Addiction, Oslo University Hospital, Oslo, Norway. 327Department of Mental Health, Faculty of Medicine and Health 326 Sciences, Norwegian University of Science and Technology, Trondheim, Norway. 328Department of Psychiatry, Hospital Namsos, Nord-Trøndelag Health Trust, Namsos, Norway. 329Department of Neurology, Akershus University Hospital, Lørenskog, Norway. 330Centre for Old Age Psychiatry Research, Innlandet Hospital Trust, Ottestad, Norway. FinnGen Aarno Palotie331, Mark Daly331, Howard Jacob332, Athena Matakidou333, Heiko Runz334, Sally John334, Robert Plenge335, Mark McCarthy336, Julie Hunkapiller336, Meg Ehm337, Dawn Waterworth337, Caroline Fox338, Anders Malarstig339, Kathy Klinger340, Kathy Call340, Tim Behrens341, Patrick Loerch342, Tomi Mäkelä343, Jaakko Kaprio331, Petri Virolainen344, Kari Pulkki344, Terhi Kilpi345, Markus Perola345, Jukka Partanen346, Anne Pitkäranta347, Riitta Kaarteenaho348, Seppo Vainio348, Miia Turpeinen348, Raisa Serpi348, Tarja Laitinen349, Johanna Mäkelä350, Veli-Matti Kosma351, Urho Kujala352, Outi Tuovila353, Minna Hendolin353, Raimo Pakkanen353, Jeff Waring332, Bridget Riley-Gillis332, Jimmy Liu334, Shameek Biswas335, Dorothee Diogo338, Catherine Marshall339, Xinli Hu339, Matthias Gossel340, Robert Graham341, Beryl Cummings342, Samuli Ripatti331, Johanna Schleutker344, Mikko Arvas346, Olli Carpén347, Reetta Hinttala348, Johannes Kettunen348, Arto Mannermaa351, Jari Laukkanen352, Valtteri Julkunen354, Anne Remes354, Reetta Kälviäinen354, Jukka Peltola355, Pentti Tienari356, Juha Rinne357, Adam Ziemann332, Jeffrey Waring332, Sahar Esmaeeli332, Nizar Smaoui332, Anne Lehtonen332, Susan Eaton334, Sanni Lahdenperä334, Janet van Adelsberg335, John Michon336, Geoff Kerchner336, Natalie Bowers336, Edmond Teng336, John Eicher336, Vinay Mehta338, Padhraig Gormley338, Kari Linden339, Christopher Whelan339, Fanli Xu337, David Pulford337, Martti Färkkilä356, Sampsa Pikkarainen356, Airi Jussila358, Timo Blomster359, Mikko Kiviniemi360, Markku Voutilainen357, Bob Georgantas332, Graham Heap332, Fedik Rahimov332, Keith Usiskin335, Tim Lu6, Danny Oh336, Kirsi Kalpala339, Melissa Miller339, Linda McCarthy337, Kari Eklund356, Antti Palomäki357, Pia Isomäki358, Laura Pirilä357, Oili Kaipiainen-Seppänen360, Johanna Huhtakangas359, Apinya Lertratanakul332, Marla Hochfeld335, Nan Bing339, Jorge Esparza Gordillo337, Nina Mars331, Margit Pelkonen360, Paula Kauppi356, Hannu Kankaanranta355, Terttu Harju359, David Close333, Steven Greenberg335, Hubert Chen336, Jo Betts337, Soumitra Ghosh337, Veikko Salomaa361, Teemu Niiranen361, Markus Juonala357, Kaj Metsärinne357, Mika Kähönen358, Juhani Junttila359, Markku Laakso354, Jussi Pihlajamäki354, Juha Sinisalo356, Marja-Riitta Taskinen356, Tiinamaija Tuomi356, Ben Challis333, Andrew Peterson336, Audrey Chu338, Jaakko Parkkinen339, Anthony Muslin340, Heikki Joensuu356, Tuomo Meretoja356, Lauri Aaltonen356, Johanna Mattson356, Annika Auranen355, Peeter Karihtala359, Saila Kauppila359, Päivi Auvinen359, Klaus Elenius357, Relja Popovic332, Jennifer Schutzman336, Andrey Loboda338, Aparna Chhibber338, Heli Lehtonen339, Stefan McDonough339, Marika Crohns340, Diptee Kulkarni337, Kai Kaarniranta354, Joni A. Turunen356, Terhi Ollila356, Sanna Seitsonen356, Hannu Uusitalo355, Vesa Aaltonen357, Hannele Uusitalo-Järvinen355, Marja Luodonpää359, Nina Hautala359, Stephanie Loomis334, Erich Strauss336, Hao Chen336, Anna Podgornaia338, Joshua Hoffman337, Kaisa Tasanen359, Laura Huilaja359, Katariina Hannula-Jouppi356, Teea Salmi358, Sirkku Peltonen356, Leena Koulu356, Ilkka Harvima354, Ying Wu9, David Choy336, Pirkko Pussinen356, Aino Salminen356, Tuula Salo356, David Rice356, Pekka Nieminen356, Ulla Palotie356, Maria Siponen354, Liisa Suominen354, Päivi Mäntylä354, Ulvi Gursoy357, Vuokko Anttonen359, Kirsi Sipilä359, Justin Wade Davis332, Danjuma Quarless332, Slavé Petrovski333, Eleonor Wigmore333, Chia-Yen Chen334, Paola Bronson334, Ellen Tsai334, Yunfeng Huang334, Joseph Maranville335, Elmutaz Shaikho335, Elhaj Mohammed335, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 431 Articles NATure GeneTiCs Samir Wadhawan362, Erika Kvikstad362, Minal Caliskan362, Diana Chang336, Tushar Bhangale336, Sarah Pendergrass336, Emily Holzinger338, Xing Chen339, Åsa Hedman339, Karen S. King337, Clarence Wang340, Ethan Xu340, Franck Auge340, Clement Chatelain340, Deepak Rajpal340, Dongyu Liu340, Katherine Call340, Tai-he Xia340, Matt Brauer341, Mitja Kurki331, Juha Karjalainen331, Aki Havulinna331, Anu Jalanko331, Priit Palta331, Pietro della Briotta Parolo331, Wei Zhou363, Susanna Lemmelä331, Manuel Rivas364, Jarmo Harju331, Arto Lehisto331, Andrea Ganna331, Vincent Llorens331, Hannele Laivuori331, Sina Rüeger331, Mari E. Niemi331, Taru Tukiainen331, Mary Pat Reeve331, Henrike Heyne331, Nina Mars331, Kimmo Palin365, Javier Garcia-Tabuenca366, Harri Siirtola366, Tuomo Kiiskinen331, Jiwoo Lee331, Kristin Tsuo331, Amanda Elliott331, Kati Kristiansson345, Kati Hyvärinen367, Jarmo Ritari367, Miika Koskinen347, Katri Pylkäs348, Marita Kalaoja348, Minna Karjalainen348, Tuomo Mantere348, Eeva Kangasniemi350, Sami Heikkinen351, Eija Laakkonen352, Csilla Sipeky368, Samuel Heron368, Antti Karlsson344, Dhanaprakash Jambulingam368, Venkat Subramaniam Rathinakannan368, Risto Kajanne331, Mervi Aavikko331, Manuel González Jiménez331, Pietro della Briotta Parola331, Arto Lehistö331, Masahiro Kanai363, Mari Kaunisto331, Elina Kilpeläinen331, Timo P. Sipilä331, Georg Brein331, Ghazal Awaisa331, Anastasia Shcherban331, Kati Donner331, Anu Loukola347, Päivi Laiho345, Tuuli Sistonen345, Essi Kaiharju345, Markku Laukkanen345, Elina Järvensivu345, Sini Lähteenmäki345, Lotta Männikkö345, Regis Wong345, Hannele Mattsson345, Tero Hiekkalinna345, Teemu Paajanen345, Kalle Pärn331 and Javier Gracia-Tabuenca366 331 Institute for Molecular Medicine Finland, HiLIFE, University of Helsinki, Helsinki, Finland. 332AbbVie, Chicago, IL, USA. 333Astra Zeneca, Cambridge, UK. 334Biogen, Cambridge, MA, USA. 335Celgene, Summit, NJ, USA. 336Genentech, San Francisco, CA, USA. 337GlaxoSmithKline, Brentford, UK. 338Merck, Kenilworth, NJ, USA. 339Pfizer, New York, NY, USA. 340Sanofi, Paris, France. 341Maze Therapeutics, San Francisco, CA, USA. 342Janssen Biotech, Beerse, Belgium. 343HiLIFE, University of Helsinki, Helsinki, Finland. 344Auria Biobank, University of Turku, Hospital District of Southwest Finland, Turku, Finland. 345 THL Biobank, The National Institute of Health and Welfare Helsinki, Helsinki, Finland. 346Finnish Red Cross Blood Service, Finnish Hematology Registry and Clinical Biobank, Helsinki, Finland. 347Helsinki Biobank, Helsinki University and Hospital District of Helsinki and Uusimaa, Helsinki, Finland. 348Northern Finland Biobank Borealis, University of Oulu, Northern Ostrobothnia Hospital District, Oulu, Finland. 349Oxford Healthy Aging Project, Clinical Trial Service Unit, University of Oxford, Oxford, UK. 350Finnish Clinical Biobank Tampere, University of Tampere, Pirkanmaa Hospital District, Tampere, Finland. 351 Biobank of Eastern Finland, University of Eastern Finland / Northern Savo Hospital District, Kuopio, Finland. 352Central Finland Biobank, University of Jyväskylä, Central Finland Health Care District, Jyväskylä, Finland. 353Business Finland, Helsinki, Finland. 354Northern Savo Hospital District, Kuopio, Finland. 355 Pirkanmaa Hospital District, Tampere, Finland. 356Hospital District of Helsinki and Uusimaa, Helsinki, Finland. 357Hospital District of Southwest Finland, Turku, Finland. 358Pirkanmaa Hospital District, Tampere, Finland. 359Northern Ostrobothnia Hospital District, Oulu, Finland. 360Northern Savo Hospital District, Kuopio, Finland. 361The National Institute of Health and Welfare Helsinki, Helsinki, Finland. 362Bristol Myers Squibb, New York, NY, USA. 363Broad Institute, Cambridge, MA, USA. 364University of Stanford, Stanford, CA, USA. 365University of Helsinki, Helsinki, Finland. 366University of Tampere, Tampere, Finland. 367Finnish Red Cross Blood Service, Helsinki, Finland. 368University of Turku, Turku, Finland. ADGC Erin Abner369, Perrie M. Adams370, Alyssa Aguirre371, Marilyn S. Albert372, Roger L. Albin373,374,375, Mariet Allen376, Lisa Alvarez377, Liana G. Apostolova378,379, Steven E. Arnold380, Sanjay Asthana381,382,383, Craig S. Atwood381,382,383, Gayle Ayres371, Clinton T. Baldwin248, Robert C. Barber377, Lisa L. Barnes384,385,386, Sandra Barral50,183,186, Thomas G. Beach387, James T. Becker388, Gary W. Beecham154, Duane Beekly389, Jennifer E. Below93,390, Penelope Benchek57, Bruno A. Benitez391, David Bennett384,386, John Bertelson392, Flanagan E. Margaret393,394, Thomas D. Bird181,395, Deborah Blacker396,397, Bradley F. Boeve398, James D. Bowen399, Adam Boxer400, James Brewer401, James R. Burke402, Jeffrey M. Burns403, Will S. Bush57, Joseph D. Buxbaum103,104,106,160, Nigel J. Cairns404, Chuanhai Cao405, Christopher S. Carlson406, Cynthia M. Carlsson382,383, Regina M. Carney407, Minerva M. Carrasquillo376, Scott Chasse408, Marie-Francoise Chesselet409, Hung-Hsin Chen93, Alessandra Chesi14, Nathaniel A. Chin381,382, Helena C. Chui410, Jaeyoon Chung248, Suzanne Craft411, 432 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles Paul K. Crane196, David H. Cribbs412, Elizabeth A. Crocco413, Carlos Cruchaga414,415, Michael L. Cuccaro88,154, Munro Cullum370, Eveleen Darby416, Barbara Davis417, Philip L. De Jager418, Charles DeCarli419, John DeToledo420, Malcolm Dick421, Dennis W. Dickson376, Beth A. Dombroski14, Rachelle S. Doody416, Ranjan Duara422, Nilüfer Ertekin-Taner376,423, Denis A. Evans424, Kelley M. Faber119, Thomas J. Fairchild425, Kenneth B. Fallon426, David W. Fardo122, Martin R. Farlow427, John J. Farrell248, Victoria Fernandez-Hernandez391, Steven Ferris428, Tatiana M. Foroud119, Matthew P. Frosch429, Brian Fulton-Howard430, Douglas R. Galasko401, Adriana Gamboa431,432, Marla Gearing433,434, Daniel H. Geschwind414, Bernardino Ghetti435, John R. Gilbert88,154, Thomas J. Grabowski181,436, Neill R. Graff-Radford377,423, Struan F. A. Grant437,438,439, Robert C. Green440, John H. Growdon441, Jonathan L. Haines56,57, Hakon Hakonarson442,443, James Hall377, Ronald L. Hamilton444, Kara L. Hamilton-Nelson154, Oscar Harari415, Lindy E. Harrell445, Jacob Haut14, Elizabeth Head446, Victor W. Henderson447,448, Michelle Hernandez420, Timothy Hohman93,449, Lawrence S. Honig50, Ryan M. Huebinger450, Matthew J. Huentelman451, Christine M. Hulette452, Bradley T. Hyman441, Linda S. Hynan370,453,454, Laura Ibanez415,455, Gail P. Jarvik196,456, Suman Jayadev181, Lee-Way Jin457, Kim Johnson420, Leigh Johnson431, M. Ilyas Kamboh388,458,459, Anna M. Karydas400, Mindy J. Katz460, Jeffrey A. Kaye461,462, C. Dirk Keene463, Aisha Khaleeq416, Ronald Kim446, Janice Knebl431, Neil W. Kowall20,464, Joel H. Kramer465, Pavel P. Kuksa14, Frank M. LaFerla466, James J. Lah467, Eric B. Larson468, Chien-Yueh Lee14, Edward B. Lee14, Alan Lerner57, Yuk Yee Leung14, James B. Leverenz469, Allan I. Levey467, Mingyao Li14, Andrew P. Lieberman470, Richard B. Lipton460, Mark Logue248,397,471, Constantine G. Lyketsos472, John Malamon14, Douglas Mains431,432, Daniel C. Marson445, Frank Martiniuk473, Deborah C. Mash474, Eliezer Masliah401,475, Paul Massman416, Arjun Masurkar428, Wayne C. McCormick196, Susan M. McCurry476, Andrew N. McDavid406, Stefan McDonough477, Ann C. McKee20,464, Marsel Mesulam393,394, Jesse Mez20, Bruce L. Miller478, Carol A. Miller479, Joshua W. Miller457, Thomas J. Montine480, Edwin S. Monuki446, John C. Morris404,415,445,481, Amanda J. Myers413, Trung Nguyen453, Sid O’Bryant482, John M. Olichney483, Marcia Ory484, Raymond Palmer485, Joseph E. Parisi486, Henry L. Paulson373,375, Valory Pavlik416, David Paydarfar371, Victoria Perez420, Elaine Peskind487, Ronald C. Petersen398, Jennifer E. Phillips-Cremins439,488, Aimee Pierce412, Marsha Polk489, Wayne W. Poon421, Huntington Potter490, Liming Qu14, Mary Quiceno491,492, Joseph F. Quinn461,462, Ashok Raj405, Murray Raskind487, Eric M. Reiman451,493,494,495, Barry Reisberg428,489, Joan S. Reisch417, John M. Ringman410, Erik D. Roberson445, Monica Rodriguear416, Ekaterina Rogaeva496, Howard J. Rosen400, Roger N. Rosenberg453, Donald R. Royall485, Mark A. Sager382, Mary Sano106, Andrew J. Saykin119,497, Julie A. Schneider384,386,498, Lon S. Schneider410,499, William W. Seeley400, Jin Sha14, Susan H. Slifer154, Scott Small50,186, Amanda G. Smith405, Janet P. Smith417, Yeunjoo E. Song56,57, Joshua A. Sonnen463, Salvatore Spina435, Peter St George-Hyslop500,501, Robert A. Stern20, Alan B. Stevens484,502,503, Stephen M. Strittmatter504, David Sultzer505, Russell H. Swerdlow403, Rudolph E. Tanzi441, Jeffrey L. Tilson506, John Q. Trojanowski14, Juan C. Troncoso507, Debby W. Tsuang395,487, Otto Valladares14, Vivianna M. Van Deerlin14, Linda J. van Eldik122, Robert Vassar393,394, Harry V. Vinters410,508, Jean-Paul Vonsattel50, Sandra Weintraub509, Kathleen A. Welsh-Bohmer402,510, Patrice L. Whitehead154, Ellen M. Wijsman196,456,511, Kirk C. Wilhelmsen408, Benjamin Williams411, Jennifer Williamson50, Henrik Wilms420, Thomas S. Wingo467, Thomas Wisniewski512,513, Randall L. Woltjer514, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 433 Articles NATure GeneTiCs Martin Woon392, Clinton B. Wright515, Chuang-Kuo Wu420, Steven G. Younkin376,423, Chang-En Yu196, Lei Yu384,386, Yuanchao Zhang442, Yi Zhao48 and Xiongwei Zhu516 369 Sanders-Brown Center on Aging, Department of Epidemiology, College of Public Health, University of Kentucky, Lexington, KY, USA. 370Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, TX, USA. 371Department of Neurology, Dell Medical School, University of Texas at Austin, Austin, TX, USA. 372Department of Neurology, Johns Hopkins University, Baltimore, MD, USA. 373Department of Neurology, University of Michigan, Ann Arbor, MI, USA. 374Geriatric Research, Education and Clinical Center (GRECC), VA Ann Arbor Healthcare System (VAAAHS), Ann Arbor, MI, USA. 375 Michigan Alzheimer’s Disease Center, University of Michigan, Ann Arbor, MI, USA. 376Department of Neuroscience, Mayo Clinic, Jacksonville, FL, USA. 377 Department of Pharmacology and Neuroscience, University of North Texas Health Science Center, Fort Worth, TX, USA. 378Departments of Neurology, Radiology, and Medical and Molecular Genetics, Indiana University School of Medicine, Indianapolis, IN, USA. 379Indiana Alzheimer’s Disease Research Center, Indiana University School of Medicine, Indianapolis, IN, USA. 380Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 381Geriatric Research, Education and Clinical Center (GRECC), University of Wisconsin, Madison, WI, USA. 382 Department of Medicine, University of Wisconsin, Madison, WI, USA. 383Wisconsin Alzheimer’s Disease Research Center, Madison, WI, USA. 384 Department of Neurological Sciences, Rush University Medical Center, Chicago, IL, USA. 385Department of Behavioral Sciences, Rush University Medical Center, Chicago, IL, USA. 386Rush Alzheimer’s Disease Center, Rush University Medical Center, Chicago, IL, USA. 387Civin Laboratory for Neuropathology, Banner Sun Health Research Institute, Phoenix, AZ, USA. 388Departments of Psychiatry, Neurology, and Psychology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA. 389National Alzheimer’s Coordinating Center, University of Washington, Seattle, WA, USA. 390Human Genetics Center, Department of Epidemiology, Human Genetics, and Environmental Sciences, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA. 391Department of Psychiatry and Hope Center Program on Protein Aggregation and Neurodegeneration, Washington University School of Medicine, St. Louis, MO, USA. 392Department of Psychiatry, University of Texas at Austin/Dell Medical School, Austin, TX, USA. 393 Department of Pathology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 394Cognitive Neurology and Alzheimer’s Disease Center, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 395VA Puget Sound Health Care System/GRECC, Seattle, WA, USA. 396 Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA. 397Department of Psychiatry, Massachusetts General Hospital/ Harvard Medical School, Boston, MA, USA. 398Department of Neurology, Mayo Clinic, Rochester, MN, USA. 399Swedish Medical Center, Seattle, WA, USA. 400 Department of Neurology, University of California San Francisco, San Francisco, CA, USA. 401Department of Neurosciences, University of California San Diego, La Jolla, CA, USA. 402Department of Medicine, Duke University, Durham, NC, USA. 403University of Kansas Alzheimer’s Disease Center, University of Kansas Medical Center, Kansas City, KS, USA. 404Department of Pathology and Immunology, Washington University, St. Louis, MO, USA. 405USF Health Byrd Alzheimer’s Institute, University of South Florida, Tampa, FL, USA. 406Fred Hutchinson Cancer Research Center, Seattle, WA, USA. 407Mental Health and Behavioral Science Service, Bruce W. Carter VA Medical Center, Miami, FL, USA. 408Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. 409Neurogenetics Program, University of California, Los Angeles, Los Angeles, CA, USA. 410Department of Neurology, University of Southern California, Los Angeles, CA, USA. 411Section of Gerontology and Geriatric Medicine Research, Wake Forest School of Medicine, Winston-Salem, NC, USA. 412Department of Neurology, University of California, Irvine, Irvine, CA, USA. 413Department of Psychiatry and Behavioral Sciences, Miller School of Medicine, University of Miami, Miami, FL, USA. 414NeuroGenomics and Informatics, Washington University in St. Louis, St. Louis, MO, USA. 415Department of Psychiatry, Washington University in St. Louis, St Louis, MO, USA. 416Alzheimer’s Disease and Memory Disorders Center, Baylor College of Medicine, Houston, TX, USA. 417Department of Population and Data Sciences, University of Texas Southwestern Medical Center, Dallas, Texas, USA. 418Center for Translational and Computational Neuroimmunology, Department of Neurology, Columbia University Medical Center, New York, NY, USA. 419Department of Neurology, University of California, Davis, Sacramento, CA, USA. 420Departments of Neurology, Pharmacology and Neuroscience, Texas Tech University Health Science Center, Lubbock, TX, USA. 421Institute for Memory Impairments and Neurological Disorders, University of California, Irvine, Irvine, CA, USA. 422Wien Center for Alzheimer’s Disease and Memory Disorders, Mount Sinai Medical Center, Miami Beach, FL, USA. 423Department of Neurology, Mayo Clinic, Jacksonville, FL, USA. 424Rush Institute for Healthy Aging, Department of Internal Medicine, Rush University Medical Center, Chicago, IL, USA. 425Office of Strategy and Measurement, University of North Texas Health Science Center, Fort Worth, TX, USA. 426Department of Pathology, University of Alabama at Birmingham, Birmingham, AL, USA. 427Department of Neurology, Indiana University, Indianapolis, IN, USA. 428Department of Psychiatry, New York University, New York, NY, USA. 429C.S. Kubik Laboratory for Neuropathology, Massachusetts General Hospital, Charlestown, MA, USA. 430Department of Neuroscience, Ronald M. Loeb Center for Alzheimer’s Disease, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 431Department of Health Behavior and Health Systems, University of North Texas Health Science Center, Fort Worth, TX, USA. 432Department of Health Management and Policy, School of Public Health, University of North Texas Health Science Center, Fort Worth, TX, USA. 433 Department of Pathology and Laboratory Medicine, Emory University, Atlanta, GA, USA. 434Emory Alzheimer’s Disease Center, Emory University, Atlanta, GA, USA. 435Department of Pathology and Laboratory Medicine, Indiana University, Indianapolis, IN, USA. 436Department of Radiology, University of Washington, Seattle, WA, USA. 437Center for Spatial and Functional Genomics, Division of Human Genetics, Children’s Hospital of Philadelphia, Philadelphia, PA, USA. 438Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 439Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 440Division of Genetics, Department of Medicine and Partners Center for Personalized Genetic Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA, USA. 441Department of Neurology, Massachusetts General Hospital/Harvard Medical School, Boston, MA, USA. 442Center for Applied Genomics, Children’s Hospital of Philadelphia, Philadelphia, PA, USA. 443Division of Human Genetics, Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA. 444Department of Pathology (Neuropathology), University of Pittsburgh, Pittsburgh, PA, USA. 445Department of Neurology, University of Alabama at Birmingham, Birmingham, AL, USA. 446Department of Pathology and Laboratory Medicine, University of California Irvine, Irvine, CA, USA. 447Department of Epidemiology and Population Health, Stanford University, Stanford, CA, USA. 448Department of Neurology and Neurological Sciences, Stanford University, Stanford, CA, USA. 449Vanderbilt Memory and Alzheimer’s Center, Department of Neurology, Vanderbilt University Medical Center, Nashville, TN, USA. 450Department of Surgery, University of Texas Southwestern Medical Center, Dallas, TX, USA. 451Neurogenomics Division, Translational Genomics Research Institute, Phoenix, AZ, USA. 452Department of Pathology, Duke University, Durham, NC, USA. 453Department of Neurology, University of Texas Southwestern Medical Center, Dallas, TX, USA. 454Department of Neurological Surgery, University of Texas Southwestern Medical Center, Dallas, TX, USA. 455Hope Center Program on Protein Aggregation and Neurodegeneration, Washington University School of Medicine, St. Louis, MO, USA. 456 Department of Genome Sciences, University of Washington, Seattle, WA, USA. 457Department of Pathology and Laboratory Medicine, University of California, Davis, Sacramento, CA, USA. 458Department of Human Genetics, University of Pittsburgh, Pittsburgh, PA, USA. 459Alzheimer’s Disease Research Center, University of Pittsburgh, Pittsburgh, PA, USA. 460Department of Neurology, Albert Einstein College of Medicine, New York, NY, USA. 461Department of Neurology, Oregon Health and Science University, Portland, OR, USA. 462Department of Neurology, Portland Veterans Affairs Medical Center, Portland, OR, USA. 463Department of Laboratory Medicine and Pathology, University of Washington, Seattle, WA, USA. 464Department of Pathology, Boston 434 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles University, Boston, MA, USA. 465Department of Neuropsychology, University of California San Francisco, San Francisco, CA, USA. 466Department of Neurobiology and Behavior, University of California Irvine, Irvine, CA, USA. 467Department of Neurology, Emory University, Atlanta, GA, USA. 468Kaiser Permanente Washington Health Research Institute, Seattle, WA, USA. 469Cleveland Clinic Lou Ruvo Center for Brain Health, Cleveland Clinic, Cleveland, OH, USA. 470Department of Pathology, University of Michigan, Ann Arbor, MI, USA. 471National Center for PTSD at Boston VA Healthcare System, Boston, MA, USA. 472Department of Psychiatry, Johns Hopkins University, Baltimore, MD, USA. 473Department of Medicine (Pulmonary), New York University, New York, NY, USA. 474Department of Neurology, Miller School of Medicine, University of Miami, Miami, FL, USA. 475Department of Pathology, University of California San Diego, La Jolla, CA, USA. 476School of Nursing Northwest Research Group on Aging, University of Washington, Seattle, WA, USA. 477Pfizer Worldwide Research and Development, New York, NY, USA. 478Weill Institute for Neurosciences, Memory and Aging Center, University of California, San Francisco, San Francisco, CA, USA. 479Department of Pathology, University of Southern California, Los Angeles, CA, USA. 480Department of Pathology, Stanford University School of Medicine, Stanford, CA, USA. 481Department of Neurology, Washington University at St. Louis, St. Louis, MO, USA. 482 Institute for Translational Research, University of North Texas Health Science Center, Fort Worth, TX, USA. 483Center for Mind and Brain and Department of Neurology, University of California, Davis, Sacramento, CA, USA. 484Center for Population Health and Aging, Texas A&M University Health Science Center, Lubbock, TX, USA. 485Department of Family and Community Medicine, University of Texas Health Science Center San Antonio, San Antonio, TX, USA. 486Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA. 487Department of Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, WA, USA. 488Department of Bioengineering, University of Pennsylvania, Philadelphia, PA, USA. 489 Alzheimer’s Disease Center, New York University, New York, NY, USA. 490Department of Neurology, University of Colorado School of Medicine, Aurora, CO, USA. 491Department of Internal Medicine and Geriatrics, University of North Texas Health Science Center, Fort Worth, TX, USA. 492Department of Medical Education, TCU/UNTHSC School of Medicine, Fort Worth, TX, USA. 493Arizona Alzheimer’s Consortium, Phoenix, AZ, USA. 494Banner Alzheimer’s Institute, Phoenix, AZ, USA. 495Department of Psychiatry, University of Arizona, Phoenix, AZ, USA. 496Tanz Centre for Research in Neurodegenerative Disease, University of Toronto, Toronto, Ontario, Canada. 497Department of Radiology and Imaging Sciences, Indiana University, Indianapolis, IN, USA. 498 Department of Pathology (Neuropathology), Rush University Medical Center, Chicago, IL, USA. 499Department of Psychiatry, University of Southern California, Los Angeles, CA, USA. 500Cambridge Institute for Medical Research, University of Cambridge, Cambridge, UK. 501Faculty of Medicine, Department of Medicine (Neurology), University of Toronto, Toronto, Ontario, Canada. 502Center for Applied Health Research, Baylor Scott & White Health, Temple, TX, USA. 503College of Medicine, Texas A&M University Health Science Center, College Station, TX, USA. 504Program in Cellular Neuroscience, Neurodegeneration and Repair, Yale University School of Medicine, New Haven, CT, USA. 505Department of Psychiatry and Human Behavior, University of California, Irvine, Irvine, CA, USA. 506Renaissance Computing Institute, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA. 507 Department of Pathology, Johns Hopkins University, Baltimore, MD, USA. 508Department of Pathology and Laboratory Medicine, University of California, Los Angeles, Los Angeles, CA, USA. 509Department of Psychiatry and Behavioral Sciences, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 510Department of Psychiatry and Behavioral Sciences, Duke University, Durham, NC, USA. 511Department of Biostatistics, University of Washington, Seattle, WA, USA. 512Department of Psychiatry, New York University Grossman School of Medicine, New York, NY, USA. 513Center for Cognitive Neurology and Departments of Neurology and Pathology, New York University Grossman School of Medicine, New York, NY, USA. 514Department of Pathology, Oregon Health and Science University, Portland, OR, USA. 515Evelyn F. McKnight Brain Institute, Department of Neurology, Miller School of Medicine, University of Miami, Miami, FL, USA. 516Department of Pathology, Case Western Reserve University, Cleveland, OH, USA. CHARGE Hieab Adams1, Rufus O. Akinyemi517, Muhammad Ali518, Nicola Armstrong519, Hugo J. Aparicio109, Maryam Bahadori130, James T. Becker520, Monique Breteler521, Daniel Chasman522, Ganesh Chauhan523, Hata Comic12, Simon Cox524, Adrienne L. Cupples21, Gail Davies525, Charles S. DeCarli526, Marie-Gabrielle Duperron527, Josée Dupuis21, Tavia Evans12, Frank Fan528, Annette Fitzpatrick529, Alison E. Fohner23, Mary Ganguli528, Mirjam Geerlings530, Stephen J. Glatt531, Hector M. Gonzalez532, Monica Goss130, Hans Grabe533, Mohamad Habes534, Susan R. Heckbert535, Edith Hofer536, Elliot Hong537, Timothy Hughes130, Xueqiu Jian130, Tiffany F. Kautz538, Maria Knol12, William Kremen12, Paul Lacaze539, Jari Lahti540, Quentin Le Grand527, Elizabeth Litkowski541, Shuo Li21, Dan Liu521, Xuan Liu21, Marisa Loitfelder542, Alisa Manning543, Pauline Maillard544, Riccardo Marioni545, Bernard Mazoyer546, Debora Melo van Lent130, Hao Mei547, Aniket Mishra527, Paul Nyquist548, Jeffrey O’Connell537, Yash Patel549, Tomas Paus550, Zdenka Pausova551, Katri Raikkonen-Talvitie540, Moeen Riaz539, Stephen Rich539, Jerome Rotter552, Jose Romero553, Gena Roshchupkin12, Yasaman Saba554, Murali Sargurupremraj554, Helena Schmidt554, Reinhold Schmidt555, Joshua M. Shulman556, Jennifer Smith557, Hema Sekhar527, Reddy Rajula527, Jean Shin558, Jeannette Simino559, Eeva Sliz559, Alexander Teumer560, Alvin Thomas561, Adrienne Tin559, Elliot Tucker-Drob562, Dina Vojinovic12, Yanbing Wang9, Galit Weinstein563, Dylan Williams564, Katharina Wittfeld565, Lisa Yanek548 and Yunju Yang566 Centre for Genomic and Precision Medicine, College of Medicine, UI, Ibadan, Nigeria. 518Washington University at St. Louis, St. Louis, MO, USA. 517 Mathematics and Statistics, Curtin University, Perth, Western Australia, Australia. 520Departments of Psychiatry, Neurology, and Psychology, University 519 of Pittsburgh, Pittsburgh, PA, USA. 521Population Health Sciences, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany. 522Brigham and Women’s Hospital, Harvard University, Boston, MA, USA. 523INSERM U1219, University of Bordeaux, Bordeaux, France. 524University of Edinburgh, Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics 435 Articles NATure GeneTiCs Edinburgh, UK. 525Centre for Cognitive Ageing and Cognitive Epidemiology, University of Edinburgh, Edinburgh, UK. 526Department of Neurology and Center for Neuroscience, University of California, Davis, Davis, CA, USA. 527Bordeaux Population Health Research Center, Team VIN-TAGE, UMR 1219, University of Bordeaux, INSERM, Bordeaux, France. 528University of Pittsburgh, Pittsburgh, PA, USA. 529Department of Family Medicine, University of Washington, Seattle, WA, USA. 530University Medical Center Utrecht, Utrecht, the Netherlands. 531Psychiatric Genetic Epidemiology & Neurobiology Laboratory (PsychGENe Lab), Department of Psychiatry and Behavioral Sciences, SUNY Upstate Medical University, Syracuse, NY, USA. 532University of California, San Diego, San Diego, CA, USA. 533Department of Psychiatry and Psychotherapy, University Medicine Greifswald, Greifswald, Germany. 534Department of Radiology, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA. 535School of Public Health, University of Texas Health Science Center at Houston, Houston, TX, USA. 536Clinical Division of Neurogeriatrics, Department of Neurology, Medical University of Graz, Graz, Austria. 537 University of Maryland, College Park, MD, USA. 538Department of General Medicine, University of Texas Health Science Center, San Antonio, TX, USA. 539 Monash University Clayton Campus, Mebourne, Victoria, Australia. 540University of Helsinki, Helsinki, Finland. 541University of Colorado Anschutz Medical Center, Aurora, CO, USA. 542Medical University of Graz, Graz, Austria. 543Massachusetts General Hospital, Harvard University, Cambridge, MA, USA. 544Imaging of Dementia and Aging (IDeA) Laboratory, Department of Neurology, University of California, Davis, Davis, CA, USA. 545University of Staffmail, Edinburgh, UK. 546University of Bordeaux, IMN, Bordeaux, France. 547University of Mississippi Medical Center, Jackson, MS, USA. 548GeneSTAR Research Program, Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, USA. 549University of Toronto, Toronto, Ontario, Canada. 550Departments of Psychiatry & Neuroscience, Centre Hospitalier Universitaire Saint-Justine, University of Montreal, Montreal, Quebec, Canada. 551Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. 552Institute for Translational Genomics and Population Sciences, Los Angeles Biomedical Research Institute and Pediatrics at Harbor-UCLA Medical Center, Torrance, CA, USA. 553Boston Medical Center, Boston, MA, USA. 554 Gottfried Schatz Research Center, Department of Molecular Biology and Biochemistry, Medical University of Graz, Graz, Austria. 555Clinical Division of Neurogeriatrics, Department of Neurology, Medical University of Graz, Graz, Austria. 556Departments of Neurology, Molecular & Human Genetics, and Neuroscience and Program in Developmental Biology, Baylor College of Medicine, Houston, TX, USA. 557Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USA. 558Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. 559University of Mississippi Medical Center, Jackson, MS, USA. 560Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany. 561University of North Carolina, Chapel Hill, NC, USA. 562University of Texas, Austin, TX, USA. 563Clinical Division of Neurogeriatrics, Department of Neurology, Medical University of Graz, Graz, Austria. 564Karolinska Institute, Stockholm, Sweden. 565German Center for Neurodegenerative Diseases (DZNE), Site Rostock/ Greifswald, Greifswald, Germany. 566Institute of Molecular Medicine, University of Texas Health Science Center at Houston McGovern Medical School, Houston, TX, U­SA­­. 436 Nature Genetics | VOL 54 | April 2022 | 412–436 | www.nature.com/naturegenetics NATure GeneTiCs Articles M­et­ho­ds level if they had a P value below 5 × 10−8 in the stage I and II meta-analysis. Samples. All of our stage I meta-analysis samples came from the following However, we excluded the chr6:32657066:G:A variant, because its frequency consortia/datasets: EADB, GR@ACE, EADI, GERAD/PERADES, DemGene, amplitude was high. Bonn, the Rotterdam study, the CCHS study, NxC and the UKBB. In the UKBB, individuals who did not report dementia or any family history of dementia were Pathway analysis. A total of 10,271 gene sets were considered for analysis used as controls; the analysis included 2,447 diagnosed cases, 46,828 proxy cases of (Supplementary Note). Gene set enrichment analyses were performed in dementia and 338,440 controls. All individuals included in stage I are of European MAGMA v1.08 (refs. 65,66), with correction for the number of variants in each ancestry; demographic data on these case–control studies are summarized in gene, LD between variants and LD between genes. LD was computed from the Supplementary Table 1, and more detailed descriptions are available in the EADB-TOPMed dataset using high-quality (probability ≥0.9) imputed genotypes. Supplementary Note. Stage II samples are from the ADGC, CHARGE and FinnGen The measure of pathway enrichment was the MAGMA ‘competitive’ test (in which consortia (Supplementary Table 1 and Supplementary Note) and are described the association statistic for genes in the pathway is compared with those for all in detail elsewhere5,6,9,10,54–56. Written informed consent was obtained from study other protein-coding genes), as recommended by De Leeuw et al.67. We used participants or, for those with substantial cognitive impairment, a caregiver, the ‘mean’ test statistic, which uses the sum of −log(variant P value) across all legal guardian or other proxy. Study protocols for all cohorts were reviewed and genes. The primary analysis assigned variants to genes if they lay within the gene approved by the appropriate institutional review boards. boundaries, although a secondary analysis used a window of 35 kb upstream and 10 kb downstream to assign variants to genes (as in Kunkle et al.5). The primary Quality control and imputation. A standard quality control was performed on analysis included all variants with an imputation quality above 0.8. We used q variants and samples from all datasets individually. The samples were then imputed values68 to account for multiple testing. with the TOPMed reference panel57,58. The Haplotype Reference Consortium (HRC) panel59 was also used for some datasets (Supplementary Table 2). For the Expression in various cell types. The expression of genes was assigned to specific UKBB, we used the provided imputed data generated from a combination of the cell classes of the adult brain, as described previously69. Briefly, middle temporal 1000 Genomes, HRC and UK10K reference panels (Supplementary Note). gyrus single-nucleus transcriptomes from the Allen Brain Atlas dataset (49,555 total nuclei derived from 8 human tissue donors aged 24–66 years) were used to Stage I analyses. Tests of the association between clinical or proxy-ADD status annotate and select six main cell classes using Seurat 3.1.1 (ref. 70): glutamatergic and autosomal genetic variants were conducted separately in each dataset by using neurons, GABAergic neurons, astrocytes, oligodendrocytes, microglia and logistic regression and an additive genetic model, as implemented in SNPTEST endothelial cells. Enrichment analyses were performed by using the mean gene 2.5.4-beta3 (ref. 60) or PLINK v1.90 (ref. 4). However, a logistic mixed model (as expression per nucleus for each cell type relative to the total expression summed implemented in SAIGE v0.36.4 (ref. 61)) was considered for the UKBB data. We across cell types as a quantitative covariate in a MAGMA gene property analysis. analyzed the genotype probabilities in SNPTEST (using the newml method) and dosages in PLINK and SAIGE. Analyses were adjusted for PCs and genotyping Functional interpretation of GWAS signals and gene prioritization. To centers, when necessary (Supplementary Table 2). For the UKBB dataset, only prioritize candidate genes in the new loci, we systematically searched for variants with a MAF above 0.01% and a minor allele count (MAC) above 3 were evidence for these genes in seven different domains: (1) variant annotation, analyzed, and effect sizes and standard errors were corrected by a factor of two, (2) eQTL-GWAS integration, (3) sQTL-GWAS integration, (4) protein QTL because proxy cases were analyzed7. This approach is appropriate for variants (pQTL)-GWAS integration, (5) mQTL-GWAS integration, (6) histone acetylation with a moderate-to-high frequency and a small effect size. For all datasets, we QTL (haQTL)-GWAS integration and (7) APP metabolism. On the basis of this filtered out duplicated variants and variants with (1) missing data on the effect size, evidence, we then defined a gene prioritization score of between 0 and 100 for each standard error or P value; (2) an absolute effect size above 5; (3) an imputation candidate gene (Supplementary Fig. 34). Detailed information on the domains, quality below 0.3; and (4) a value below 20 for the product of the MAC and the categories (e.g., the tissue or cell type for QTL-GWAS integration domains) and imputation quality (MAC-info score). For datasets not imputed with the TOPMed subcategories (for the type of evidence) is given in Supplementary Table 19. A brief reference panel, we also excluded (1) variants for which conversion of position or summary of how evidence was assessed in each domain is provided below, together alleles from the GRCh37 assembly to the GRCh38 assembly was not possible or with a detailed description of the gene prioritization strategy. problematic or (2) variants with very large difference of frequency between the TOPMed reference panel and the reference panels used to perform imputation. Candidate genes. We considered protein-coding candidate genes within a ±1-Mb Results were then combined across studies in a fixed-effect meta-analysis window of the new lead variants. The genes in overlapping loci (i.e., L28, L30 with an inverse-variance weighted approach, as implemented in METAL v2011- and L37) were assigned to their respective loci based on proximity to the lead 03-25 software62. We filtered out (1) variants with a heterogeneity P value below variants, and the distal genes were not considered for gene prioritization in 5 × 10−8, (2) variants analyzed in less than 20% of the total number of cases and (3) the investigated loci. Moreover, we did not perform gene prioritization in the variants with frequency amplitude above 0.4 (defined as the difference between complex IGH gene cluster locus (L27), as this telomeric region contains complex the maximum and minimum frequencies across all the studies). We also excluded splicing events (spanning a high number of IGH genes) that probably result from variants not analyzed in the EADB-TOPMed dataset. known fusion events18. The genomic inflation factor lambda was computed with the GenABEL 1.8-0 R package63 and a median approach after exclusion of the APOE region The variant annotation domain. In this domain, we determined whether the (44–46 Mb on chromosome 19 in GRCh38). The LD score regression intercept candidate gene was the nearest protein-coding gene to the lead variant and/or was computed with LDSC v1.0.1 software using the ‘baselineLD’ LD scores built whether the lead variant was a rare variant (MAF < 1%) and/or protein-altering from 1000 Genomes phase 3 (ref. 64). The analysis was restricted to HapMap 3 variant of the investigated candidate gene. variants and excluded multiallelic variants, variants without an rs ID and variants in the APOE region. Molecular QTL–GWAS integration domains. To study the downstream effects of new ADD-associated variants on molecular phenotypes (i.e., expression, splicing, Definition of associated loci. A region of ±500 kb was defined around each protein expression, methylation and histone acetylation) in various AD-relevant variant with a stage I P value below 1 × 10−5. These regions were then merged tissues, cell types and brain regions, molecular cis-QTL information (i.e., the (using bedtools v2.27.0 software; https://bedtools.readthedocs.io/en/latest/) to genetic variants that regulate these molecular phenotypes) was integrated with define nonoverlapping regions. The region corresponding to the APOE locus was the stage I ADD GWAS results in genetic colocalization analyses, TWASs and excluded. We then used the PLINK clumping procedure to define independent a genetically driven DNA methylation scan. These molecular QTLs include hits in each region. An iterative clumping procedure was applied to all variants eQTLs, sQTLs, pQTLs, mQTLs and haQTLs. We mapped and prepared eQTL/ with a stage I P value below 1 × 10−5, starting with the variant with the lowest sQTL catalogs in AD-relevant bulk brain regions from AMP-AD cohorts71–74 P value (referred to as the index variant). Variants with a stage I P value below and in LCLs from the EADB Belgian cohort. We used additional eQTL/sQTL 1 × 10−5, located within 500 kb of this index variant and in LD with the index information in AD-relevant bulk brain regions from GTEx75 and microglia from variant (r2 above 0.001) were assigned to the index variant’s clump. The clumping the MiGA study76. Furthermore, eQTLs in monocytes and macrophages from procedure was then applied until all the variants had been clumped. LD in the various datasets77–82 (as prepared by eQTL Catalogue83) were included in the EADB-TOPMed dataset was computed using high-quality (probability ≥0.8) analyses. Data on pQTLs84, mQTLs85 and haQTLs85 were available for DLPFC. imputed genotypes. Using each molecular QTL catalogue, the effect of the lead variants was queried and significant associations were reported. Moreover, genetic colocalization studies Stage II analyses. Variants with a stage I P value below 1 × 10−5 were followed up were conducted by comparing ADD association signals with the eQTL/sQTL (Supplementary Note). Results were combined across all stage I and II studies signals from AMP-AD bulk brain, MiGA microglia and EADB LCL cohorts. We in a fixed-effect meta-analysis with an inverse variance weighted approach, as also conducted eTWASs and splicing TWASs (sTWAS) of the ADD risk, along with implemented in METAL. In each clump, we then reported the variants with fine mapping of the eTWAS results. To this end, we trained functional expression positive follow-up results (i.e., the same direction of effect in stage I and stage and splicing reference panels based on the AMP-AD bulk brain and EADB LCL II, and a stage II P value below 0.05) and the lowest P value in the meta-analysis. cohorts, and we leveraged precalculated reference panel weights86 for the GTEx Those variants were considered to be associated at the genome-wide significance dataset75 in tissues and cells of interest. Lastly, for the mQTL-GWAS integration Nature Genetics | www.nature.com/naturegenetics Articles NATure GeneTiCs domain, we also tested for associations between ADD and genetically driven the multiplicative scale. In the primary analysis, conversion to AD was used as the DNA methylation (MetaMeth analysis) in blood (with blood–brain methylation outcome (conversions to non-AD dementias were coded as being censored at time correlation estimates obtained from BECon87) using the procedures described of conversion), but analyses were repeated using all-cause dementia as the outcome. by Freytag et al.88 and Barbeira et al.89. A detailed description of the datasets and To quantify the effect size of the potential association between the GRS and methods used for each of these analyses is given in the Supplementary Note. conversion to dementia regarding predictive performance, we computed three different indices measuring different aspects of the predictive performance of the APP metabolism domain. We assessed the functional impact of gene GRS in our prospective, longitudinal cohort studies91: the continuous version of the underexpression on APP metabolism for all candidate genes based on a C-index,92,93 the continuous NRI94 and IPA95 (Supplementary Note). For all indices, genome-wide high-content short interfering RNA screen17 (Supplementary Note). we provide point estimates and 95% CIs. In the main analysis, indices were computed at the time point for which all Gene prioritization score. We computed a gene prioritization score for each cohorts in a specific setting (i.e., population-based studies or memory clinics, candidate gene as the weighted sum of the evidence identified in the seven respectively) provided follow-up observations (that is 5 years for population-based domains. We specified a weight for each type of evidence, as detailed in cohorts and 3 years for MCI cohorts). In a sensitivity analysis, indices for longer Supplementary Table 19. For the molecular QTL-GWAS integration domains, or shorter follow-up periods were also derived (that is 3 years and 10 years for we gave more weight to replicated hits (i.e., evidence in several datasets) than to population-based cohorts and 5 years for MCI cohorts). Standard errors for indices single hits. We also gave more weight to hits observed in brain (the bulk brain were derived by non-parametric bootstrapping with 1,000 samples. and microglia datasets) than to hits observed in other tissues/cell types (LCLs, To determine the average effect of the GRS across the various cohorts monocytes, macrophages and blood). To avoid score inflation, several specific examined, individual cohort results were subjected to both inverse-variance rules were applied: (1) for the results of sQTL- and mQTL-based analyses, weighted meta-analyses (primary analyses) and random effects meta-analysis multiple splice junctions or CpGs annotated for the same genes were aggregated (Supplementary Note). To facilitate comparisons of results for different time prior to weighting due to correlated data; (2) if we observed a fine-mapped points, cohorts with longer follow-up periods were meta-analyzed separately. eTWAS association for a gene, its other significant (but not fine-mapped) Furthermore, two memory clinic cohorts with a limited sample size (N < 50) were eTWAS associations were not considered; (3) for genes having several significant excluded to assess their impact on the final meta-analysis results. Meta-analyses CpGs (prior to aggregation) in MetaMeth analyses, the associated CpGs with a were performed using the ‘metafor’ (3.0.2) R package96. low (<75% percentile) blood–brain methylation correlation estimate were not To further illustrate the clinical relevance of the GRS, we pooled computed considered if the gene also had associated CpGs with a high (≥75% percentile) GRSs across four population-based cohorts (3C, AgeCoDe, VITA and MAS) blood–brain methylation correlation estimate. and computed deciles of the GRS distribution for use as a common reference for all cohorts. We then computed the increase in risk when augmenting the GRS Gene prioritization strategy. After obtaining a total weighted score per gene, we value from the first decile (GRS = 50.76) to the ninth decile (GRS = 59.74) of the ranked genes per locus according to their prioritization scores and compared distribution. To represent this risk increase in the HR, we rescaled the HR derived the relative score differences between the highest ranked gene and other genes from our meta-analyses results using the equation elog(HR)∗(GRS9thdecile −GRS1stdecile ). in the investigated locus. If this relative difference was at least 20% and the gene Importantly, this approach yields exactly the same results as transforming the GRS prioritization score for the highest ranked gene was ≥4, then we classified this gene so that a one unit increment corresponds to the increase from the lowest decile to as a tier 1 prioritized gene in the investigated locus (i.e., a greater likelihood of the highest decile. being the true risk gene responsible for the ADD signal). If this absolute threshold Furthermore, we approximated the probability of conversion to AD at 3 and 5 was not met, then the highest ranked gene was classified as a tier 2 prioritized gene years in memory clinic patients with MCI by using Cox models implemented in (i.e., a lower level of confidence and absence of the minimum level of evidence for the ‘PredictCox’ function from the ‘riskRegression’ (2020.12.8) R package97. We a true risk gene). Furthermore, other genes in a locus harboring a tier 1 gene were did not derive AD conversion probabilities for two cohorts with very small sample classified as tier 2 prioritized genes if the relative score difference versus the highest sizes (N < 50). Predicted AD conversion probabilities were derived and averaged ranked (tier 1) gene was between 20% and 50%. Lastly, when the relative score for all patients in each of the groups formed by the decile of the GRS distribution difference between the highest ranked gene and other genes in the same locus was in each cohort. The difference between the groups with the highest and lowest <20%, then both the highest ranked gene and all genes with a score difference GRSs was computed in each cohort. We report the median (range) results in each <20% were classified as tier 2 prioritized genes in the investigated locus; based on group formed by the GRS deciles. the current evidence, it is difficult to prioritize two or more similarly scored genes. The gene prioritization strategy is summarized in Supplementary Fig. 34. Detailed Reporting Summary. Further information on research design is available in the descriptions and discussions of prioritized genes and tier levels in each investigated Nature Research Reporting Summary linked to this article. new locus can be found in the Supplementary Note. Data Availability GRS analysis. Eight longitudinal MCI cohorts and seven population-based studies Genome-wide summary statistics have been deposited to the European were included in the analysis and are fully described in the Supplementary Note Bioinformatics Institute GWAS Catalog (https://www.ebi.ac.uk/gwas/) under and Supplementary Table 33. The GRSs were calculated as previously described90. accession no. GCST90027158. Briefly, we considered variants with genome-wide significant evidence of The significant eQTLs/sQTLs mapped and eTWAS/sTWAS functional reference association with ADD in our study. We did not include any APOE variants in the panel weights generated for this study (in AD-relevant bulk brain regions from GRS. Variants were directly genotyped or imputed (R² ≥ 0.3). Imputation was AMP-AD cohorts and in LCLs from the EADB Belgian cohort) are publicly performed using the HRC panel59 for subcohorts from the Rotterdam study and the available at https://doi.org/10.5281/zenodo.5745927 and https://doi.org/10.5281/ TOPMed panel for the other cohorts57. For HRC-imputed data, LD proxies were zenodo.5745929. considered for variants that were not available in this reference panel. The GRS was Anonymized aligned reads of the amplicon-based long-read Nanopore cDNA calculated as the weighted average of the number of risk-increasing alleles for each sequencing experiment conducted for the TSPAN14 splicing analysis are available variant, using dosages. Weights were based on the respective log(OR) obtained in through the European Nucleotide Archive under accession PRJEB49234. stage II. The GRS was then multiplied by the number of included variants. Thus, Moreover, the following data used in the gene prioritization are publicly available: the HR measured the effect of carrying one additional average risk allele. AMP-AD rnaSeqReprocessing Study (https://www.synapse. To assess whether the new variants in this study contribute to the risk of org/#!Synapse:syn9702085); conversion to AD (in addition to known AD genes), we calculated two GRSs: MayoRNAseq whole-genome sequencing variant call formats (WGS VCFs) one based solely on variants known before this study (GRSknown, n = 39; Table 1) (https://www.synapse.org/#!Synapse:syn11724002); and another based on variants identified in the present study (GRSnovel, n = 44; ROSMAP WGS VCFs (https://www.synapse.org/#!Synapse:syn11724057); Table 2). These GRSs were calculated in the same way as the GRS encompassing MSBB WGS VCFs (https://www.synapse.org/#!Synapse:syn11723899); all the variants. eQTLGen (https://www.eqtlgen.org/); The association between the GRS and the risk of progression to dementia in eQTL Catalogue database (https://www.ebi.ac.uk/eqtl/); individuals from population-based cohorts or patients with MCI from memory Brain xQTL serve (http://mostafavilab.stat.ubc.ca/xqtl/); clinics was tested statistically using Cox proportional hazards models. The models GTEx v8 eQTL and sQTL catalogs (https://www.gtexportal.org/); were adjusted for age, sex, the first four PCs (to correct for potential population GTEx v8 expression and splicing prediction models (http://predictdb.org/); stratification) and the number of APOE-ε4 and APOE- ε2 alleles (assuming MiGA eQTLs (https://doi.org/10.5281/zenodo.4118605); an additive effect). In the FHS study, the generation was used as an additional MiGA sQTLs (https://doi.org/10.5281/zenodo.4118403); covariate. In the 3C study, the analysis was adjusted for age, sex, the number of MiGA meta-analysis (https://doi.org/10.5281/zenodo.4118676); and APOE alleles, the two first PCs and center. The PCs used were generated for each Wingo et al.84 pQTL data (https://www.synapse.org/#!Synapse:syn23627957). cohort, using the same variants as in the case/control study’s PC analysis. The number of APOE-ε4 alleles was obtained from direct genotyping or, if missing, the Code availability genotypes (with probability >0.8) derived from the TOPMed imputations. The We used publicly available software for all analyses. The software are listed in the interaction between the GRS and the number of APOE-ε4 alleles was tested on Supplementary Note with their appropriate citations and/or URLs. Nature Genetics | www.nature.com/naturegenetics NATure GeneTiCs Articles References 84. Wingo, A. P. et al. Integrating human brain proteomes with genome-wide 54. Psaty, B. M. et al. Cohorts for Heart and Aging Research in Genomic association data implicates new proteins in Alzheimer’s disease pathogenesis. Epidemiology (CHARGE) Consortium design of prospective meta-analyses of Nat. Genet. 53, 143–146 (2021). genome-wide association studies from 5 cohorts. Circulation: Cardiovasc. 85. Ng, B. et al. An xQTL map integrates the genetic architecture of the human Genet. 2, 73–80 (2009). brain’s transcriptome and epigenome. Nat. Neurosci. 20, 1418–1426 (2017). 55. Naj, A. C. et al. Common variants at MS4A4/MS4A6E, CD2AP, CD33 and 86. Barbeira, A. N. et al. Exploiting the GTEx resources to decipher the EPHA1 are associated with late-onset Alzheimer’s disease. Nat. Genet. 43, mechanisms at GWAS loci. Genome Biol. 22, 49 (2021). 436–441 (2011). 87. Edgar, R. D., Jones, M. J., Meaney, M. J., Turecki, G. & Kobor, M. S. BECon: a 56. Jun, G. et al. Meta-analysis confirms CR1, CLU, and PICALM as Alzheimer tool for interpreting DNA methylation findings from blood in the context of disease risk loci and reveals interactions with APOE genotypes. Arch. Neurol. brain. Transl. Psychiatry 7, e1187 (2017). 67, 1473–1484 (2010). 88. Freytag, V. et al. Genetic estimators of DNA methylation provide insights into 57. Taliun, D. et al. Sequencing of 53,831 diverse genomes from the NHLBI the molecular basis of polygenic traits. Transl. Psychiatry 8, 31 (2018). TOPMed Program. Nature 590, 290–299 (2021). 89. Barbeira, A. N. et al. Exploring the phenotypic consequences of tissue specific 58. Das, S. et al. Next-generation genotype imputation service and methods. Nat. gene expression variation inferred from GWAS summary statistics. Nat. Genet. 48, 1284–1287 (2016). Commun. 9, 1825 (2018). 59. McCarthy, S. et al. A reference panel of 64,976 haplotypes for genotype 90. Chouraki, V. et al. Evaluation of a genetic risk score to improve risk imputation. Nat. Genet. 48, 1279–1283 (2016). prediction for Alzheimer’s disease. J. Alzheimers Dis. 53, 921–932 (2016). 60. Marchini, J., Howie, B., Myers, S., McVean, G. & Donnelly, P. A new 91. Steyerberg, E. W. et al. Assessing the performance of prediction models: a multipoint method for genome-wide association studies by imputation of framework for traditional and novel measures. Epidemiology 21, 128–138 (2010). genotypes. Nat. Genet. 39, 906–913 (2007). 92. Harrell, F. E., Califf, R. M., Pryor, D. B., Lee, K. L. & Rosati, R. A. Evaluating 61. Zhou, W. et al. Efficiently controlling for case-control imbalance and sample the yield of medical tests. JAMA 247, 2543–2546 (1982). relatedness in large-scale genetic association studies. Nat. Genet. 50, 93. Gerds, T. A., Kattan, M. W., Schumacher, M. & Yu, C. Estimating a 1335–1341 (2018). time-dependentconcordance index for survival prediction models with 62. Willer, C. J., Li, Y. & Abecasis, G. R. METAL: fast and efficient meta-analysis covariate dependent censoring. Stat. Med. 32, 2173–2184 (2013). of genomewide association scans. Bioinformatics 26, 2190–2191 (2010). 94. Pencina, M. J., D’Agostino, R. B. & Steyerberg, E. W. Extensions of net 63. Aulchenko, Y. S., Ripke, S., Isaacs, A. & van Duijn, C. M. GenABEL: reclassification improvement calculations to measure usefulness of new an R library for genome-wide association analysis. Bioinformatics 23, biomarkers. Stat. Med. 30, 11–21 (2011). 1294–1296 (2007). 95. Kattan, M. W. & Gerds, T. A. The index of prediction accuracy: an intuitive 64. Bulik-Sullivan, B. et al. LD score regression distinguishes confounding measure useful for evaluating risk prediction models. Diagnostic Progn. Res. from polygenicity in genome-wide association studies. Nat. Genet. 47, 2, 7 (2018). 291–295 (2015). 96. Viechtbauer, W. Conducting meta-analyses in R with the metafor. J. Stat. 65. de Leeuw, C. A., Mooij, J. M., Heskes, T. & Posthuma, D. MAGMA: Softw. 36, 1–48 (2010). generalized gene-set analysis of GWAS data. PLoS Comput. Biol. 11, e1004219 97. Ozenne, B., Sørensen, A. L., Scheike, T., Torp-Pedersen, C. & Gerds, T. (2015). riskRegression: predicting the risk of an event using Cox regression models. 66. Yurko, R., Roeder, K., Devlin, B. & G’Sell, M. H‐MAGMA, inheriting a shaky R. J. 9, 440–460 (2017). statistical foundation, yields excess false positives. Ann. Hum. Genet. 85, 97–100 (2021). Acknowledgements 67. De Leeuw, C. A., Neale, B. M., Heskes, T. & Posthuma, D. The statistical We thank the many study participants, researchers and staff for collecting and properties of gene-set analysis. Nat. Rev. Genet. 17, 353–364 (2016). contributing to the data, the high-performance computing service at the University 68. Storey, J. D. & Tibshirani, R. Statistical significance for genomewide studies. of Lille and the staff at CEA-CNRGH for their help with sample preparation and Proc. Natl Acad. Sci. U. S. A. 100, 9440–9445 (2003). genotyping and excellent technical assistance. We thank Antonio Pardinas for his help. 69. Coelho, D. M., L. I. da Cruz Carvalho, Melo-de-Farias, A. R., Lambert, J.-C. We thank the Netherlands Brain Bank. This research was conducted using the UKBB & Costa, M. R. Differential transcript usage unravels gene expression resource (application number 61054). This work was funded by a grant (EADB) from the alterations in Alzheimer’s disease human brains. NPJ Aging Mech. Dis. 7, EU Joint Programme – Neurodegenerative Disease Research. INSERM UMR1167 is also 2 (2021). funded by the INSERM, Institut Pasteur de Lille, Lille Métropole Communauté Urbaine 70. Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating and French government’s LABEX DISTALZ program (development of innovative single-cell transcriptomic data across different conditions, technologies, and strategies for a transdisciplinary approach to AD). Full consortium acknowledgements species. Nat. Biotechnol. 36, 411–420 (2018). and funding are in the Supplementary Note. 71. Allen, M. et al. Human whole genome genotype and transcriptome data for Alzheimer’s and other neurodegenerative diseases. Sci. Data 3, 160089 (2016). Author contributions 72. Mostafavi, S. et al. A molecular network of the aging human brain provides EADB coordination: K. Mather, F.J., M.T., R.F.-S., J. Clarimon, J.-F. Deleuze, O.A.A., insights into the pathology and cognitive decline of Alzheimer’s disease. Nat. M.I., M. Hiltunen, K.S., C.M.v.D., R.S., W.M.v.d.F., A. Ruiz, A. Ramirez and J.-C.L. Data Neurosci. 21, 811–819 (2018). analyses: C. Bellenguez, F.K., I.E.J., L.K., S.M.-G., N.A., R.C., B.G.-B., V. Andrade, P.A.H., 73. Bennett, D. A. et al. Religious orders study and rush memory and aging R.C.-M., V.D., S.J.v.d.L., M.R.C., T.K., I.R., J. Chapuis and P.G.-G. ADGC analysis and project. J. Alzheimer’s Dis. 64, S161–S189 (2018). coordination: A.C.N., W.S.B., L.A.F., J.L.H., K.L.H.-N., P.P.K., B.W.K., C.-Y.L. and E.R.M., 74. Wang, M. et al. The Mount Sinai cohort of large-scale genomic, transcriptomic R. Mayeux, M.A.P.-V., J.S., L.-S.W., Y.Z. and G.D.S. Charge analysis and coordination: and proteomic data in Alzheimer’s disease. Sci. Data 5, 180185 (2018). Q.Y., J.C.B., A.D.S., C.S., B.M.P., R.W., O. Lopez. and S. Seshadri. FinnGen analysis: 75. GTEx Consortium, The GTEx Consortium atlas of genetic regulatory effects T.K. and M. Hiltunen. Rotterdam analysis: A.Y., I.P.-N, M. Ghanbari and M.A.I. Sample across human tissues. Science 369, 1318–1330 (2020). contribution: S. Ahmad, V. Giedraitis, D. Aarsland, P.V.Ã., D.G.-G., C. Abdelnour, 76. De, K. et al. Atlas of genetic effects in human microglia transcriptome across E.A.-M., D. Alcolea, M. Alegret, I. Alvarez, V. Alvarez, N.J.A., A. Tsolaki, C. Antúnez, brain regions, aging and disease pathologies. Preprint at bioRxiv https://doi. I. Appollonio, M. Arcaro, S. Archetti, A.A.P., B.A., L.A., H. Bailly, N.B., M. Baquero, S. org/10.1101/2020.10.27.356113 (2020). Barral, A. Beiser, A.B.P., J.E.B., P. Benchek, L.B., C. Berr, C. Besse, V. Bessi, G. Binetti, 77. Alasoo, K. et al. Shared genetic effects on chromatin and gene expression A. Bizarro, R.B., M. Boada, E.B., B.B., S. Boschi, P. Bossù, G. Bråthen, J.B., C. Bresner, indicate a role for enhancer priming in immune response. Nat. Genet. 50, H. Brodaty, K.J.B., L.I.B., D.B.-R., K.B., V. Burholt, W.S.B., M.C., L.B.C., G.C., J. Chung, 424–431 (2018). M.L.C., Ã.C., R.C., L.C.-C., C. Charbonnier, H.-H.C., C. Chilotti, S.C., J.A.C., C. Clark, 78. Nédélec, Y. et al. Genetic ancestry and natural selection drive population E. Conti, A.C.-G., E. Costantini, C. Custodero, D.D., M.C.D., A. Daniele, E. Dardiotis, differences in immune responses to pathogens. Cell 167, 657–669.e21 (2016). J-F. Dartigues, P.P.d.D., K.d.P.L., L.D.d.W., S. Debette, J.D., T.d.S., N.D., A. DeStefano, 79. Chen, L. et al. Genetic drivers of epigenetic and transcriptional variation in M.D., J.D.-S., M.D.-F., P.D.R., S. Djurovic, E. Duron, E. Duzel, C.D., G.E., S.E., V.E.-P., human immune cells. Cell 167, 1398–1414.e24 (2016). A.E., M.E., K.M.F., T. Fabrizio, S.F.N., D.W.F., L. Farotti, C.F., M.F.-F., R.F., C.B.F., E.F., 80. Momozawa, Y. et al. IBD risk loci are enriched in multigenic regulatory B. Fin, P.F., T. Fladby, K.F., B. Fongang, M.F., J.F., T.M.F., S.F., N.C.F., E.F.-M., M.J.B., modules encompassing putative causative genes. Nat. Commun. 9, A.F.-G., L. Froelich, B.F.-H., D.G., J.M.G.-A., S.G.-M., G.G.-R., R.G., I.G., G. Giorgio, 2427 (2018). A.M.G., O.G., D.G.-F., A.G.-P., C.G., G. Grande, E. Green, T.G., E. Grunblatt, M. Grunin, 81. Fairfax, B. P. et al. Innate immune activity conditions the effect of regulatory V. Gudnason, T.G.-B., A.H., G.H., J.L.H., K.L.H.-N., H. Hampel, O.H., J. Hardy, A.M.H., variants upon monocyte gene expression. Sci. (80-.). 343, 1246949 (2014). L.H., J. Harwood, S.H.-H., S.H., M.T.H., I.H., M.J.H., P.H., C.H., H. Holstege, R.H.V., M. 82. Quach, H. et al. Genetic adaptation and neandertal admixture shaped the Hulsman, J. Humphrey, G.J.B., X.J., C.J., G.R.J., Y.K., J. Kauwe, P.G.K., L. Kilander, immune system of human populations. Cell 167, 643–656.e17 (2016). A.K.S., M.K., A.K., J. Kornhuber, M.H.K., W.A.K., P.P.K., B.W.K., A.B.K., C.L., E.J.L., L. 83. Kerimov, N. et al. eQTL catalogue: a compendium of uniformly processed Launer, A. Lauria, C.-Y.L., J.L., O.Ler., A. Lleó, W.L.J., O. Lopez, A.L.d.M., S.L., M.L., L. human gene expression and splicing QTLs. Preprint at bioRxiv https://doi. Luckcuck, K.L.L., Y.M., J.M., C.A.M., W.M., F. Mangialasche, M. Spallazzi, M. Marquié, org/10.1101/2020.01.29.924266 (2020). R. Marshall, E.R.M., A.M.M., C.M.R., C. Masullo, R. Mayeux, S. Mead, P. Mecocci, Nature Genetics | www.nature.com/naturegenetics Articles NATure GeneTiCs M. Menéndez-González, A.M., S. Mehrabian, S. Mendoza, M.M.-G., P. Mir, S. Moebus, and exclusively his own opinion. He serves as Senior Associate Editor for the Alzheimers M. Mol, L.M.-P., L. Montrreal, L. Morelli, F. Moreno, K. Morgan, T. Mosley, M.M.N., & Dementia journal and has not received any fees or honoraria since May 2019. Before C. Muchnik, S. Mukherjee, B.N., T.N., G.N., B.G.N., R.O., A.O., M.O., G.O., A.P., C. May 2019, H. Hampel received lecture fees from Servier, Biogen and Roche; research Paollo., G. Papenberg, L.P., F.P., P. Pastor, G. Peloso, A.P.-C., J.P.-T., P. Pericard, O.P., grants from Pfizer, Avid and MSD Avenir (paid to the institution); travel funding from Y.A.P., J.A.P., G.P.-R., C. Pisanu, T.P., J. Popp, D.P., J. Priller, R.P., O.Q., I.Q., J.Q.T., A. Eisai, Functional Neuromodulation, Axovant, Eli Lilly and Company, Takeda, Zinfandel Rábano, I. Rainero, F.R., I. Ramakers, L.M.R., M.J.R., C.R., D.R.-D., P. Ridge, S.R.-H., P. Pharmaceuticals, GE Healthcare and Oryzon Genomics; and consultancy fees from Riederer, N.R., E.R.-R., A. Rongve, I.R.A., M.R.-R., J.L.R., E.R., D.R., M.E.S., P. Sakka, Qynapse, Jung Diagnostics, Cytox, Axovant, Anavex, Takeda, Zinfandel Pharmaceuticals, I.S., Ã.S., M.B.S.-A., F.S.-G., P.S.J., R.S.-V., S.B.S., C.S., C.L.S., M. Scamosci, N. Scarmeas, GE Healthcare, Oryzon Genomics and Functional Neuromodulation. He served as a E. Scarpini, P. Scheltens, N. Scherbaum, M. Scherer, M. Schmid, A. Schneider, J.M.S., scientific advisory board member for Functional Neuromodulation, Axovant, Eisai, Eli G. Selbæk, D.S., M. Serrano, J.S., A.A.S., O.S., S. Slifer, G.J.L.S., H.S., V.S., A. Solomon, Lilly and Company, Cytox, GE Healthcare, Takeda and Zinfandel, Oryzon Genomics and Y.S., S. Sorbi, O.S.-G., G. Spalletta, A. Spottke, A. Squassina, E. Stordal, J.P.T., L. Tárraga, Roche Diagnostics. The remaining authors declare no competing interests N.T., A. Thalamuthu, T.T., G.T., L. Traykov, L. Tremolizzo, A.T.-H., A. Uitterlinden, A. Ullgren, I.U., S.V., O.V., C.V.B., J. Vance, B.N.V., A.v.d.L., J.V.D., J.v.R., J.v.S., R.V., F.V., J.-S.V., J. Vogelgsang, M.V., M.W., D.W., L.-S.W., R.W., L.W., J. Wiltfnag, G.W., B.W., M.Y., Additional information H.Z., Y.Z., X.Z., C.Z., M.Z., L.A.F., B.M.P., M. Ghanbari, T.R., P. Sachdev, K. Mather, F.J., Supplementary information The online version contains supplementary material M.A.I., A.d.M., J. Hort, M.T. and M.A.P.-V. Core writing group: C. Bellenguez, F.K., V. available at https://doi.org/10.1038/s41588-022-01024-z. Andrade, B.G.-B., P.A.H., R.C.-M., L.K., S.J.v.d.L., K.S., A. Ruiz, A. Ramirez and J.-C.L. Correspondence and requests for materials should be addressed to Céline Bellenguez or Jean-Charles Lambert. Competing interests Peer review file Nature Genetics thanks the anonymous reviewers for their contribution H. Hampel is an employee of Eisai. The present article was initiated and prepared as part to the peer review of this work. Peer reviewer reports are available. of his academic position at Sorbonne University (Paris, France), and it reflects entirely Reprints and permissions information is available at www.nature.com/reprints. Nature Genetics | www.nature.com/naturegenetics