{"paper_id":"068d3e8d-72e9-4269-b3a5-8d312abc2a1b","body_text":"The genetic architecture of brainstem structures | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article The genetic architecture of brainstem structures Chunshui Yu, Hui Xue, Jilian Fu, Zuojun Geng, Jingliang Cheng, and 42 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5060768/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 10 Dec, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract The brainstem contains numerous nuclei and tracts with vital functions. Genome-wide associations with brainstem substructure volumes are explored in European individuals, however other ancestries are under-represented. Here, we conducted the cross-ancestry genome-wide association meta-analyses in 72,717 individuals for brainstem and 48,522 for eight substructure volumes, including 7,096 Chinese Han individuals. We identified 122 genetic loci associated with brainstem and substructure volumes at P < 5.56 ×10 − 9 , including 46 new loci. Three associations had different effect sizes and 292 associations had similar effect sizes between ancestries. We prioritized 550 genes for these brainstem volumetric traits, primarily enriching for neural development. We identified the shared and distinct genetic loci, genes, and pathways for midbrain, pons, and medulla volumes, and the shared genetic architectures with brainstem-related neuropsychiatric disorders and physiological functions. The results provide new insight into genetic architectures of brainstem and substructure volumes and their genetic associations with brainstem physiologies and pathologies. Biological sciences/Genetics/Genetic association study/Genome-wide association studies Biological sciences/Neuroscience/Neurogenesis Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction The brainstem anatomically connects the cerebrum to cerebellum and spinal cord, and is composed of medulla, pons, and midbrain. It contains numerous grey matter nuclei and white matter tracts that are critical for our life. For instance, the vital center in the medulla is responsible for the autonomic control of breathing, blood pressure, and heart rate 1 – 3 . The locus coeruleus in the pons is the main source of noradrenaline playing key roles in vigilance and cognition, and the dysfunction of this nucleus may lead to arousal-related disorders, such as attention deficit hyperactivity disorder (ADHD) and major depressive disorder (MDD) 4 . The substantia nigra in the midbrain contains enormous dopaminergic neurons, generating dopamine (DA) essential for the control of voluntary movement and the regulation of emotion. Damage to this structure is associated with schizophrenia (SCZ), Parkinson’s disease (PD), MDD, and ADHD 5 , 6 . In addition to the grey matter nuclei, white matter fibers in the brainstem are also important for various physiological functions. For example, the reticular formation is crucial for maintaining consciousness 7 and the corticospinal tract is vital for the control of limb movement. Based on the individual’s brain structural magnetic resonance imaging (MRI) data, the brainstem and substructure (medulla, pons, and midbrain) volumes can be estimated by approaches, such as the Bayesian brainstem segmentation 8 . Significant differences in brainstem substructure volumes are observed between healthy controls and patients with neuropsychiatric disorders 8 – 11 , such as Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS), migraine, and autism spectrum disorder (ASD), indicating that the brainstem and substructure volumes are potential neuroimaging markers for these neuropsychiatric disorders. To investigate the genetic architectures of these brainstem volumetric traits, five genome-wide association studies (GWASs) are conducted for the whole brainstem volume in up to 40,282 individuals 12 – 16 , including two studies 12 , 13 also performing GWASs for brainstem substructure volumes in up to 33,224 individuals. Although one study involving the GWAS for whole brainstem volume also includes 7,058 individuals with East Asian ancestry (EAS) 16 , other studies only include individuals with European ancestry (EUR) 12 – 15 , preventing from distinguishing ancestry-specific from ancestry-shared associations for brainstem substructure volumes. In the present study, we conducted the largest GWAS for whole brainstem volume in 72,717 non-overlapping individuals, among which 7,096 EAS participants were from the Chinese Imaging Genetics (CHIMGEN) study 17 and 65,621 EUR participants were from the UK Biobank (UKBB) study 18 , the Adolescent Brain Cognitive Development (ABCD) study 19 , the Cohorts of Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium 20 , and the Enhancing Neuroimaging Genetics through Meta-Analysis (ENIGMA) consortium 21 . We performed the largest GWASs for brainstem substructure volumes in 48,522 individuals (7,096 EAS participants from CHIMGEN and 41,426 EUR participants from ABCD and UKBB). We identified ancestry-shared and ancestry-specific genetic associations with brainstem volumetric traits based on the EAS-GWAS and EUR-GWAS summary statistics. We detected potential causal variants by statistical fine-mapping and prioritized genes by integrating with genomic location, gene expression, and chromatin interaction data. We searched for shared and distinct genetic loci, genes, and biological pathways between brainstem substructure volumes, and explored shared genetic architectures of brainstem and substructure volumes with brainstem-related physiological and pathological phenotypes by conducting genetic correlation, genetic colocalization, and conditional and conjunctional false discovery rate (condFDR/conjFDR) analyses. The study design is presented in Supplementary Fig. 1 . By increasing sample size and population diversity, we found more genetic loci associated with brainstem and substructure volumes, providing a better understanding of the genetic architecture of brainstem volumetric traits. Results Overview of brainstem volumetric traits and GWAS strategies We included three sets of raw neuroimaging genetics datasets (CHIMGEN, ABCD, and UKBB) and a set of GWAS summary data for whole brainstem volume from ENIGMA, CHARGE, and UKBB (first release) datasets 14 . Consisting with the prior GWASs 12 , 13 for brainstem substructure volumes, we used the same Bayesian approach 8 to obtain the absolute volumes of whole brainstem, medulla, pons, midbrain, and superior cerebellar peduncle (SCP) (Fig. 1 a) of each CHIMGEN, ABCD, or UKBB participant, although SCP is not a classical brainstem substructure. The obtained whole brainstem (including SCP) volume was included as an additional covariate to perform GWASs for relative volumes of medulla, pons, midbrain, and SCP 12 . As the previous GWAS 14 applied the automatic subcortical segmentation method to obtain whole brainstem (not including SCP) volume, we also used the same method to calculate the whole brainstem volume for each CHIMGEN, ABCD, or UKBB participant, which was used in GWASs for the trait. We investigated the consistency of whole brainstem volumes obtained by the two segmentation methods by calculating intraclass correlation coefficient (ICC) of whole brainstem volumes derived from the two methods for CHIMGEN, ABCD, and UKBB datasets, respectively. We found high ICCs (0.956–0.997) ( Supplementary Table 1 ). Based on the participants from each dataset who had neuroimaging data acquired at two different time points, we also calculated the ICC of each substructure volume to assess the test-retest reliability of the Bayesian brainstem segmentation, and found high ICCs (0.868–0.992) ( Supplementary Table 1 ). The demographic and brainstem volumetric data are presented in Supplementary Table 2 . We used the mixed linear model (MLM) in fastGWA 22 to conduct GWASs for brainstem volumetric traits in CHIMGEN, ABCD, and UKBB participants, respectively. Based on the obtained GWAS summary statistics and those from the prior EUR-GWAS for whole brainstem volume 14 , we used inverse variance weighted (IVW) fixed effect model in METAL 23 to conduct EUR and cross-ancestry GWAS meta-analyses to make full use of the available data resources. As the prior GWAS for whole brainstem volume 14 included the first released UKBB data, we excluded the participants and reperformed UKBB-GWAS for whole brainstem volume, and utilized the obtained GWAS summary statistics for the trait in the meta-analyses with the prior study 14 . The participants and genetic variants included in each GWAS are shown in Supplementary Table 3 . Due to lacking any independent replication, we only reported the study-wide significant associations ( P < 5.56 × 10 − 9 , Bonferroni correction for nine traits). The genomic control inflation factor (λ GC ) and linkage disequilibrium score regression (LDSC) intercepts 24 were used to identify population stratification for each GWAS, and did not show any population stratification ( Supplementary Table 4 ). We used imputed genotype data from 7,096 CHIMGEN and 35,611 UKBB participants to construct EAS, EUR, and cross-ancestry LD references, which were used to identify independent variant-trait and locus-trait associations, and LD-independent loci in EAS, EUR, and cross-ancestry GWASs, respectively. Genetic discovery in EAS-GWASs We conducted EAS-GWASs ( P < 5.56 × 10 − 9 , Bonferroni corrected) for nine brainstem volumetric traits at 8,790,144 autosomal and 227,168 X-chromosomal variants with information score (info) ≥ 0.6 and minor allele frequency (MAF) ≥ 0.5% in 7,096 CHIMGEN participants. We only identified one independent variant-trait association ( P = 2.61 × 10 − 9 ; Fig. 1 b, Supplementary Fig. 2 and Supplementary Table 5 ) between rs6547787 at 2p11.2 and pons absolute volume. The variant rs6547787 is an expression quantitative trait locus (eQTL) of EIF2AK3 , a risk gene for progressive supranuclear palsy (PSP) characterized by brainstem pathology 25 . Genetic discovery in EUR-GWASs We conducted EUR-GWASs ( P < 5.56 × 10 − 9 , Bonferroni corrected) for the brainstem volumetric traits at 9,064,819 autosomal and 278,985 X-chromosomal variants with MAF ≥ 0.5% and imputation r 2 > 0.3 in 6,060 ABCD participants. We identified three independent variant-trait associations and three LD-independent loci ( Supplementary Fig. 3 and Supplementary Table 6 ), including the variant-trait associations between rs10983073 at 9q33.1 and pons absolute volume ( P = 2.25 × 10 − 10 ), rs6775519 at 3p25.1 and pons relative volume ( P = 1.16 × 10 − 9 ), and rs12449302 at 17q11.2 and midbrain absolute volume ( P = 1.39 × 10 − 10 ; Fig. 1 c). We conducted EUR-GWASs ( P < 5.56 × 10 − 9 , Bonferroni corrected) for brainstem volumetric traits at 10,042,001 autosomal and 390,192 X-chromosomal variants with MAF ≥ 0.5% and info ≥ 0.6 in 35,611 UKBB participants. We found 268 independent variant-trait associations (involving 191 autosomal and four X-chromosomal variants), 200 locus-trait associations, and 91 LD-independent loci ( Supplementary Fig. 4 and Supplementary Table 7 ). Among these variant-trait associations, four variants were missense mutations, linking rs13388394 ( RASGRP3 ) to whole brainstem volume and pons absolute volume, rs1805165 ( EIF2AK3 ) and rs2234675 ( PAX3 ) to pons absolute volume, and rs2273171 ( STRN3 ) to midbrain absolute volume. We also identified one stop-gained variant, linking rs2231142 ( ABCG2 ) to whole brainstem volume and pons absolute volume. We conducted EUR-GWAS meta-analyses ( P < 5.56 × 10 − 9 , Bonferroni corrected) for the eight brainstem substructure volumes at 9,637,747 autosomal and 316,973 X-chromosomal variants in 41,426 EUR participants (5,815 from ABCD and 35,611 from UKBB). As the previous EUR-GWAS (28,809 individuals from ENIGMA, CHARGE, and UKBB) for whole brainstem volume 14 included first released UKBB participants, we excluded these participants and reperformed GWAS for the trait in 30,752 UKBB participants. Based on the GWAS summary data of the prior EUR-GWAS (n = 28,809), UKBB-GWAS (n = 30,752), and ABCD-GWAS (n = 6,060), we conducted the EUR-GWAS meta-analysis for whole brainstem volume in 65,621 participants at 9,755,044 autosomal and 317,092 X-chromosomal variants using METAL 23 , which could output summary statistics for variants included in any GWAS, providing the largest number of EUR-GWAS association statistics. In the EUR-GWAS meta-analyses for the brainstem volumetric traits ( P < 5.56 × 10 − 9 , Bonferroni corrected), we identified 367 independent variant-trait associations (involving 276 autosomal and four X-chromosomal variants), 261 locus-trait associations, and 112 LD-independent loci ( Supplementary Fig. 5 and Supplementary Table 8 ). For example, we found a meaningful association between rs803903 at 9q33.1 and medulla relative volume ( P = 5.39 × 10 − 15 ; Fig. 1 d). The lead variant is mapped to ASTN2 , involving in neuronal migration and linking to migraine without aura 26 , characterized by tractus solitarius impairment in medulla 27 . Genetic findings in cross-ancestry GWASs With the same approach as EUR-GWAS meta-analyses, we conducted cross-ancestry GWAS meta-analyses ( P < 5.56 × 10 − 9 , Bonferroni corrected) for brainstem volumetric traits in up to 72,717 participants by further including EAS-GWASs (n = 7,096). In cross-ancestry GWASs, we found 375 independent variant-trait associations (involving 285 autosomal and four X-chromosomal variants), 260 locus-trait associations, and 108 LD-independent loci ( Supplementary Fig. 6 and Supplementary Table 9 ). The cross-ancestry GWASs identified many associations that were not significant in EAS-GWASs and EUR-GWASs. For example, the association between rs3213141 at 20q11.22 and pons absolute volume (Fig. 1 e) was significant in cross-ancestry GWAS ( P = 3.13 × 10 − 9 ) rather than EAS-GWAS ( P = 3.76 × 10 − 3 ) and EUR-GWAS ( P = 1.70 × 10 − 7 ). Pooling significant associations and identifying new loci As most of the prior GWASs for brainstem and substructure volumes were conducted in EUR individuals and EUR samples were at least six times larger than EAS samples in this study, we used the EUR-LD reference to pool GWAS results and identify new associations and loci. We pooled all study-wide significant variant-trait associations ( P < 5.56 × 10 − 9 ) from all GWASs and identified 430 independent variant-trait associations, 296 independent locus-trait associations, and 122 LD-independent loci (Fig. 1 f, g and Supplementary Tables 10 and 11 ). To identify new variant-trait associations and loci, we integrated all previous GWASs for brainstem and substructure volumes 12 – 16 to define the references ( P < 5.56 × 10 − 9 ) of known variant-trait (n = 211) and locus-trait (n = 162) associations, and LD-independent loci (n = 74). Using the EUR-LD reference and the same threshold ( P < 5.56 × 10 − 9 ), we defined a new variant-trait association if the variant was 500 kb away from and not in LD (r 2 < 0.1) with any variants of the same trait in the list of known variant-trait associations; a new locus-trait association when all lead variants in the locus were 500 kb away from and not in LD (r 2 < 0.1) with any lead variants in the loci of all known locus-trait associations; and a novel locus when all lead variants in the locus were 500 kb away from and not in LD (r 2 < 0.1) with any lead variants in known loci. By definition, we found 287 new variant-trait associations, 175 new locus-trait associations, and 63 new LD-independent loci. Even using a lenient threshold of P < 5 × 10 − 8 to define the known associations (334 known variant-trait and 236 locus-trait associations, and 104 known LD-independent loci), we still found 241 new variant-trait and 150 new locus-trait associations, and 46 new LD-independent loci (Fig. 1 g and Supplementary Tables 10 and 11 ). For example, the new locus at 3p14.1 was associated with both absolute and relative volumes of SCP (a white matter tract formed by enormous axons), its lead variant rs2306272 is a missense variant of LRIG1 , which controls axonal extension, guidance, and branching by regulating growth factor signals during neural development 28 . Heritability of brainstem and substructure volumes After excluding the major histocompatibility complex (MHC) region from the genome, we utilized LDSC 24 , 29 to estimate the single nucleotide polymorphism (SNP)-based heritability of the nine brainstem volumetric traits in EAS, EUR, and cross-ancestry populations based on the EAS, EUR, and cross-ancestry GWAS summary data and LD references, respectively. All brainstem volumetric traits showed significant SNP-based heritability (0.17–0.40, all P < 1.17 × 10 − 29 ; Supplementary Table 12 ) in the EUR and cross-ancestry individuals, although the SNP-based heritability of two traits was not significant ( P > 0.05) in EAS individuals. Cross-ancestry genetic correlations of brainstem volumetric traits After excluding the genetic variants in the MHC region, we used Popcorn 30 to calculate the genetic-effect correlation (without considering MAF information) of each brainstem volumetric trait between EAS and EUR, and identified relatively high correlations (0.44–1.13; Supplementary Table 13 ). We also calculated genetic-impact correlations (considering MAF information) for these brainstem volumetric traits between EAS and EUR, and also found relatively high correlations (0.49–1.12). Ancestry-shared and ancestry-specific variant-trait associations To identify the ancestry-shared and ancestry-specific variant-trait associations for nine brainstem volumetric traits, we used the Cochran's Q (CQ) test to quantify effect size differences between EAS and EUR for the pooled variant-trait associations ( P < 5.56 × 10 − 9 ) from EAS-GWASs, EUR-GWASs, and cross-ancestry GWASs. Of the 430 pooled independent variant-trait associations, the variants of 369 associations were included in both CHIMGEN and UKBB datasets, which were used for comparing the effect size differences between EAS and EUR. Based on the CQ-tests, we defined the associations with P ≥ 0.05 as ancestry-shared associations and P < 1.36 × 10 − 4 (Bonferroni corrected) as ancestry-specific associations. We found three ancestry-specific associations and 313 ancestry-shared associations (Fig. 2 a, b and Supplementary Table 14 ). The variants of the three ancestry-specific associations and 300 ancestry-shared associations were also included in the ABCD dataset. Using ABCD to replace UKBB data, we replicated all three ancestry-specific associations (CQ-test: P < 0.0167, Bonferroni corrected) and 292/300 ancestry-shared associations (CQ-test: P ≥ 0.05) (Fig. 2 b and Supplementary Table 15 ). The three ancestry-specific associations were found between rs111883632 and pons absolute volume, rs606599 and pons relative volume, and rs151057105 and midbrain absolute volume (Fig. 2 c). As an example of ancestry-shared associations, the association (Fig. 2 d) between rs16863657 and pons absolute volume was significant in UKBB-GWAS ( P = 6.72 × 10 − 16 ) and nearly genome-wide significant in CHIMGEN-GWAS ( P = 6.88 × 10 − 8 ). The variant is at the upstream of PAX3 , a transcription factor regulating fetal neural development 31 . Statistical fine-mapping For each locus of the pooled locus-trait associations, we used the matched LD reference to perform statistical fine-mapping by estimating the posterior probability (PP) of each variant to be a causal variant (PP > 0.8) using the probabilistic annotation integrator (PAINTOR) tool 32 , 33 with the Markov chain Monte Carlo (MCMC) model that allows multiple causal variants. In the 296 pooled locus-trait associations, one was significant in all three categories (EAS, EUR, and cross-ancestry) of GWASs, 36 significant only in EUR-GWASs, 36 significant only in cross-ancestry GWASs, 223 significant in both EUR and cross-ancestry GWASs ( Supplementary Table 10 ). We conducted statistical fine-mapping for one locus based on EAS-GWAS using EAS-LD reference, 260 loci based on EUR-GWASs using EUR-LD reference, and 260 loci based on cross-ancestry GWASs using merged LD reference, respectively. We found 250 unique causal variants for brainstem volumetric traits, of which 25 were identified only from EUR analyses, 223 only from cross-ancestry analyses, and two from both EUR and cross-ancestry analyses (Fig. 3 a and Supplementary Table 16 ). For the 223 locus-trait associations significant in both EUR and cross-ancestry GWASs, we also conducted PAINTOR with one causal variant assumption and reported the results in Supplementary Table 17 . We then used the Wilcoxon rank-sum test ( P < 0.05) to investigate whether cross-ancestry fine-mapping can reduce the 95% credible sets for these trait-associated loci compared to EUR-specific fine-mapping. We found that the 95% credible sets (median = 2) from cross-ancestry analyses were significantly smaller ( P = 2.20 × 10 − 16 ; Fig. 3 b) than those (median = 13) from EUR analyses. For example, although the locus-trait association between 2p11.2 and pons absolute volume was significant in all GWASs, we failed to identify any causal variants based on the EAS-GWAS and EUR-GWAS, but found two causal variants (rs2090217: PP = 1.00; rs335124: PP = 1.00) based on the cross-ancestry GWAS (Fig. 3 c). Functional annotations To investigate functional consequences of genetic variants associated with brainstem volumetric traits, we used FUMA 34 to perform functional annotations for the unique variants with PP > 0.1 in statistical fine-mapping. Based on the genomic location and functional consequence, we categorized the four variants with PP > 0.1 in the only locus-trait association identified by EAS-GWAS, and found one variant in the intronic region, two variants in the intergenic region, and one variant as a missense variant of EIF2AK3 ( Supplementary Table 18 ). We also categorized 2,076 unique variants with PP > 0.1 in the 260 locus-trait associations identified by EUR-GWASs, and found that the variants were mainly in the intronic (44.9%) and intergenic (37.7%) regions (Fig. 3 d and Supplementary Table 19 ). We also found 28 variants in the coding sequence and 48 variants in the untranslated region (UTR), of which three were stop gained and 12 were missense variants. For instance, rs2231142 (4q22.1) associated with brainstem ( P = 4.37 × 10 − 14 , PP = 0.16) and pons absolute ( P = 8.70 × 10 − 11 , PP = 0.18) volumes is a stop gained variant of ABCG2 , a regulator of self-renewal of neural stem cells 35 . As for the 3,944 unique genetic variants (PP > 0.1) in 260 locus-trait associations identified by cross-ancestry GWASs, they were also mainly in the intronic (46.0%) and intergenic (35.9%) regions (Fig. 3 d and Supplementary Table 20 ). We found 40 coding and 70 UTR variants, including two splice region variants within 2bp of the splicing junction, one stop-gained variant, and 24 missense variants. For the 5,846 unique variants with PP > 0.1, we also used combined annotation-dependent depletion (CADD) score 36 to assess deleteriousness of 5,844 variants included in the CADD database, from which we identified 255 pathogenic variants (CADD score > 12.37; Supplementary Tables 18–20) . We then used RegulomeDB (RDB) score 37 to identify variants with regulatory function, and found 112 most likely regulatory variants (RDB ≤ 1f; Supplementary Tables 18–20 ) from 5,152 fine-mapped variants included in the RDB database. Identifying genes associated with brainstem volumetric traits We identified genes associated with brainstem volumetric traits based on the location, gene expression, and chromatin interaction using the gene-based, transcriptome-wide, and chromatin interaction association analyses, respectively. Gene-based association analyses. We conducted the gene-based association analyses to identify genes associated with brainstem volumetric traits ( P < 0.05/17,550/9 = 3.16 × 10 − 7 , Bonferroni correction for 17,550 genes and nine traits) based on the EUR-GWASs using Multivariate Analysis of Genomic Annotation (MAGMA) 38 . We found 189 unique genes associated with brainstem volumetric traits, including 84 with whole brainstem volume, 49 with medulla absolute volume, 29 with medulla relative volume, 81 with pons absolute volume, 51 with pons relative volume, 51 with midbrain absolute volume, 29 with midbrain relative volume, ten with SCP absolute volume, and three with SCP relative volume ( Supplementary Fig. 7 and Supplementary Table 21 ). Transcriptome-wide association analyses. Based on the EUR-GWAS summary data of nine brainstem volumetric traits and eQTL data of 13 brain tissues provided by GTEx 39 , we used S-PrediXcan 40 to perform the transcriptome-wide association studies (TWASs) between predicted gene expression in each tissue and each brainstem volumetric trait ( Supplementary Fig. 8 ). Based on the gene-trait associations of 13 tissues, we then used S-MultiXcan 41 to conduct the multi-tissue TWAS to test the joint effects of gene expression on brainstem volumetric traits across 13 brain tissues ( P < 0.05/15,375/9 = 3.61 × 10 − 7 , Bonferroni correction for 15,375 genes and nine traits). We identified 151 unique genes whose brain expression was associated with brainstem volumetric traits, including 75 with whole brainstem volume, 37 with medulla absolute volume, 27 with medulla relative volume, 56 with pons absolute volume, 42 with pons relative volume, 39 with midbrain absolute volume, 19 with midbrain relative volume, 12 with SCP absolute volume, and one with SCP relative volume ( Supplementary Fig. 9 and Supplementary Table 22 ). Chromatin interaction association analyses. Hi-C-coupled MAGMA (H-MAGMA) 42 was used to identify the genes associated with brainstem volumetric traits based on the chromatin interaction profiles in six brain tissues and cells, including adult brain 43 , fetal brain 44 , cortical neuron 45 , induced pluripotent stem cells (iPSC) derived astrocyte 46 , iPSC derived neuron 46 , and midbrain DA neuron 47 . We identified 2,315 significant gene-trait associations ( P < 0.05/107,575/9 = 5.16 × 10 − 8 , Bonferroni correction for 107,575 genes for six tissues and cells and nine traits) ( Supplementary Fig. 10 and Supplementary Table 23 ). These associations contained 465 unique genes, including 231 with whole brainstem volume, 121 with medulla absolute volume, 73 with medulla relative volume, 165 with pons absolute volume, 132 with pons relative volume, 114 with midbrain absolute volume, 60 with midbrain relative volume, 17 with SCP absolute volume, and seven with SCP relative volume. Pooling genes from three association analyses. By integrating genes identified by the three approaches, we prioritized 550 unique genes, including 267 associated with whole brainstem volume, 154 with medulla absolute volume, 86 with medulla relative volume, 194 with pons absolute volume, 142 with pons relative volume, 134 with midbrain absolute volume, 75 with midbrain relative volume, 28 with SCP absolute volume, and eight with SCP relative volume (Fig. 4 a). Pathways associated with brainstem volumetric traits Based on pathways from Gene Ontology (GO) 48 biological processes and Reactome 49 , we performed pathway enrichment analyses for the prioritized 550 genes associated with brainstem volumetric traits using g:Profiler 50 ( https://biit.cs.ut.ee/gprofiler/gost ). We identified 138 significant enrichment pathways ( P c < 0.05, g:SCS (Set Counts and Sizes) corrected; Fig. 4 b and Supplementary Table 24 ), mainly including metabolic process regulation, animal organ development, activity kinase regulation, neurogenesis, neuron generation, cell proliferation, and neuron differentiation regulation. Shared and distinct genetic architectures between brainstem substructures Although sharing some white matter tracts, the three brainstem substructures (medulla, pons, and midbrain) are originated from different embryonic structures (medulla and pons from hindbrain and midbrain from mesencephalon) and contain different gray matter nuclei and white matter tracts, indicating the coexistence of shared and distinct genetic architectures. Thus, we investigated the shared and distinct genetic loci, genes, and enriched pathways between each pair of the three brainstem substructures. Shared and distinct genetic loci between brainstem substructures. Among the 122 LD-independent loci associated with brainstem volumetric traits, 111 were associated with the absolute or relative volume of medulla, pons, or midbrain. From the 111 loci, we searched for loci that were associated with volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume based on the independent locus-trait associations. We identified 62 substructure-specific loci (17 for medulla, 27 for pons, and 18 for midbrain) and 49 substructure-shared loci (20 shared by all substructures, six by medulla and pons, five by medulla and midbrain, and 18 by pons and midbrain) (Fig. 5 a and Supplementary Table 25 ). For example, we found a locus (rs9428966) at 1q43 shared by all brainstem substructures, a medulla-specific locus (rs12448813) at 16q23.2, a pons-specific locus (rs72927168) at 2q31.1, and a midbrain-specific locus (rs2331753) at 1q25.3 (Fig. 5 b). Shared and distinct genes between brainstem substructures. Among the prioritized 550 genes associated with brainstem volumetric traits, 468 were associated with the absolute or relative volume of medulla, pons, or midbrain. From the 468 prioritized genes, we searched for genes prioritized for one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume. We identified 295 substructure-specific genes (100 for medulla, 123 for pons, and 72 for midbrain) and 173 substructure-shared genes (64 shared by all substructures, 42 shared by medulla and pons, 13 shared by medulla and midbrain, and 54 shared by pons and midbrain) (Fig. 5 c, Supplementary Table 26 ). For example, UBE4B was a medulla-specific gene involving axon regrowth 51 ; FOXO6 was a pons-specific gene involving the regulation of synaptic function 52 ; and FRAT1 and FRAT2 were midbrain-specific genes affecting midbrain morphogenesis by regulating WNT signaling 53 . Shared and distinct biological pathways between brainstem substructures. We pooled the prioritized genes for medulla, pons, and midbrain regardless of absolute or relative volume, and obtained 219 genes for medulla volume, 283 for pons volume, and 203 for midbrain volume. We then used g:Profiler 50 to conduct pathway enrichment analyses for the three groups of genes, respectively. We found 90 unique enrichment pathways ( P c < 0.05, g:SCS corrected), including 33 pathways for medulla, 85 for pons, and one for midbrain volumes ( Supplementary Table 27 ). Then, we searched for the pathways enriched by one, two, or three brainstem substructures, and found 62 substructure-specific pathways (five for medulla and 57 for pons) and 28 substructure-shared pathways (one shared by all substructures and 27 shared by medulla and pons; Fig. 5 d) ( Supplementary Table 28 ). The pathway shared by all substructures was the regulation of developmental process. A medulla-specific pathway was the noncanonical activation of NOTCH3, which regulates neurogenesis and neuronal differentiation in hindbrain 54 . Two pons-specific enrichment pathways contained HOX , involving in the regulation of pontine neuronal migration 55 . Genetic architectures shared by brainstem volumetric traits and other phenotypes Based on the available GWAS summary statistics for non-imaging phenotypes that are potentially associated with brainstem functions, we used three complementary methods (genetic correlation, genetic colocalization, and condFDR/conjFDR) to identify shared genetic architectures between brainstem volumetric traits and non-imaging phenotypes. As most GWASs are conducted in EUR population and most approaches require GWAS samples from same ancestry, we included 26 non-imaging phenotypes ( Supplementary Table 29 ) with EUR-GWAS summary statistics and conducted the three genetic sharing analyses based on the EUR-GWAS summary data for brainstem volumetric traits. Genetic correlation analyses . We used LDSC 24 , 56 to calculate the genetic correlations between nine brainstem volumetric traits and 26 non-imaging phenotypes based on the EUR-GWAS summary data. Using a Bonferroni-corrected P < 0.05/26/9 = 2.14 × 10 − 4 , we identified six significant genetic correlations between brainstem volumetric traits and non-imaging phenotypes (Fig. 6 a and Supplementary Table 30 ), including genetic correlations of whole brainstem volume (r = -0.13, P = 1.85 × 10 − 6 ), midbrain absolute volume (r = -0.18, P = 1.59 × 10 − 8 ), and midbrain relative volume (r = -0.13, P = 4.58 × 10 − 6 ) with ADHD, whole brainstem volume (r = -0.08, P = 1.00 × 10 − 4 ) and midbrain absolute volume with MDD (r = -0.11, P = 1.60 × 10 − 5 ), and midbrain absolute volume with PD (r = 0.16, P = 2.00 × 10 − 4 ). Genetic colocalization analyses. We used coloc 57 to identify the genetic loci shared by brainstem volumetric traits and non-imaging phenotypes based on the EUR-GWAS summary data. Among the 260 loci associated with brainstem volumetric traits in EUR-GWASs ( P < 5.56 × 10 − 9 ), we found genetic colocalization (PP.H4 > 0.8) between eight brainstem volumetric traits and ten non-imaging phenotypes at 59 genetic loci (19 LD-independent loci) (Fig. 6 b and Supplementary Table 31 ), including the non-imaging phenotypes of brain disorders (MDD, SCZ, and AD), cardiovascular functions (resting heart rate, systolic and diastolic blood pressure, and pulse pressure), circadian rhythms (chronotype and morning person), and subjective well-being. For instance, the two circadian rhythms phenotypes showed colocalizations (PP.H4 = 0.94–0.99) at 19p13.11 with whole brainstem volume, medulla absolute volume, medulla relative volume, and pons relative volume. SCZ showed colocalizations at 2q33.1 with whole brainstem volume (PP.H4 = 0.89) and medulla absolute volume (PP.H4 = 0.84). The lead SNPs in the locus are eQTLs of TYW5 , a regulator of neurodevelopment 58 . CondFDR/conjFDR analyses. Based on the EUR-GWAS summary data, we conducted condFDR/conjFDR analyses using pleioFDR ( https://github.com/precimed/pleiofdr ) 59 to identify genetic variants shared by each pair of nine brainstem volumetric traits and 26 non-imaging phenotypes. We first generated the conditional Q-Q plots to assess the polygenetic enrichment for each brainstem volumetric trait conditioned on P -values of the association with each non-imaging phenotype. Among the 234 trait-phenotype pairs, 116 pairs showed polygenetic enrichment with successive leftward shifts from the null distribution ( Supplementary Fig. 11 ), involving eight brainstem volumetric traits and 23 non-imaging phenotypes. Then, we performed the conjFDR analysis to identify the shared genetic variants (conjFDR < 0.05) between each trait-phenotype pair, in which the conjFDR value of each variant was defined as the maximal FDR value of this variant in the two mutual condFDR analyses. Among the 234 potential trait-phenotype pairs, 201 pairs showed shared genetic variants (conjFDR < 0.05; Supplementary Fig. 12 ), involving nine brainstem volumetric traits and 25 non-imaging phenotypes. These trait-phenotype pairs shared 7,263 variants and 5,826 loci (Fig. 6 c and Supplementary Table 32 ). Numerous genetic sharing findings were consistent with genetic correlation and/or colocalization analyses. For example, both conjFDR and colocalization analyses demonstrated that the locus (7q21.2) was shared between brainstem volumetric traits (absolute volumes of medulla and midbrain) and blood pressure phenotypes (systolic blood pressure and pulse pressure) (Fig. 6 d). Its lead SNP rs42039 is located in the UTR3 region of CDK6 , a key player in cell cycle progression 60 . We also found genetic sharing between brainstem volumetric traits and 13 non-imaging phenotypes that were observed in neither genetic correlation nor genetic colocalization analyses. Discussion By including all available neuroimaging genetics data, especially those from the ABCD and CHIMGEN studies, we conducted cross-ancestry GWAS meta-analyses for whole brainstem volume in 72,717 individuals and substructure volumes in 48,522 individuals. We identified 122 loci and 550 genes associated with nine brainstem volumetric traits, including 46 new loci. We discovered three (0.8%) ancestry-specific and 292 (79.1%) ancestry-shared associations, consistent with high cross-ancestry genetic correlations between EAS and EUR. We provided new evidence for the merit of cross-ancestry fine-mapping by identifying 225 causal variants (PP > 0.8), which are much greater than 27 from EUR fine-mapping. We revealed shared and distinct loci, genes, and pathways for midbrain, pons, and medulla volumes, and shared genetic architectures of the brainstem volumetric traits with 25 brainstem-related physiological and pathological phenotypes, in line with the importance of the brainstem in the control of heart rate, blood pressure, circadian rhythms, as well as its associations with neuropsychiatric disorders, such us AD, ADHD, MDD, PD, and SCZ. The first contribution of this study is the discovery of nearly two times of variant-trait and locus-trait associations compared to those significant at P < 5.56 × 10 − 9 in the previous GWASs for brainstem volumetric traits 12 – 16 . Even compared to associations significant at P < 5 × 10 − 8 in previous GWASs, we still found 241 new variant-trait and 150 new locus-trait associations and 46 new LD-independent loci. The large number of new genetic discoveries may improve the understanding of the genetic architectures of brainstem and substructure volumes. For instance, in the cross-ancestry GWAS meta-analysis for pons absolute volume, we found a new locus-trait association at 20q11.22. Its lead variant rs3213141 is in the upstream of E2F1 , a cell cycle suppressor regulating neuronal survival and death by interacting with protein product of the retinoblastoma (RB) gene 61 , 62 . RB1-deficient mutation can lead to neuronal apoptosis in the hindbrain 63 including the pons as a main component. Another new locus-trait association was found between 12q13.13 and pons relative volume, its lead variant rs56098072 is an eQTL of HOX3 , interacting with other HOX genes to specify the rhombomere identity in the developing hindbrain 64 . We also found a locus at Xp11.22, the only new locus on X-chromosome, which was associated with whole brainstem and pons absolute volumes. Its lead variant rs7060542 is an eQTL of the CASK-interacting nucleosome assembly protein ( CINAP ), influencing brain development by binding to CASK 65 , while CASK mutation can lead to an X-linked brain malformation with brainstem hypoplasia 66 . The second contribution is the novel genetic findings derived from cross-ancestry analyses. We revealed high cross-ancestry genetic correlations (0.44–1.13) in brainstem volumetric traits between EUR and EAS populations, consistent with 97 times more ancestry-shared associations (n = 292) than ancestry-specific associations (n = 3). These results indicate that EAS and EUR individuals have similar genetic architectures for brainstem and substructure volumes. We also conducted statistical fine-mapping for the locus-trait associations of brainstem volumetric traits, and found 250 causal variants, among which 223 (89.2%) were only identified by cross-ancestry fine-mapping. These results further highlight the value of cross-ancestry fine-mapping in detecting causal variants for brain imaging phenotypes 67 . For instance, rs17010085 was identified as a causal variant (PP = 0.96-1.00) for whole brainstem volume and pons absolute volume in both EUR and cross-ancestry fine-mapping. The variant is an eQTL of RYBP , which is involved in neural differentiation 68 . The third contribution of this study is the discovery of shared and distinct genetic architectures (loci, genes, and pathways) between brainstem substructure (midbrain, pons, and medulla) volumes. We found 20 loci and 64 genes shared by all substructures, mainly involving neural development processes shared by brain structures. For example, we found a locus at 1q43 shared by all brainstem substructures. Its lead SNP rs9428966 is in the UTR3 region of AKT3 , an AKT kinase involving a wide variety of biological processes, including brain development 69 , neuronal survival 70 , myelination 71 , and the regulation of blood pressure 72 and breathing 73 . We also found 62 loci and 295 genes that were specific to one brainstem substructure, providing more specific insight into the genetic architecture of each brainstem substructure. For instance, a locus at 16q23.2 was a medulla-specific locus and its lead variant rs12448813 is an eQTL of the giant axonal neuropathy (GAN) gene. GAN encodes gigaxonin, affecting neuronal survival by regulating the degradation of the light chain of microtubule-associated protein 1B (MAP1B) 74 . The disorganization of the neurofilament network due to GAN mutation is associated GAN 75 . These findings are consistent with the compact axon arrangement in the medulla. Among the 33 enrichment pathways for medulla volume, 28 (84.9%) were also enriched for pons volume, in line with their shared developmental origin from the hindbrain. The last contribution is the identification of shared genetic architectures between brainstem volumetric traits and brainstem-related non-imaging phenotypes. A previous study has conducted genetic correlation and conjunctional FDR analyses to investigate the shared genetic architectures between brainstem volumetric traits and eight common brain disorders 12 , but failed to identify any significant genetic correlations after multiple testing correction, although overlapped genetic loci are observed for all eight disorders. We extended the study by including 26 brainstem-related non-imaging phenotypes and by further conducting genetic colocalization analyses, generating many novel findings. We found genetic correlations of midbrain volume with PD, MDD and ADHD, all of which are linked to abnormal DA levels. The findings agree with the role of midbrain DA neurons in the regulation of voluntary movement, reward, salience, motivation, and emotion 5 , 6 . In addition to genetic colocalization with brain disorders (MDD, SCZ, and AD), we also identified genetic colocalization between brainstem volumetric traits and physiological phenotypes (heart rate, blood pressure, chronotype, and morning person) associated with cardiovascular function and circadian rhythm, well-known functions of the brainstem 2 , 3 , 76 . The shared genetic architectures with physiological phenotypes were further confirmed by the conjunctional FDR analyses, in which brainstem volumetric traits shared more loci with physiological phenotypes than any other phenotypes. Several limitations should be noted when interpreting our findings. First, although we included 7,069 Chinese Han individuals from the CHIMGEN study, the sample size of EAS individuals is much smaller than that of EUR individuals, which may bias the cross-ancestry GWASs for brainstem volumetric traits and the effect size comparisons between EAS and EUR populations. More non-EUR individuals from EAS and other ancestral populations should be included in the future cross-ancestry GWASs. Second, despite we controlled for the effect of age in GWASs, we cannot exclude the bias from the age differences among the participants from ABCD (aged 8–11), CHIMGEN (aged 18–30) and UKBB (aged 40–70). Third, we reported the study-wide significant genetic associations ( P < 5.56 × 10 − 9 ) for brainstem volumetric traits, but we cannot completely exclude the false-positive findings due to lacking independent data replication. Methods Participants and data preparation for GWASs In GWASs for brainstem volumetric traits, we included three sets of raw neuroimaging genetics datasets (CHIMGEN, ABCD, and UKBB) and one set of GWAS summary data for whole brainstem volume 14 from ENIGMA, CHARGE, and UKBB datasets. In line with prior GWASs for brainstem substructure volumes 12 , 13 , we used the same Bayesian approach 8 to obtain the absolute volumes of whole brainstem, medulla, pons, midbrain, and superior cerebellar peduncle (SCP) of CHIMGEN, ABCD, and UKBB participants. The obtained whole brainstem (including SCP) volume was included as an additional covariate to perform GWASs for relative volumes of medulla, pons, midbrain, and SCP. As the previous GWAS 14 used the automatic subcortical segmentation method to obtain whole brainstem (not including SCP) volume, we also applied the same approach to calculate whole brainstem volume for each CHIMGEN, UKBB, or ABCD participant, which was used in GWASs for whole brainstem volume throughout our study. As the previous GWAS for whole brainstem volume 14 included first released UKBB data, we excluded these participants and reperformed the UKBB-GWAS for whole brainstem volume, and utilized the obtained GWAS summary statistics in the meta-analyses with the prior study 14 . The number of participants included in each GWAS are shown in Supplementary Table 3 . CHMGEN participants and data preparation . All the EAS participants were recruited from the CHIMGEN study ( http://chimgen.tmu.edu.cn/ ), which collected genomic and neuroimaging data from 7,306 healthy Chinese Han participants aged 18–30 years from 32 centers using the predefined inclusion and exclusion criteria ( Supplementary Table 33 ). The CHIMGEN study was approved by the Medical Research Ethics Committees of Tianjin Medical University General Hospital and all other institutions, and written informed consent was obtained from each participant. Among the 7,306 participants, 7,195 participants with DNA samples were genotyped by Illumina ASA-750K (Asian Screening Array) that was specially designed for Asian individuals. PLINK v2.084 77 ( http://www.cog-genomics.org/plink2 ) was applied for quality control of genetic data. Details for sample-level and variant-level quality control, principal component analysis (PCA), and genetic data imputation are provided in our prior studies 16 , 67 . After quality control, we included 7,163 participants and 8,790,144 imputed autosomal and 227,168 X-chromosomal bi-allelic variants (MAF ≥ 0.5%, info ≥ 0.6, and P ≥ 1 × 10 − 7 in Hardy–Weinberg equilibrium (HWE)) in EAS-GWASs. The brain structural MRI data were acquired by ten types of 3.0-Tesla MRI scanners and 12 sets of scanning parameters ( Supplementary Table 34 ). After excluding 61 participants without qualified structural MRI data, we calculated the volumes of whole brainstem, medulla, pons, midbrain, and SCP for the remaining 7,102 participants using the Bayesian segmentation algorithm 8 implemented in FreeSurfer v7.0 ( https://surfer.nmr.mgh.harvard.edu ). With FreeSurfer v7.0, we also used the automatic subcortical segmentation method to obtain the whole brainstem volume (not including SCP). For each of the five brainstem volumetric traits, we removed the participants with volumes greater than five times the median absolute deviation (MAD) from the median value, and finally included 7,094 − 7,096 participants in EAS-GWASs. The quality control procedures for CHIMGEN data are presented in Supplementary Fig. 13 . ABCD participants and data preparation. The ABCD study ( https://abcdstudy.org/ ) is a publicly available longitudinal dataset containing over 10,000 participants aged 9–10 years at their baseline assessment from 21 research centers. Procedures of the study in most research centers were approved by a central Institutional Review Board (IRB) at the University of California, San Diego, and by local IRB in a few research centers 78 . All parents or caregivers provided the written informed consent and children provided the written assent. We accessed to the data under application ID 17607. From the 11,099 participants with qualified imputed genotype data, we included 6,605 EUR participants whose genetic ancestry was largely (> 80%, a previously recommended threshold 79 , 80 ) European estimated by SNPweights v2.1 81 based on the SNP weights for European, West African, East Asian, and Native American populations. We applied MAF ≥ 0.5%, imputation quality r 2 > 0.3, and P HWE ≥ 1 × 10 − 7 to filter the variants and transformed them from GRCh38/hg38 to GRCh37/hg19 to be consistent with the genetic data from CHIMGEN and UKBB. In the ABCD-GWASs, we included 9,064,819 autosomal and 278,985 X chromosomal bi-allelic variants. Among the 6,605 participants, 6,060 had qualified whole brainstem (not including SCP) volume data obtained by automatic subcortical segmentation using FreeSurfer v7.0. Of the 6,060 participants, 5,815 had raw brain structural MRI data, from which we calculated volumes of whole brainstem, medulla, pons, midbrain, and SCP using the Bayesian segmentation algorithm 8 in FreeSurfer v7.0. After removing the participants with volumes greater than five times MAD from median, we included 5,804-6,060 participants ( Supplementary Table 3 ) in ABCD-GWASs. The detailed quality control procedures for ABCD data are presented in Supplementary Fig. 14 . UKBB participants and data preparation. Most of the included EUR participants were recruited from the UKBB study ( https://www.ukbiobank.ac.uk/ ) 82 , which collected approximately 500,000 participants aged 40–69 years at recruitment from 22 research centers across the United Kingdom. The UKBB study was approved by the National Health Service (NHS) Research Ethics Service (21/NW/0157), and written informed consent was obtained from each participant. We accessed to the data under application number 75556. After initial genetic data quality control and imputation, the remaining 487,207 participants were included in the further sample-level quality control. After excluding 651 participants with sex chromosome aneuploidy, 186 with sex mismatch, and 78,257 non-Caucasians, we then included 408,113 EUR participants with qualified genomic data. Using the filtering criteria of MAF ≥ 0.5%, info ≥ 0.6, and P HWE ≥ 1 × 10 − 7 , we finally included 10,042,001 autosomal and 390,192 X-chromosomal bi-allelic variants in UKBB-GWASs. Among the 408,113 participants, 36,533 had the volumetric data of brainstem and substructures obtained by the two segmentation approaches using FreeSurfer 7.0. We then visually checked raw brain structural images and the brainstem segmentation images, and further excluded 914 participants with brain tumors, imaging artifacts, incomplete brainstem coverage, or incorrected brainstem segmentation. After further removing participants with volumes greater than five times MAD from median, we finally included 35,521 − 35,611 participants ( Supplementary Table 3 ) in UKBB-GWASs. The quality control procedures for UKBB data are shown in Supplementary Fig. 15 . Reproducibility of brainstem segmentation In 7,096 CHIMGEN, 5,815 ABCD, and 35,611 UKBB participants, we calculated the intraclass correlation coefficient (ICC) of whole brainstem volumes obtained from the two brainstem segmentation methods, respectively. Although the automatic subcortical segmentation generated whole brainstem volume not including SCP and the Bayesian segmentation generated whole brainstem volume including SCP, we found high ICCs (0.956–0.997; Supplementary Table 1 ) in whole brainstem volumes obtained by the two brainstem segmentation methods. Among these participants, 24 CHIMGEN, 4,077 ABCD, and 2,698 UKBB participants had brain structural MRI data acquired at two time points. For each participant, we calculated the brainstem and substructure volumes using the Bayesian brainstem segmentation based on the MRI data acquired at the two time points. For each dataset, we calculated the ICC of each volumetric trait obtained from the two time points to assess the test-retest reliability of the Bayesian brainstem segmentation, and found high ICCs (0.868–0.992; Supplementary Table 1 ). Harmonization and normalization of brainstem and substructure volumes The brain structural MRI data of the CHIMGEN, ABCD, and UKBB participants were acquired by different MRI scanners, which may bring bias to the integrated analyses of brainstem and substructure volumetric data from multiple centers. To remove the bias, for each brainstem volumetric trait from each dataset, the Combat method was used to harmonize the volume data calculated based on MRI data acquired by different scanners, which can remove between-scanner variation and preserve biological variability 83 . We tested the effect of ComBat harmonization in two participants who traveled to different centers and were scanned at 28 MRI scanners. In each participant, we segmented and calculated the five brainstem and substructure volumes based on the MRI data acquired from each scanner and used the coefficient of variation (CV) to assess between-scanner variations of these traits. In the two participants, we found that CVs of these volumetric traits before harmonization significantly reduced (Wilcoxon rank-sum test: P = 0.018) after harmonization ( Supplementary Fig. 16 ). As the skewed data distribution would violate the assumption of normal distribution when using linear regression model to perform GWASs, quantile normalization was applied to the harmonized brainstem and substructure volumetric data. Covariates for GWASs We controlled for age at imaging, genetic-determined sex, age × sex, total intracranial volume (TIV), and first genetic principal components (PCs) in CHIMGEN-GWASs, ABCD-GWASs, and UKBB-GWASs, and further controlled for genotyping batches in ABCD-GWASs and UKBB-GWASs. For each participant, TIV was also estimated by FreeSurfer 7.0 and followed ComBat harmonization and quantile normalization. We controlled for the first ten genetic PCs in CHIMGEN-GWASs, 32 in ABCD-GWASs, and 40 in UKBB-GWASs, which were selected based on the population complexity. GWASs for brainstem and substructure volumes GWASs for single dataset. We used the mixed linear model (MLM) from fastGWA 22 to conduct GWASs (additive effect) with the predefined covariates for absolute volumes of brainstem, medulla, pons, midbrain, and SCP at both autosomal and X-chromosomal variants in CHIMGEN, ABCD, and UKBB participants, respectively. Considering that brainstem substructure volumes were highly correlated with whole brainstem volume (r = 0.556–0.982; Supplementary Table 35 ), we also performed GWASs for relative volumes of medulla, pons, midbrain, and SCP by additionally accounting for the whole brainstem volume, which can reveal genetic signals beyond commonality in volume 12 , 84 . The participants and genetic variants on both autosomes and X-chromosome included in each GWAS are presented in Supplementary Table 3 . As the fastGWA cannot output summary statistics for all input variants, only variants with output summary statistics ( Supplementary Table 3 ) were included in the subsequent GWAS meta-analyses. Due to lacking independent replication, we reported the study-wide significant associations ( P < 5.56 × 10 − 9 , Bonferroni corrected for the nine traits) for all GWASs. EUR-GWAS meta-analyses. Based on the summary statistics of ABCD-GWASs and UKBB-GWASs for eight brainstem substructure volumes, we used the inverse variance weighted (IVW) fixed effect model in METAL 23 to conduct EUR-GWAS meta-analyses for the eight brainstem volumetric traits. For each trait, in addition to the identification of new genetic associations by increasing the sample size, the meta-analysis could also output summary statistics for variants included in either of the two GWASs, providing the largest number of EUR-GWAS association statistics. In EUR-GWAS meta-analysis for whole brainstem volume, to make full use of the available data resources, we also included the GWAS summary data for whole brainstem volume (n = 28,809) from ENIGMA, CHARGE, and UKBB (first release) datasets 14 . As the GWAS included first released UKBB data, we then excluded these participants and reperformed GWAS for whole brainstem volume in 30,752 UKBB participants. Based on the GWAS summary data of UKBB-GWAS (n = 30,752), ABCD-GWAS (n = 6,060), and previous EUR-GWAS (n = 28,809) 14 , we conducted EUR-GWAS meta-analysis for whole brainstem volume. The participants and genetic variants included in EUR-GWAS meta-analyses for brainstem volumetric traits are presented in Supplementary Table 3 . Cross-ancestry GWAS meta-analyses. Based on the obtained summary statistics from CHIMGEN-GWASs and EUR-GWAS meta-analyses, we conducted the cross-ancestry GWAS meta-analyses for these nine brainstem volumetric traits using the IVW fixed effect model in METAL 23 . The participants and genetic variants included in the cross-ancestry GWAS meta-analyses are also presented in Supplementary Table 3 . Population stratification estimation. For each GWAS, we used the genomic control inflation factor (λ GC ) and linkage disequilibrium score regression (LDSC) intercepts 24 to estimate population stratification. λ GC was calculated as the median of the resulting chi-squared (χ 2 ) test statistics (z scores) divided by 0.4549, the expected median of the χ 2 distribution with one degree of freedom. As high λ GC indicates either genomic inflation or polygenicity, we used LDSC intercept to identify genomic inflation based on LD scores. Defining LD references and independent associations and loci We used imputed genotype data from 7,096 CHIMGEN and 35,611 UKBB participants with qualified genetic and brainstem volumetric data to construct EAS-LD and EUR-LD references, respectively. We also constructed a cross-ancestry LD reference using the sample-weighted method based on the two datasets. For each brainstem volumetric trait, based on the summary statistics of CHIMGEN, ABCD, UKBB, EUR, or cross-ancestry GWAS, we used the matched LD reference to identify independent variant-trait associations by PLINK clumping 85 with the following steps: (1) all significant variants were included in a list of candidate variants; (2) the most significant variant was defined as the first lead variant (independent variant), and variants within 500 kb from and in LD with (r 2 > 0.1) the lead variant were clumped; (3) the remaining variants formed a new list of candidate variants, and then step (2) was repeated; and (4) the iterative process stopped until the list was empty. We identified independent locus-trait associations by: (1) creating loci for independent variants by adding 500 kb to both sides of each variant; (2) merging loci within 500kb; (3) merging loci if any independent variant of one locus was in LD (r 2 > 0.1) with any independent variant of another locus; and (4) merging loci overlapped with the major histocompatibility complex (MHC) or 8p23.1 region into one locus. The associations of the remaining loci with this trait were defined as independent locus-trait associations. As EUR samples were at least six times larger than EAS samples in this study, for each trait, we used the EUR-LD reference to pool GWAS results from the five different categories of GWASs. We used the above-mentioned strategies to identify independent variant-trait and locus-trait associations for each trait. We identified LD-independent loci by merging the loci of the pooled locus-trait associations within 500kb or their lead variants with LD (r 2 > 0.1). As the prior GWASs for brainstem and substructure volumes 12 – 16 were conducted mainly in EUR participants, we used the same strategies and EUR-LD reference to pool these GWASs to identify known independent variant-trait and locus-trait associations and LD-independent loci using two significance thresholds ( P < 5 × 10 − 8 and P < 5.56 × 10 − 9 ), respectively. Using EUR-LD reference and each known list of associations and loci, we defined a new variant-trait association if the variant was 500 kb away from and not in LD (r 2 < 0.1) with any variants of the same trait in the list of known variant-trait associations; a new locus-trait association when the locus was 500 kb away from loci of all known locus-trait associations and all lead variants in the locus were not in LD (r 2 < 0.1) with any lead variants in the loci of all known locus-trait associations; and a novel locus when the locus was 500 kb away from all known loci and all lead variants in the locus were not in LD (r 2 < 0.1) with any lead variants in known loci. SNP-based heritability and cross-ancestry genetic correlation We used LDSC 24 to estimate the SNP-based heritability for brainstem volumetric traits in EAS and EUR based on the ancestry-specific LD reference and the GWAS summary statistics for autosomal variants. We applied the covariate-adjusted LDSC 29 method to estimate the cross-ancestry SNP-based heritability for these traits based on the cross-ancestry LD reference and GWAS summary statistics for autosomal variants, while adjusting for 40 genetic PCs derived from both CHIMGEN and UKBB genetic data. After excluding variants in the MHC region, based on EAS-LD and EUR-LD references and cross-ancestry GWAS summary statistics for autosomal variants, we used Popcorn ( https://github.com/brielin/Popcorn ) 30 to estimate the cross-ancestry genetic-effect (not considering MAF information) and genetic-impact (considering MAF information) correlations between EAS and EUR for each brainstem volumetric trait. Allele effect heterogeneity between EAS and EUR We used Cochran’s Q test (CQ-test) to assess the allelic effect heterogeneity between EAS and EUR by comparing the differences in effect sizes of candidate variants derived from EAS-GWASs and EUR-GWASs. Cochran’s Q was estimated based on the χ 2 test with a null hypothesis that homogeneity existed between the two ancestral populations and an alternative hypothesis that heterogeneity existed between ancestries. We defined the associations with CQ-test P ≥ 0.05 as ancestry-shared associations and those with P c < 0.05 (Bonferroni correction for the total number of associations tested) as ancestry-specific associations. The candidate variants were those included in the pooled variant-trait associations ( P < 5.56 × 10 − 9 ) from EAS, EUR, and cross-ancestry GWASs. Using UKBB-GWASs to represent EUR-GWASs, we conducted CQ-test for the variants of the pooled variant-trait associations included in GWAS summary statistics from both CHIMGEN and UKBB. The ancestry-shared and ancestry-specific associations were validated by replacing UKBB-GWASs by ABCD-GWASs. A successful verification for ancestry-specific associations was defined as P c < 0.05 (Bonferroni correction for the total number of the discovered ancestry-specific associations). Statistical fine-mapping For each locus of the pooled locus-trait association, we used the ancestry-matched LD references and GWAS summary data to perform statistical fine-mapping by estimating the posterior probability (PP) of each variant to be a causal variant (PP > 0.8) using the probabilistic annotation integrator (PAINTOR) tool 32 , 33 with the Markov chain Monte Carlo (MCMC) model that allows multiple causal variants. We conducted fine-mapping using the ancestry-specific LD reference for pooled locus-trait associations significant only in EAS-GWASs or EUR-GWASs, and the cross-ancestry LD reference for pooled locus-trait associations significant only in cross-ancestry GWASs. When a locus-trait association was significant in two or more categories of GWASs, we performed fine-mapping using matched LD references, respectively. For pooled locus-trait associations significant in both EUR and cross-ancestry GWASs, we also conducted PAINTOR with one causal variant assumption and used the Wilcoxon rank-sum test to assess whether cross-ancestry fine mapping can reduce the 95% credible sets for these loci compared to EUR-specific fine mapping. Functional annotations We used FUMA 34 to perform functional annotations for unique variants with PP > 0.1 in statistical fine-mapping for the pooled locus-trait associations identified by EAS-GWASs, EUR-GWASs, and cross-ancestry GWASs, respectively. We categorized these variants based on genic position, such as exon, intron, untranslated region (UTR), and intergenic region. We used the CADD score to prioritize deleterious and pathogenic variants, and a variant was considered as pathogenic when the CADD score was above 12.37 36 . We also used the RegulomeDB score to prioritize the variants in non-coding regions with classification scheme based on known and predicted regulatory elements 37 . Identifying genes associated with brainstem volumetric traits Based on the EUR-GWAS summary data for brainstem volumetric traits, we identified genes associated with these traits using gene-based, transcriptome-wide, and chromatin interaction association analyses, respectively. Gene-based association analyses. We mapped the variants included in the EUR-GWAS summary statistics for brainstem volumetric traits to 17,550 protein-coding genes based on location. We then conducted gene-based association analyses to identify the genes associated with brainstem volumetric traits ( P < 0.05/17,550/9 = 3.16 × 10 − 7 , Bonferroni correction for 17,550 genes and nine traits) based on the EUR-GWAS summary data and the EUR-LD reference from 1000 Genomes Project using Multivariate Analysis of Genomic Annotation (MAGMA) 38 . Transcriptome-wide association analyses. Based on the EUR-GWAS summary data of nine brainstem volumetric traits and eQTL data of 13 brain tissues provided by GTEx ( https://predictdb.org/ ) 39,86,87 , we used S-PrediXcan 40 to perform the transcriptome-wide association study (TWAS) between the predicted gene expression in each tissue and each brainstem volumetric trait. Based on the identified gene-trait associations of 13 brain tissues, we then used S-MultiXcan 41 to conduct multi-tissue TWAS to test the joint effects of gene expression on these brainstem volumetric traits across the 13 brain tissues ( P < 0.05/15,375/9 = 3.61 × 10 − 7 , Bonferroni correction for 15,375 genes under consideration and nine traits). Chromatin interaction association analyses. We used the Hi-C-coupled MAGMA (H-MAGMA) 42 , 88 to identify genes associated with brainstem volumetric traits based on the chromatin interaction profiles in six brain tissues and cells, including adult brain 43 , fetal brain 44 , cortical neuron 45 , induced pluripotent stem cells (iPSC) derived astrocyte 46 , iPSC derived neuron 46 , and midbrain dopamine (DA) neuron 47 . By mapping variants of EUR-GWASs for nine brainstem volumetric traits to protein-coding genes included in the variant-gene annotation files of the six brain tissues and cells based on the EUR-LD reference derived from the 1000 Genomes Project, we obtained 17,938 genes for adult brain, 17,955 for fetal brain, 17,952 for cortical neuron, 17,926 for astrocyte and 17,931 for neuron derived from iPSC, and 17,873 for midbrain DA neuron. We conducted H-MAGMA to identify the genes associated with these brainstem volumetric traits at P < 0.05/107,575/9 = 5.16 × 10 − 8 (Bonferroni correction for the nine brainstem volumetric traits and the total number of protein-coding genes for six brain tissues and cells). Prioritized genes and pathways We pooled significant genes in MAGMA, S-MultiXcan, and H-MAGMA analyses, and considered them as prioritized genes associated with brainstem volumetric traits, which were then submitted to g:Profiler 50 ( https://biit.cs.ut.ee/gprofiler/gost ), a web server for functional enrichment analysis, to perform pathway enrichment analyses based on pre-specified pathways from GO 48 (15,472 biological processes) and Reactome 49 (2,562 terms) databases. We corrected for multiple testing using the method ( P c < 0.05, g:SCS corrected) recommended by the g:Profiler tool. The significant GO biological processes and Reactome terms were visualized as a graph using Cytoscape 89 , 90 . Shared and distinct genetic architectures between brainstem substructures Shared and distinct genetic loci between brainstem substructures. From the pooled LD-independent loci associated with brainstem volumetric traits, we searched for loci that were associated with volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume based on the independent locus-trait associations. By this way, we identified medulla-, pons-, and midbrain-specific loci, as well as loci shared by any two substructures or all three substructures. Shared and distinct genes between brainstem substructures. From the prioritized genes for brainstem volumetric traits, we searched for genes prioritized for volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume. Thus, we identified medulla-, pons-, and midbrain-specific genes, as well as genes shared by any two substructures or all three substructures. Shared and distinct enrichment pathways between brainstem substructures. Based on the prioritized genes for medulla, pons, and midbrain regardless of absolute or relative volume, we used the above-mentioned method to perform pathway enrichment analyses for genes associated with medulla, pons, and midbrain volumes, respectively. Then, we searched for pathways enriched by volume(s) of one, two, or three substructures to identify the medulla-, pons-, and midbrain-specific pathways, and pathways shared by any two substructures or all three substructures. Genetic architectures shared by brainstem volumetric traits and other phenotypes Based on the available GWAS summary statistics for non-imaging phenotypes that are potentially associated with brainstem functions, we used three complementary methods (genetic correlation, genetic colocalization, and condFDR/conjFDR) to identify shared genetic architectures between brainstem volumetric traits and non-imaging phenotypes. As GWASs for both traits and phenotypes are mainly conducted in the EUR population and most approaches require GWAS samples from the same ancestry, we only included 26 non-imaging phenotypes ( Supplementary Table 29 ) with EUR-GWAS summary statistics and conducted the three genetic sharing analyses based on the EUR-GWAS summary data and the EUR-LD reference. Genetic correlation analyses . We used LDSC 24 , 56 to calculate the genetic correlations between nine brainstem volumetric traits and 26 non-imaging phenotypes based on the EUR-GWAS summary statistics and the EUR-LD reference. We corrected for the nine brainstem volumetric traits and 26 non-imaging phenotypes, generating a Bonferroni-corrected threshold of P < 0.05/26/9 = 2.14 × 10 − 4 . Genetic colocalization analyses. We used a Bayesian colocalization method named coloc ( https://chr1swallace.github.io/coloc/ ) 57 to identify the loci shared by each pair of nine brainstem volumetric traits (locus-trait associations with P < 5.56 × 10 − 9 ) and 26 non-imaging phenotypes (locus-phenotype associations with P < 5 × 10 − 8 ) based on the EUR-GWAS summary statistics. With the default priors ( P 1 = 1 × 10 − 4 , P 2 = 1 × 10 − 4 , and P 12 = 1 × 10 − 5 ), we considered evidence for colocalization if PP.H4 (the posterior probability of shared causal variant) was greater than 0.8. CondFDR/conjFDR analyses. Based on the EUR-GWAS summary data and EUR-LD reference from 1000 Genomes Project, we conducted the condFDR/conjFDR analyses using pleioFDR ( https://github.com/precimed/pleiofdr ) 59 to identify the variants shared by each pair of nine brainstem volumetric traits and 26 non-imaging phenotypes. We generated conditional Q-Q plots to assess polygenetic enrichment for each brainstem volumetric trait conditioned on P -values ( P < 0.10, P < 0.01, and P < 0.001) of the association with each non-imaging phenotype. We defined polygenetic enrichment as the curves with more significant P -values for the phenotype showed successive leftward shifts from the null distribution. Then, we performed the conjFDR analysis to identify shared genetic variants (conjFDR < 0.05) between each trait-phenotype pair, in which the conjFDR value of each variant was defined as the maximal FDR value of this variant in the two mutual conditional FDR analyses. In the analyses, we excluded variants in the MHC and 8p23.1 regions. We pooled the shared variants for each trait-phenotype pair with the same standard (< 500kb and LD r 2 > 0.1) based on the EUR-LD reference. We also annotated the shared independent variants for each trait-phenotype pair using the same strategies. Declarations Data availability The GWAS summary statistics used in this work from following publicly available dataset: the ENIGMA study ( https://enigma.ini.usc.edu/research/download-enigma-gwas-results/ ). All GWAS summary statistics from EAS, EUR and cross-ancestry meta-analyses of the brainstem and substructures volumes are publicly available at Zendo ( https://zenodo.org/records/13382122?preview=1&token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjhkNTgwNjM0LTFhNzUtNDFiZC1iNjk5LTM5M WE3YWMwMmQyYyIsImRhdGEiOnt9LCJyYW5kb20iOiI4NzRkYjM1MzY4OTc2NjRiN2ZhZWI1Njk 5MWNiNmIxMiJ9.TiWysH_4b6FO4fReeYT2pzQ-SOyjDZ1wPnsl07KAd5K0pFhVm3Z-Mh7LA_CMkd1fX888VR2_20IyierI9zoSP-g ). Code availability We made use of publicly available software and tools. All codes used to generate results reported in this paper are publicly available ( https://github.com/xuehui2014/The-genetic-architecture-of-brainstem-structures ). Competing interests The authors declare no competing interests. Author contributions C.Y., H.X. and J.F. designed the study. C.Y. and H.X. wrote the article. H.X. analyzed the data. C.Y., Z.G., S.Q. and W.Z. supervised this work. C.Y., Z.G., J.C., M.W., L.Z., G.C., Y.Y., W.L., H.Z., B.G., X.X., T.H., Z.Y., Q.Z., W.Q., F.L., M.L., S.W., Q.X., J.X., C.W., N.L., Y.J., H.X., P.Z., W.L., W.W., D.S., S.L., Z.Y., F.C., J.Z., W.S., Y.M., D.W., J.-H.G., Y.Y., K.X., J.X., B.Z., X.Z., X.-N.Z., M.J.L., Z.Y., S.Q., W.Z. acquired the data. All authors critically reviewed the manuscript. Acknowledgements We are grateful to all participants and researchers from CHIMGEN, UKBB, and ABCD, who generously donated their time to make these resources available. We acknowledge funding from the National Natural Science Foundation of China (82430063, 82030053, 81425013 to C.Y.). We are grateful to the ENIGMA for providing the GWAS summary statistics of whole brainstem volume. References Del Negro CA, Funk GD, Feldman JL (2018) Breathing matters. Nat Rev Neurosci 19:351–367 Guyenet PG (2006) The sympathetic control of blood pressure. Nat Rev Neurosci 7:335–346 Herring N, Kalla M, Paterson DJ (2019) The autonomic nervous system and cardiac arrhythmias: current concepts and emerging therapies. Nat Reviews Cardiol 16:707–726 Poe GR et al (2020) Locus coeruleus: a new look at the blue spot. Nat Rev Neurosci 21:644–659 Smidt MP, Burbach JPH (2007) How to make a mesodiencephalic dopaminergic neuron. Nat Rev Neurosci 8:21–32 Castellanos FX, Tannock R (2002) Neuroscience of attention-deficit/hyperactivity disorder: The search for endophenotypes. Nat Rev Neurosci 3:617–628 Steriade M (1996) Arousal: revisiting the reticular activating system. Science 272:225 Iglesias JE et al (2015) Bayesian segmentation of brainstem structures in MRI. NeuroImage 113:184–195 Milella G et al (2022) Medulla oblongata volume as a promising predictor of survival in amyotrophic lateral sclerosis. Neuroimage-Clinical 34 Chong CD, Plasencia JD, Frakes DH, Schwedt TJ (2017) Structural alterations of the brainstem in migraine. Neuroimage-Clinical 13:223–227 Bosco P et al (2019) Brainstem enlargement in preschool children with autism: Results from an intermethod agreement study of segmentation algorithms. Hum Brain Mapp 40:7–19 Elvsåshagen T et al (2020) The genetic architecture of human brainstem structures and their involvement in common brain disorders. Nat Commun 11:4016 Smith SM et al (2021) An expanded set of genome-wide association studies of brain imaging phenotypes in UK Biobank. Nat Neurosci 24:737–745 Satizabal CL et al (2019) Genetic architecture of subcortical brain structures in 38,851 individuals. Nat Genet 51:1624–1636 Zhao B et al (2019) Genome-wide association analysis of 19,629 individuals identifies variants influencing regional brain volumes and refines their genetic co-architecture with cognitive and mental health traits. Nat Genet 51:1637–1644 Fu JL et al (2024) Cross-ancestry genome-wide association studies of brain imaging phenotypes. Nat Genet 56:1110–1120 Xu Q et al (2020) CHIMGEN: a Chinese imaging genetics cohort to enhance cross-ethnic and cross-geographic brain research. Mol Psychiatry 25:517–529 Miller KL et al (2016) Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat Neurosci 19:1523–1536 Casey BJ et al (2018) The Adolescent Brain Cognitive Development (ABCD) study: Imaging acquisition across 21 sites. Dev Cogn Neurosci 32:43–54 Psaty BM et al (2009) Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium Design of Prospective Meta-Analyses of Genome-Wide Association Studies From 5 Cohorts. Circulation-Cardiovascular Genet 2:73–80 Thompson PM et al (2014) The ENIGMA Consortium: large-scale collaborative analyses of neuroimaging and genetic data. Brain Imaging Behav 8:153–182 Jiang LD et al (2019) A resource-efficient tool for mixed model association analysis of large-scale data. Nat Genet 51:1749–1755 Willer CJ, Li Y, Abecasis GR (2010) METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics 26:2190–2191 Bulik-Sullivan BK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47:291–295 Stutzbach LD et al (2013) The unfolded protein response is activated in disease-affected brain regions in progressive supranuclear palsy and Alzheimer's disease. Acta Neuropathol Commun 1:31 Freilinger T et al (2012) Genome-wide association analysis identifies susceptibility loci for migraine without aura. Nat Genet 44:777–782 Nyholt DR, Borsook D, Griffiths LR (2017) Migrainomics - identifying brain and genetic markers of migraine. Nat Reviews Neurol 13:725–741 Mandai K et al (2009) LIG Family Receptor Tyrosine Kinase-Associated Proteins Modulate Growth Factor Signals during Neural Development. Neuron 63:614–627 Luo Y et al (2021) Estimating heritability and its enrichment in tissue-specific gene sets in admixed populations. Hum Mol Genet 30:1521–1534 Brown BC, Ye CJ, Price AL, Zaitlen N, Network AGE (2016) Transethnic Genetic-Correlation Estimates from Summary Statistics. Am J Hum Genet 99:76–88 Monsoro-Burq AH (2015) PAX transcription factors in neural crest development. Semin Cell Dev Biol 44:87–96 Kichaev G et al (2014) Integrating Functional Data to Prioritize Causal Variants in Statistical Fine-Mapping Studies. PLoS Genet 10 Kichaev G et al (2017) Improved methods for multi-trait fine mapping of pleiotropic risk loci. Bioinformatics 33:248–255 Watanabe K, Taskesen E, van Bochoven A, Posthuma D (2017) Functional mapping and annotation of genetic associations with FUMA. Nat Commun 8:1826 Lin TT, Islam O, Heese K (2006) ABC transporters, neural stem cells and neurogenesis - a different perspective. Cell Res 16:857–871 Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M (2019) CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res 47:D886–D894 Dong SC et al (2023) Annotating and prioritizing human non-coding variants with RegulomeDB v.2. Nat Genet 55:724–726 de Leeuw CA, Mooij JM, Heskes T, Posthuma D (2015) MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput Biol 11:e1004219 Aguet F et al (2020) The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369:1318–1330 Barbeira AN et al (2018) Exploring the phenotypic consequences of tissue specific gene expression variation inferred from GWAS summary statistics. Nat Commun 9:1825 Barbeira AN et al (2019) Integrating predicted transcriptome from multiple tissues improves association detection. PLoS Genet 15:e1007889 Sey NYA et al (2020) A computational tool (H-MAGMA) for improved prediction of brain-disorder risk genes by incorporating brain chromatin interaction profiles. Nat Neurosci 23:583–589 Wang D et al (2018) Comprehensive functional genomic resource and integrative model for the human brain. Science 362:1266 Won HJ et al (2016) Chromosome conformation elucidates regulatory relationships in developing human brain. Nature 538:523–527 Hu B et al (2021) Neuronal and glial 3D chromatin architecture informs the cellular etiology of brain disorders. Nat Commun 12:3968 Rajarajan P et al (2018) Neuron-specific signatures in the chromosomal connectome associated with schizophrenia risk. Science 362:1269 Sey NYA et al (2022) Chromatin architecture in addiction circuitry identifies risk genes and potential biological mechanisms underlying cigarette smoking and alcohol use traits. Mol Psychiatry 27:3085–3094 Ashburner M et al (2000) Gene Ontology: tool for the unification of biology. Nat Genet 25:25–29 Gillespie M et al (2022) The reactome pathway knowledgebase 2022. Nucleic Acids Res 50:D687–D692 Kolberg L et al (2023) g:Profiler-interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Res 51:W207–W212 Jin S et al (2023) Downregulation of UBE4B promotes CNS axon regrowth and functional recovery after stroke. iScience 26:105885 Salih DAM et al (2012) FoxO6 regulates memory consolidation and synaptic function. Genes Dev 26:2780–2801 Nordstrom U, Jessell TM, Edlund T (2002) Progressive induction of caudal neural character by graded Wnt signaling. Nat Neurosci 5:525–532 Hevia CF, Engel-Pizcueta C, Udina F, Pujades C (2022) The neurogenic fate of the hindbrain boundaries relies on Notch3-dependent asymmetric cell divisions. Cell Rep 39:110915 Geisen MJ et al (2008) Hox paralog group 2 genes control the migration of mouse pontine neurons through slit-robo signaling. PLoS Biol 6:e142 Bulik-Sullivan B et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47:1236–1241 Wallace C (2020) Eliciting priors and relaxing the single causal variant assumption in colocalisation analyses. PLoS Genet 16:e1008720 Li S et al (2022) Regulatory variants at 2q33.1 confer schizophrenia risk by modulating distal gene TYW5 expression. Brain 145:770–786 Andreassen OA et al (2013) Improved Detection of Common Variants Associated with Schizophrenia and Bipolar Disorder Using Pleiotropy-Informed Conditional False Discovery Rate. PLoS Genet 9:e1003455 Fassl A, Geng Y, Sicinski P (2022) CDK4 and CDK6 kinases: From basic science to cancer therapy. Science 375:6577 Liu DX, Greene LA (2001) Regulation of neuronal survival and death by E2F-dependent gene repression and derepression. Neuron 32:425–438 Harbour JW, Dean DC (2000) Rb function in cell-cycle regulation and apoptosis. Nat Cell Biol 2:E65–E67 Zhao S et al (2024) Role of RB1 in neurodegenerative diseases: inhibition of post-mitotic neuronal apoptosis via Kmt5b. Cell Death Discovery 10:182 Gaufo GO, Thomas KR, Capecchi MR (2003) Hox3 genes coordinate mechanisms of genetic suppression and activation in the generation of branchial and somatic motoneurons. Development 130:5191–5201 Wang GS et al (2004) Transcriptional modification by a CASK-interacting nucleosome assembly protein. Neuron 42:113–128 Najm J et al (2008) Mutations of CASK cause an X-linked brain malformation phenotype with microcephaly and hypoplasia of the brainstem and cerebellum. Nat Genet 40:1065–1067 Liu N et al (2023) Cross-ancestry genome-wide association meta-analyses of hippocampal and subfield volumes. Nat Genet 55:1126–1137 Liu YJ et al (2023) Functional dissection of PRC1 subunits RYBP and YAF2 during neural differentiation of embryonic stem cells. Nat Commun 14:7164 Baek ST et al (2015) An AKT3-FOXG1-reelin network underlies defective migration in human focal malformations of cortical development. Nat Med 21:1445–1454 Dudek H et al (1997) Regulation of neuronal survival by the serine-threonine protein kinase Akt. Science 275:661–665 Flores AI et al (2008) Constitutively active Akt induces enhanced myelination in the CNS. J Neurosci 28:7174–7183 Huang HN et al (2004) In situ Akt phosphorylation in the nucleus tractus solitarii is involved in central control of blood pressure and heart rate. Circulation 110:2476–2483 Felix MS et al (2014) Activation of Akt/FKHR in the medulla oblongata contributes to spontaneous respiratory recovery after incomplete spinal cord injury in adult rats. Neurobiol Dis 69:93–107 Allen E et al (2005) Gigaxonin-controlled degradation of MAP1B light chain is critical to neuronal survival. Nature 438:224–228 Mahammad S et al (2013) Giant axonal neuropathy-associated gigaxonin mutations impair intermediate filament protein degradation. J Clin Invest 123:1964–1975 Hastings MH, Maywood ES, Brancaccio M (2018) Generation of circadian rhythms in the suprachiasmatic nucleus. Nat Rev Neurosci 19:453–469 Chang CC et al (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4:7 Auchter AM et al (2018) A description of the ABCD organizational structure and communication framework. Dev Cogn Neurosci 32:8–15 Mukamel RE et al (2023) Repeat polymorphisms underlie top genetic risk loci for glaucoma and colorectal cancer. Cell 186:3659–3673 Ahern J, Thompson W, Fan CC, Loughnan R (2023) Comparing Pruning and Thresholding with Continuous Shrinkage Polygenic Score Methods in a Large Sample of Ancestrally Diverse Adolescents from the ABCD Study. Behav Genet 53:292–309 Chen CY et al (2013) Improved ancestry inference using weights from external reference panels. Bioinformatics 29:1399–1406 Bycroft C et al (2018) The UK Biobank resource with deep phenotyping and genomic data. Nature 562:203–209 Fortin JP et al (2018) Harmonization of cortical thickness measurements across scanners and sites. NeuroImage 167:104–120 van der Meer D et al (2020) Brain scans from 21,297 individuals reveal the genetic architecture of hippocampal subfield volumes. Mol Psychiatry 25:3053–3065 Purcell S et al (2007) A tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81 PLINK:559–575 Gamazon ER et al (2015) A gene-based association method for mapping traits using reference transcriptome data. Nat Genet 47:1091–1098 Barbeira AN et al (2021) Exploiting the GTEx resources to decipher the mechanisms at GWAS loci. Genome Biol 22:49 Sey NYA, Pratt BM, Won H (2023) Annotating genetic variants to target genes using H-MAGMA. Nat Protoc 18:22–35 Shannon P et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498–2504 Reimand J et al (2019) Pathway enrichment analysis and visualization of omics data using g:Profiler, GSEA, Cytoscape and EnrichmentMap. Nat Protoc 14:482–517 Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryTables20240907.xlsx Supplementary Tables Supplementaryinformation20240907.docx Supplementary Information and Figures Cite Share Download PDF Status: Published Journal Publication published 10 Dec, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-5060768\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":354044223,\"identity\":\"8c1b0f8e-1c29-4d7d-b230-5b2665d04da4\",\"order_by\":0,\"name\":\"Chunshui Yu\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYDACCSBmbGDgYWBgPgAROUC8FrbEBpK0AAGPIXFa+Gf3mD38ueOwjMGNnO+PbrYxyPHdSGD8XIDPkjtnzI15zxzmMbiRu7E5t43BWPJGArP0DDxaDCRyzKQZ2xBaEjfcSGBj5iGgRfInWEvOQ5CWeqK0SPBCtDCCtCQYENIicSOtTJq3LZ1H8swzw9k55yQMZ5552CyNTwv/jORtQIdZ2/MdT37wOafMRh7IOPgZnxY4UDgAsZUBFk2EgTyR6kbBKBgFo2AEAgCXqEy+lZ+xdAAAAABJRU5ErkJggg==\",\"orcid\":\"https://orcid.org/0000-0001-5648-5199\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Chunshui\",\"middleName\":\"\",\"lastName\":\"Yu\",\"suffix\":\"\"},{\"id\":354044224,\"identity\":\"366d8cbb-4de9-4168-bd7b-eec16f1ed7d7\",\"order_by\":1,\"name\":\"Hui Xue\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hui\",\"middleName\":\"\",\"lastName\":\"Xue\",\"suffix\":\"\"},{\"id\":354044225,\"identity\":\"ed4961ad-fd06-40cb-92e8-c2a59dee7216\",\"order_by\":2,\"name\":\"Jilian Fu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jilian\",\"middleName\":\"\",\"lastName\":\"Fu\",\"suffix\":\"\"},{\"id\":354044226,\"identity\":\"94cff8b4-54fa-4692-a014-575cd39546aa\",\"order_by\":3,\"name\":\"Zuojun Geng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The Second Hospital of Hebei Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zuojun\",\"middleName\":\"\",\"lastName\":\"Geng\",\"suffix\":\"\"},{\"id\":354044227,\"identity\":\"4649cd12-ed70-40aa-8e3f-cc549dd493e6\",\"order_by\":4,\"name\":\"Jingliang Cheng\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-9944-8419\",\"institution\":\"The First Affiliated Hospital of Zhengzhou University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jingliang\",\"middleName\":\"\",\"lastName\":\"Cheng\",\"suffix\":\"\"},{\"id\":354044228,\"identity\":\"4c05302f-08aa-4b1e-bb54-16579074167c\",\"order_by\":5,\"name\":\"Meiyun Wang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-4454-5005\",\"institution\":\"Henan Provincial People's Hospital \\u0026 the People's Hospital of Zhengzhou University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Meiyun\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":354044229,\"identity\":\"fb5f7b80-f385-4ed5-8ec9-3f978c6c573d\",\"order_by\":6,\"name\":\"Long Jiang Zhang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-6664-7224\",\"institution\":\"Jinling hospital, Affiliated Hospital of Medical School, Nanjing University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Long\",\"middleName\":\"Jiang\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044230,\"identity\":\"cece139a-8f8a-432d-ad98-8ebba1c3ec46\",\"order_by\":7,\"name\":\"Guangbin Cui\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tangdu Hospital, Air Force Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Guangbin\",\"middleName\":\"\",\"lastName\":\"Cui\",\"suffix\":\"\"},{\"id\":354044231,\"identity\":\"8eced17d-93e8-4cef-b4de-436365f574cc\",\"order_by\":8,\"name\":\"Yongqiang Yu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Anhui Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yongqiang\",\"middleName\":\"\",\"lastName\":\"Yu\",\"suffix\":\"\"},{\"id\":354044232,\"identity\":\"9c1bf49e-9165-4621-a3a6-5eea74110fa9\",\"order_by\":9,\"name\":\"Weihua Liao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Xiangya Hospital, Central South University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Weihua\",\"middleName\":\"\",\"lastName\":\"Liao\",\"suffix\":\"\"},{\"id\":354044233,\"identity\":\"29dd499e-fd88-49cf-9e57-1fa235784eb6\",\"order_by\":10,\"name\":\"Hui Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Hospital of Shanxi Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hui\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044234,\"identity\":\"2db04674-82fb-455b-a826-6030e6c5e9c7\",\"order_by\":11,\"name\":\"Bo Gao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The Affiliated Hospital of Guizhou Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Bo\",\"middleName\":\"\",\"lastName\":\"Gao\",\"suffix\":\"\"},{\"id\":354044235,\"identity\":\"8bb92633-e801-405a-8064-4534803f2600\",\"order_by\":12,\"name\":\"Xiaojun Xu\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-0127-2812\",\"institution\":\"The Second Affiliated Hospital of Zhejiang University, School of Medicine\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xiaojun\",\"middleName\":\"\",\"lastName\":\"Xu\",\"suffix\":\"\"},{\"id\":354044236,\"identity\":\"042b55e9-aa9f-485b-b76f-106679d3b87c\",\"order_by\":13,\"name\":\"Tong Han\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Neurosurgery, Huanhu Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Tong\",\"middleName\":\"\",\"lastName\":\"Han\",\"suffix\":\"\"},{\"id\":354044237,\"identity\":\"b95ed44f-b318-4662-92aa-985e1f61e462\",\"order_by\":14,\"name\":\"Zhenwei Yao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Huashan Hospital, Fudan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zhenwei\",\"middleName\":\"\",\"lastName\":\"Yao\",\"suffix\":\"\"},{\"id\":354044238,\"identity\":\"72f0c265-335e-41b9-93be-09b6fe3b6610\",\"order_by\":15,\"name\":\"Quan Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Characteristic Medical Center of Chinese People's Armed Police Force\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Quan\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044239,\"identity\":\"0cfabd92-1d88-4896-8820-ae8d8bf89f10\",\"order_by\":16,\"name\":\"Wen Qin\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-9121-8296\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wen\",\"middleName\":\"\",\"lastName\":\"Qin\",\"suffix\":\"\"},{\"id\":354044240,\"identity\":\"dd362ce3-892e-4532-aed6-4f5b8d35e80e\",\"order_by\":17,\"name\":\"Feng Liu\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-3570-4222\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Feng\",\"middleName\":\"\",\"lastName\":\"Liu\",\"suffix\":\"\"},{\"id\":354044241,\"identity\":\"d553c023-00f6-4322-a5e7-320dd6d690c4\",\"order_by\":18,\"name\":\"Meng Liang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Meng\",\"middleName\":\"\",\"lastName\":\"Liang\",\"suffix\":\"\"},{\"id\":354044242,\"identity\":\"0d668781-5ce3-473b-befc-faa07b832a2a\",\"order_by\":19,\"name\":\"Sijia Wang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-8949-303X\",\"institution\":\"School of Medical Imaging and Tianjin Key Laboratory of Functional Imaging, Tianjin Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Sijia\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":354044243,\"identity\":\"c1d05ab1-f430-4f9c-88a2-35ca1ac6a88d\",\"order_by\":20,\"name\":\"Qiang Xu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Qiang\",\"middleName\":\"\",\"lastName\":\"Xu\",\"suffix\":\"\"},{\"id\":354044244,\"identity\":\"6686617c-c3c8-439f-96ac-20fdaff7f4ae\",\"order_by\":21,\"name\":\"Jiayuan Xu\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-9473-1047\",\"institution\":\"Department of Radiology and Tianjin Key Laboratory of Functional Imaging, Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jiayuan\",\"middleName\":\"\",\"lastName\":\"Xu\",\"suffix\":\"\"},{\"id\":354044245,\"identity\":\"bbed588e-d938-47e2-9d1e-746453417a14\",\"order_by\":22,\"name\":\"Caihong Wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Zhengzhou University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Caihong\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":354044246,\"identity\":\"9c106c98-da09-494c-81fa-592c5b90ca31\",\"order_by\":23,\"name\":\"Nana Liu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Nana\",\"middleName\":\"\",\"lastName\":\"Liu\",\"suffix\":\"\"},{\"id\":354044247,\"identity\":\"e0e49c29-e387-4afd-b6f2-c07683ae786b\",\"order_by\":24,\"name\":\"Yuan Ji\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-2874-5904\",\"institution\":\"Tianjin Medical University General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yuan\",\"middleName\":\"\",\"lastName\":\"Ji\",\"suffix\":\"\"},{\"id\":354044248,\"identity\":\"acc866ce-5f2b-4abb-bb7d-44ecb260ecce\",\"order_by\":25,\"name\":\"Peng Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University Cancer Institute and Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Peng\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044249,\"identity\":\"e6844964-559b-4193-8f6e-aa2c4ff5a00d\",\"order_by\":26,\"name\":\"Wei Li\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tianjin Medical University Cancer Institute and Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wei\",\"middleName\":\"\",\"lastName\":\"Li\",\"suffix\":\"\"},{\"id\":354044250,\"identity\":\"c49d1293-d687-43d4-8579-e9125d974958\",\"order_by\":27,\"name\":\"Wei Wei\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Henan Provincial People’s Hospital \\u0026 the People’s Hospital of Zhengzhou University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wei\",\"middleName\":\"\",\"lastName\":\"Wei\",\"suffix\":\"\"},{\"id\":354044251,\"identity\":\"d3a17c76-688b-4252-ac50-d652a617da9a\",\"order_by\":28,\"name\":\"Dapeng Shi\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Henan Provincial People's Hospital \\u0026 Zhengzhou University People's Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Dapeng\",\"middleName\":\"\",\"lastName\":\"Shi\",\"suffix\":\"\"},{\"id\":354044252,\"identity\":\"6e736fa3-6154-4d55-b9c5-032027af7979\",\"order_by\":29,\"name\":\"Su Lui\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"West China Hospital of Sichuan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Su\",\"middleName\":\"\",\"lastName\":\"Lui\",\"suffix\":\"\"},{\"id\":354044253,\"identity\":\"829fe222-ceba-469e-b901-c3316774d3e2\",\"order_by\":30,\"name\":\"Zhihan Yan\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zhihan\",\"middleName\":\"\",\"lastName\":\"Yan\",\"suffix\":\"\"},{\"id\":354044254,\"identity\":\"4389e688-c2a6-4492-8935-679c1fbb59c1\",\"order_by\":31,\"name\":\"Feng Chen\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-9129-7895\",\"institution\":\"Hainan General Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Feng\",\"middleName\":\"\",\"lastName\":\"Chen\",\"suffix\":\"\"},{\"id\":354044255,\"identity\":\"8e7c2a6c-22c2-4ad1-a279-7715884804a2\",\"order_by\":32,\"name\":\"Jing Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Lanzhou University Second Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jing\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044256,\"identity\":\"6ea17e7e-0a51-4951-9350-bcc290cd70d2\",\"order_by\":33,\"name\":\"Wen Shen\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Radiology, Tianjin First Center Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wen\",\"middleName\":\"\",\"lastName\":\"Shen\",\"suffix\":\"\"},{\"id\":354044257,\"identity\":\"bfa22503-b0ab-4d1c-b456-970af5f3c6f5\",\"order_by\":34,\"name\":\"Yanwei Miao\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Dalian Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yanwei\",\"middleName\":\"\",\"lastName\":\"Miao\",\"suffix\":\"\"},{\"id\":354044258,\"identity\":\"e7b85af8-64bb-4d85-a273-53a36557627b\",\"order_by\":35,\"name\":\"Dawei Wang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Radiology, Qilu Hospital of Shandong University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Dawei\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":354044259,\"identity\":\"5e5285df-501e-4bc9-869f-86354086e88f\",\"order_by\":36,\"name\":\"Jia-Hong Gao\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-9311-0297\",\"institution\":\"\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jia-Hong\",\"middleName\":\"\",\"lastName\":\"Gao\",\"suffix\":\"\"},{\"id\":354044260,\"identity\":\"560b6c15-408e-4ec7-ac1b-7eae83f2e70c\",\"order_by\":37,\"name\":\"Yunjun Yang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Wenzhou Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yunjun\",\"middleName\":\"\",\"lastName\":\"Yang\",\"suffix\":\"\"},{\"id\":354044261,\"identity\":\"5e5ceee7-0762-4190-8934-45dd25dc932f\",\"order_by\":38,\"name\":\"Kai Xu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The Affiliated Hospital of Xuzhou Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Kai\",\"middleName\":\"\",\"lastName\":\"Xu\",\"suffix\":\"\"},{\"id\":354044262,\"identity\":\"e471391b-3315-4f1d-a495-e8e40ee274c2\",\"order_by\":39,\"name\":\"Junfang Xian\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-2191-9393\",\"institution\":\"Beijing Tongren Hospital, Capital Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Junfang\",\"middleName\":\"\",\"lastName\":\"Xian\",\"suffix\":\"\"},{\"id\":354044263,\"identity\":\"50fdd37d-30c8-4af1-91da-d74abe04ea08\",\"order_by\":40,\"name\":\"Bing Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Drum Tower Hospital, Medical School of Nanjing University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Bing\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044264,\"identity\":\"15133718-125e-4ea6-95a9-9748c79eafb8\",\"order_by\":41,\"name\":\"Xiaochu Zhang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-7541-0130\",\"institution\":\"University of Science and Technology of China\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xiaochu\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":354044265,\"identity\":\"aadd7b4b-27fb-4aa0-a1fe-5216970397e9\",\"order_by\":42,\"name\":\"Xi-Nian Zuo\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-9110-585X\",\"institution\":\"Beijing Normal University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Xi-Nian\",\"middleName\":\"\",\"lastName\":\"Zuo\",\"suffix\":\"\"},{\"id\":354044266,\"identity\":\"fe6841d1-b439-4af4-ad7b-ced1cfa13f10\",\"order_by\":43,\"name\":\"Mulin Jun Li\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-3598-3679\",\"institution\":\"School of Basic Medical Sciences, Tianjin Medical University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mulin\",\"middleName\":\"Jun\",\"lastName\":\"Li\",\"suffix\":\"\"},{\"id\":354044267,\"identity\":\"31e03076-8930-4b89-9646-bbdd593a43f9\",\"order_by\":44,\"name\":\"Zhaoxiang Ye\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-3157-8393\",\"institution\":\"Tianjin Medical University Cancer Institute and Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Zhaoxiang\",\"middleName\":\"\",\"lastName\":\"Ye\",\"suffix\":\"\"},{\"id\":354044268,\"identity\":\"7cab405e-ff85-4861-a73b-78f1bad7a3d6\",\"order_by\":45,\"name\":\"Shijun Qiu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"The First Affiliated Hospital of Guangzhou University of Traditional Chinese Medicine\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Shijun\",\"middleName\":\"\",\"lastName\":\"Qiu\",\"suffix\":\"\"},{\"id\":354044269,\"identity\":\"93c2c4b4-faed-4ea5-8eb5-879b2b5d667c\",\"order_by\":46,\"name\":\"Wenzhen Zhu\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-6252-9450\",\"institution\":\"Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wenzhen\",\"middleName\":\"\",\"lastName\":\"Zhu\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-09-10 01:00:33\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-5060768/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-5060768/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1038/s41467-025-67221-6\",\"type\":\"published\",\"date\":\"2025-12-10T05:00:00+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":66240658,\"identity\":\"d814297f-a664-4754-a5a4-78c1bc53f710\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:33:56\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1260950,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eGenetic discovery of GWASs for brainstem and substructure volumes. a\\u003c/strong\\u003e, Diagrams show brainstem and four substructures (medulla, pons, midbrain, and SCP). We conduct GWASs for nine brainstem volumetric traits, including whole brainstem volume and absolute and relative volumes of each substructure. \\u003cstrong\\u003eb-e\\u003c/strong\\u003e, The regional plots show examples of variant-trait associations in EAS (\\u003cstrong\\u003eb\\u003c/strong\\u003e), ABCD (\\u003cstrong\\u003ec\\u003c/strong\\u003e), EUR (\\u003cstrong\\u003ed\\u003c/strong\\u003e), and cross-ancestry (\\u003cstrong\\u003ee\\u003c/strong\\u003e) GWASs.\\u003cstrong\\u003e \\u003c/strong\\u003eThe red and blue lines indicate genome- (\\u003cem\\u003eP\\u003c/em\\u003e = 5 × 10\\u003csup\\u003e−8\\u003c/sup\\u003e) and study-wide (\\u003cem\\u003eP\\u003c/em\\u003e = 5.56 × 10\\u003csup\\u003e−9\\u003c/sup\\u003e, Bonferroni correction for nine traits) significance thresholds. \\u003cstrong\\u003ef\\u003c/strong\\u003e,\\u003cstrong\\u003e g\\u003c/strong\\u003e, Ideograms show study-wide significant (\\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 5.56 × 10\\u003csup\\u003e−9\\u003c/sup\\u003e) variant-trait associations (\\u003cstrong\\u003ef\\u003c/strong\\u003e) and LD-independent loci (\\u003cstrong\\u003eg\\u003c/strong\\u003e). The grey and pink dots (\\u003cstrong\\u003eg\\u003c/strong\\u003e) represent known and new LD-independent loci compared to all previous GWASs (\\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 5 × 10\\u003csup\\u003e−8\\u003c/sup\\u003e). Abbreviation: SCP, superior cerebellar peduncle.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/809d81bdf2ebbc3bb43c42cb.png\"},{\"id\":66240659,\"identity\":\"1e2fe7af-95c5-4f79-b85a-dd82dd214b2c\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:33:56\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":368352,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eAncestry-shared and ancestry-specific genetic associations with brainstem volumetric traits. a\\u003c/strong\\u003e, The bar chart shows counts of ancestry-specific (blue; Cochran’s \\u003cem\\u003eQ\\u003c/em\\u003e test: \\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 1.36 × 10\\u003csup\\u003e−4\\u003c/sup\\u003e, Bonferroni correction for 369 variant-trait associations included in both CHIMGEN and UKBB datasets) and ancestry-shared (red; Cochran’s \\u003cem\\u003eQ\\u003c/em\\u003e test: \\u003cem\\u003eP\\u003c/em\\u003e ≥ 0.05) associations between UKBB-GWASs and CHIMGEN-GWASs. \\u003cstrong\\u003eb\\u003c/strong\\u003e, Pie plots show counts and ratios of the discovered (left: CHIMGEN vs UKBB) and replicated (right: CHIMGEN vs ABCD) ancestry-specific and ancestry-shared associations. \\u003cstrong\\u003ec\\u003c/strong\\u003e, The z-score differences in three ancestry-specific associations between EAS-GWASs and EUR-GWASs in the discovery (left: CHIMGEN vs UKBB) and replication (right: CHIMGEN vs ABCD) stages. \\u003cstrong\\u003ed\\u003c/strong\\u003e, The regional plots show an example of the ancestry-shared association (Cochran’s \\u003cem\\u003eQ\\u003c/em\\u003e test: \\u003cem\\u003eP\\u003c/em\\u003e = 0.30) between rs16863657 and pons absolute volume in UKBB-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e = 6.72 × 10\\u003csup\\u003e−16\\u003c/sup\\u003e) and EAS-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e = 6.88 × 10\\u003csup\\u003e−8\\u003c/sup\\u003e). The red and blue lines indicate the genome-wide (\\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 5 × 10\\u003csup\\u003e−8\\u003c/sup\\u003e) and study-wide (\\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 5.56 × 10\\u003csup\\u003e−9\\u003c/sup\\u003e) significance thresholds of GWASs.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/795bb96f00c6fe415c3538e9.png\"},{\"id\":66240665,\"identity\":\"0992e02a-c3ea-475a-a4e6-bc49dbe2e599\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:33:56\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":606760,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eStatistical fine-mapping and functional annotation. a\\u003c/strong\\u003e, The ideogram shows the causal variants (PP \\u0026gt; 0.8) identified by statistical fine-mapping of 296 pooled locus-trait associations for brainstem volumetric traits. We identify 250 unique causal variants, including 25 only from EUR analyses, 223 only from cross-ancestry analyses, and two from both analyses. \\u003cstrong\\u003eb\\u003c/strong\\u003e, Comparison of the sizes of 95% credible sets of the 223 trait-associated loci from fine-mapping for cross-ancestry (orange) and EUR (blue) GWAS analyses. Two-sided \\u003cem\\u003eP\\u003c/em\\u003e value is calculated by the Wilcoxon rank-sum test. The sizes of 95% credible sets are followed by log-transformation and reported as medians and IQRs.\\u003cstrong\\u003e c\\u003c/strong\\u003e, The regional plots show a locus-trait association between 2p11.2 and pons absolute volume with improved resolution in cross-ancestry fine-mapping. \\u003cstrong\\u003ed\\u003c/strong\\u003e, Categorizing the causal variants (PP \\u0026gt; 0.1) from fine-mapping for EUR (left) and cross-ancestry (right) GWASs according to the genomic location and functional consequence.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/038cc9bc2f878a11878844d8.png\"},{\"id\":66241444,\"identity\":\"eab1141e-aafb-45b2-802c-1b51477d91f6\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:41:56\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":765908,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ePrioritized genes associated with brainstem and substructure volumes and their enriched pathways. a\\u003c/strong\\u003e, The upset plot (blue) shows the counts of prioritized genes for nine brainstem volumetric traits and their intersections. The inserted upset plot (red) shows the counts of prioritized genes for brainstem and substructures (irrespective of absolute or relative volume) and their intersections. \\u003cstrong\\u003eb\\u003c/strong\\u003e, The enrichment map shows the hierarchical relationships of biological pathways enriched by 550 prioritized genes for brainstem volumetric traits. Each circle represents an enriched pathway and the line thickness represents the number of genes shared by the two pathways.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/4e931acf18e960f6b6db05f2.png\"},{\"id\":66240662,\"identity\":\"6419d7d0-564c-4990-9797-b6ea5f7063dc\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:33:56\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":488111,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eShared and distinct genetic architectures between brainstem substructures. a\\u003c/strong\\u003e, The ideogram shows shared (sphere: shared by three substructures; triangle: shared by two substructures) and distinct (diamond) loci (\\u003cem\\u003eP\\u003c/em\\u003e \\u0026lt; 5.56 × 10\\u003csup\\u003e−9\\u003c/sup\\u003e) among the medulla (red), pons (purple), and midbrain (blue) volumes. \\u003cstrong\\u003eb\\u003c/strong\\u003e, The radar chart shows an example of a locus with lead variant rs9428966 shared by all three brainstem substructures and three examples of substructure-specific loci (rs12448813 for medulla; rs72927168 for pons; and rs2331753 for midbrain). Colors show different brainstem substructures and scale is the significance of the most significant variant-trait association of each variant with its corresponding absolute and relative substructure volumes in all GWASs. \\u003cstrong\\u003ec\\u003c/strong\\u003e, The Venn diagram shows the shared and distinct prioritized genes for medulla (red), pons (purple), and midbrain (blue) volumes. \\u003cstrong\\u003ed\\u003c/strong\\u003e, The compound bar chart shows the 28 shared enrichment pathways (\\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05) for medulla (red), pons (purple), and midbrain (blue) volumes.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/7a9be5107093be4c258fad61.png\"},{\"id\":66241442,\"identity\":\"dd95e0cb-9135-4e25-b6ca-545a42c8477f\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:41:56\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":625025,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eShared genetic architectures between brainstem volumetric traits and non-imaging phenotypes in EUR individuals. a\\u003c/strong\\u003e, The heat plot shows genetic correlations between brainstem volumetric traits and non-imaging phenotypes. Asterisk (*) means genetic correlation significant at Bonferroni-corrected\\u003cem\\u003e P\\u003c/em\\u003e \\u0026lt; 2.14 × 10\\u003csup\\u003e−4\\u003c/sup\\u003e. \\u003cstrong\\u003eb\\u003c/strong\\u003e, The ideogram shows genetic colocalizations (PP.H4 \\u0026gt; 0.8) between brainstem volumetric traits and non-imaging phenotypes. \\u003cstrong\\u003ec\\u003c/strong\\u003e, The heat plot shows the counts of loci shared by each pair of brainstem volumetric traits and non-imaging phenotypes in conjunctional FDR (conjunctional FDR \\u0026lt; 0.05) analyses. \\u003cstrong\\u003ed\\u003c/strong\\u003e, The regional plots show a locus at 7q21.2 with lead variant rs42039 shared by brainstem volumetric traits (medulla and midbrain absolute volumes) and blood pressure phenotypes (SBP and PP) identified by both conjunctional FDR and colocalization analyses. Abbreviation: AD, Alzheimer’s disease; ADHD, attention deficit hyperactivity disorder; ALS, amyotrophic lateral sclerosis; ASD, Autism spectrum disorders; DBP, diastolic blood pressure; GGE, genetic generalized epilepsy; iRBD, isolated rapid eye movement sleep behavior disorder; LBD, Lewy body dementia; MDD, major depressive disorder; PD, Parkinson's disease; PP, pulse pressure; PP.H4, posterior probability of shared causal variant; PTSD, posttraumatic stress disorder; SBP, systolic blood pressure; SCP, superior cerebellar peduncle.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Fig.6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/7861097f9283a10ddfa91c42.png\"},{\"id\":100295557,\"identity\":\"7b4b9220-1b10-48d6-8390-7ee1884d7bd8\",\"added_by\":\"auto\",\"created_at\":\"2026-01-15 08:06:05\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":6350235,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/74e03dc4-55c5-4d8e-a372-ddd702c670b3.pdf\"},{\"id\":66241545,\"identity\":\"0f8ac9d4-a502-4f7f-8c6e-b307d7e43676\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:49:56\",\"extension\":\"xlsx\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":2268640,\"visible\":true,\"origin\":\"\",\"legend\":\"Supplementary Tables\",\"description\":\"\",\"filename\":\"SupplementaryTables20240907.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/687a18a8adafd044ac8a4df6.xlsx\"},{\"id\":66240667,\"identity\":\"669d5bca-9eba-410b-9c10-797f399081c2\",\"added_by\":\"auto\",\"created_at\":\"2024-10-09 06:33:59\",\"extension\":\"docx\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":254578364,\"visible\":true,\"origin\":\"\",\"legend\":\"Supplementary Information and Figures\",\"description\":\"\",\"filename\":\"Supplementaryinformation20240907.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5060768/v1/0345f6f0e04a597ae10624bf.docx\"}],\"financialInterests\":\"There is \\u003cb\\u003eNO\\u003c/b\\u003e Competing Interest.\",\"formattedTitle\":\"The genetic architecture of brainstem structures\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003eThe brainstem anatomically connects the cerebrum to cerebellum and spinal cord, and is composed of medulla, pons, and midbrain. It contains numerous grey matter nuclei and white matter tracts that are critical for our life. For instance, the vital center in the medulla is responsible for the autonomic control of breathing, blood pressure, and heart rate\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR2\\\" citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e\\u003c/sup\\u003e. The locus coeruleus in the pons is the main source of noradrenaline playing key roles in vigilance and cognition, and the dysfunction of this nucleus may lead to arousal-related disorders, such as attention deficit hyperactivity disorder (ADHD) and major depressive disorder (MDD)\\u003csup\\u003e\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e\\u003c/sup\\u003e. The substantia nigra in the midbrain contains enormous dopaminergic neurons, generating dopamine (DA) essential for the control of voluntary movement and the regulation of emotion. Damage to this structure is associated with schizophrenia (SCZ), Parkinson\\u0026rsquo;s disease (PD), MDD, and ADHD\\u003csup\\u003e\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e\\u003c/sup\\u003e. In addition to the grey matter nuclei, white matter fibers in the brainstem are also important for various physiological functions. For example, the reticular formation is crucial for maintaining consciousness\\u003csup\\u003e\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e\\u003c/sup\\u003e and the corticospinal tract is vital for the control of limb movement.\\u003c/p\\u003e \\u003cp\\u003eBased on the individual\\u0026rsquo;s brain structural magnetic resonance imaging (MRI) data, the brainstem and substructure (medulla, pons, and midbrain) volumes can be estimated by approaches, such as the Bayesian brainstem segmentation\\u003csup\\u003e\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u003c/sup\\u003e. Significant differences in brainstem substructure volumes are observed between healthy controls and patients with neuropsychiatric disorders\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR9 CR10\\\" citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e\\u003c/sup\\u003e, such as Alzheimer\\u0026rsquo;s disease (AD), amyotrophic lateral sclerosis (ALS), migraine, and autism spectrum disorder (ASD), indicating that the brainstem and substructure volumes are potential neuroimaging markers for these neuropsychiatric disorders. To investigate the genetic architectures of these brainstem volumetric traits, five genome-wide association studies (GWASs) are conducted for the whole brainstem volume in up to 40,282 individuals\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR13 CR14 CR15\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e\\u003c/sup\\u003e, including two studies\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e\\u003c/sup\\u003e also performing GWASs for brainstem substructure volumes in up to 33,224 individuals. Although one study involving the GWAS for whole brainstem volume also includes 7,058 individuals with East Asian ancestry (EAS)\\u003csup\\u003e\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e\\u003c/sup\\u003e, other studies only include individuals with European ancestry (EUR)\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR13 CR14\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e\\u003c/sup\\u003e, preventing from distinguishing ancestry-specific from ancestry-shared associations for brainstem substructure volumes.\\u003c/p\\u003e \\u003cp\\u003eIn the present study, we conducted the largest GWAS for whole brainstem volume in 72,717 non-overlapping individuals, among which 7,096 EAS participants were from the Chinese Imaging Genetics (CHIMGEN) study\\u003csup\\u003e\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e\\u003c/sup\\u003e and 65,621 EUR participants were from the UK Biobank (UKBB) study\\u003csup\\u003e\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e\\u003c/sup\\u003e, the Adolescent Brain Cognitive Development (ABCD) study\\u003csup\\u003e\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e\\u003c/sup\\u003e, the Cohorts of Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium\\u003csup\\u003e\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e\\u003c/sup\\u003e, and the Enhancing Neuroimaging Genetics through Meta-Analysis (ENIGMA) consortium\\u003csup\\u003e\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e\\u003c/sup\\u003e. We performed the largest GWASs for brainstem substructure volumes in 48,522 individuals (7,096 EAS participants from CHIMGEN and 41,426 EUR participants from ABCD and UKBB). We identified ancestry-shared and ancestry-specific genetic associations with brainstem volumetric traits based on the EAS-GWAS and EUR-GWAS summary statistics. We detected potential causal variants by statistical fine-mapping and prioritized genes by integrating with genomic location, gene expression, and chromatin interaction data. We searched for shared and distinct genetic loci, genes, and biological pathways between brainstem substructure volumes, and explored shared genetic architectures of brainstem and substructure volumes with brainstem-related physiological and pathological phenotypes by conducting genetic correlation, genetic colocalization, and conditional and conjunctional false discovery rate (condFDR/conjFDR) analyses. The study design is presented in \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;1\\u003c/b\\u003e. By increasing sample size and population diversity, we found more genetic loci associated with brainstem and substructure volumes, providing a better understanding of the genetic architecture of brainstem volumetric traits.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eOverview of brainstem volumetric traits and GWAS strategies\\u003c/h2\\u003e \\u003cp\\u003eWe included three sets of raw neuroimaging genetics datasets (CHIMGEN, ABCD, and UKBB) and a set of GWAS summary data for whole brainstem volume from ENIGMA, CHARGE, and UKBB (first release) datasets\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e. Consisting with the prior GWASs\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e\\u003c/sup\\u003e for brainstem substructure volumes, we used the same Bayesian approach\\u003csup\\u003e\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u003c/sup\\u003e to obtain the absolute volumes of whole brainstem, medulla, pons, midbrain, and superior cerebellar peduncle (SCP) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea) of each CHIMGEN, ABCD, or UKBB participant, although SCP is not a classical brainstem substructure. The obtained whole brainstem (including SCP) volume was included as an additional covariate to perform GWASs for relative volumes of medulla, pons, midbrain, and SCP\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u003c/sup\\u003e. As the previous GWAS\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e applied the automatic subcortical segmentation method to obtain whole brainstem (not including SCP) volume, we also used the same method to calculate the whole brainstem volume for each CHIMGEN, ABCD, or UKBB participant, which was used in GWASs for the trait. We investigated the consistency of whole brainstem volumes obtained by the two segmentation methods by calculating intraclass correlation coefficient (ICC) of whole brainstem volumes derived from the two methods for CHIMGEN, ABCD, and UKBB datasets, respectively. We found high ICCs (0.956\\u0026ndash;0.997) (\\u003cb\\u003eSupplementary Table\\u0026nbsp;1\\u003c/b\\u003e). Based on the participants from each dataset who had neuroimaging data acquired at two different time points, we also calculated the ICC of each substructure volume to assess the test-retest reliability of the Bayesian brainstem segmentation, and found high ICCs (0.868\\u0026ndash;0.992) (\\u003cb\\u003eSupplementary Table\\u0026nbsp;1\\u003c/b\\u003e). The demographic and brainstem volumetric data are presented in \\u003cb\\u003eSupplementary Table\\u0026nbsp;2\\u003c/b\\u003e. We used the mixed linear model (MLM) in fastGWA\\u003csup\\u003e\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e\\u003c/sup\\u003e to conduct GWASs for brainstem volumetric traits in CHIMGEN, ABCD, and UKBB participants, respectively. Based on the obtained GWAS summary statistics and those from the prior EUR-GWAS for whole brainstem volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e, we used inverse variance weighted (IVW) fixed effect model in METAL\\u003csup\\u003e\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e\\u003c/sup\\u003e to conduct EUR and cross-ancestry GWAS meta-analyses to make full use of the available data resources. As the prior GWAS for whole brainstem volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e included the first released UKBB data, we excluded the participants and reperformed UKBB-GWAS for whole brainstem volume, and utilized the obtained GWAS summary statistics for the trait in the meta-analyses with the prior study\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e. The participants and genetic variants included in each GWAS are shown in \\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e. Due to lacking any independent replication, we only reported the study-wide significant associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni correction for nine traits). The genomic control inflation factor (λ\\u003csub\\u003eGC\\u003c/sub\\u003e) and linkage disequilibrium score regression (LDSC) intercepts\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u003c/sup\\u003e were used to identify population stratification for each GWAS, and did not show any population stratification (\\u003cb\\u003eSupplementary Table\\u0026nbsp;4\\u003c/b\\u003e). We used imputed genotype data from 7,096 CHIMGEN and 35,611 UKBB participants to construct EAS, EUR, and cross-ancestry LD references, which were used to identify independent variant-trait and locus-trait associations, and LD-independent loci in EAS, EUR, and cross-ancestry GWASs, respectively.\\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003eGenetic discovery in EAS-GWASs\\u003c/h3\\u003e\\n\\u003cp\\u003eWe conducted EAS-GWASs (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected) for nine brainstem volumetric traits at 8,790,144 autosomal and 227,168 X-chromosomal variants with information score (info)\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.6 and minor allele frequency (MAF)\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5% in 7,096 CHIMGEN participants. We only identified one independent variant-trait association (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;2.61 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eb, \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;2 and Supplementary Table\\u0026nbsp;5\\u003c/b\\u003e) between rs6547787 at 2p11.2 and pons absolute volume. The variant rs6547787 is an expression quantitative trait locus (eQTL) of \\u003cem\\u003eEIF2AK3\\u003c/em\\u003e, a risk gene for progressive supranuclear palsy (PSP) characterized by brainstem pathology\\u003csup\\u003e\\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e\\n\\u003ch3\\u003eGenetic discovery in EUR-GWASs\\u003c/h3\\u003e\\n\\u003cp\\u003eWe conducted EUR-GWASs (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected) for the brainstem volumetric traits at 9,064,819 autosomal and 278,985 X-chromosomal variants with MAF\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5% and imputation r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.3 in 6,060 ABCD participants. We identified three independent variant-trait associations and three LD-independent loci (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;3\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;6\\u003c/b\\u003e), including the variant-trait associations between rs10983073 at 9q33.1 and pons absolute volume (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;2.25 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;10\\u003c/sup\\u003e), rs6775519 at 3p25.1 and pons relative volume (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.16 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e), and rs12449302 at 17q11.2 and midbrain absolute volume (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.39 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;10\\u003c/sup\\u003e; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ec).\\u003c/p\\u003e \\u003cp\\u003eWe conducted EUR-GWASs (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected) for brainstem volumetric traits at 10,042,001 autosomal and 390,192 X-chromosomal variants with MAF\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5% and info\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.6 in 35,611 UKBB participants. We found 268 independent variant-trait associations (involving 191 autosomal and four X-chromosomal variants), 200 locus-trait associations, and 91 LD-independent loci (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;4\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;7\\u003c/b\\u003e). Among these variant-trait associations, four variants were missense mutations, linking rs13388394 (\\u003cem\\u003eRASGRP3\\u003c/em\\u003e) to whole brainstem volume and pons absolute volume, rs1805165 (\\u003cem\\u003eEIF2AK3\\u003c/em\\u003e) and rs2234675 (\\u003cem\\u003ePAX3\\u003c/em\\u003e) to pons absolute volume, and rs2273171 (\\u003cem\\u003eSTRN3\\u003c/em\\u003e) to midbrain absolute volume. We also identified one stop-gained variant, linking rs2231142 (\\u003cem\\u003eABCG2\\u003c/em\\u003e) to whole brainstem volume and pons absolute volume.\\u003c/p\\u003e \\u003cp\\u003eWe conducted EUR-GWAS meta-analyses (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected) for the eight brainstem substructure volumes at 9,637,747 autosomal and 316,973 X-chromosomal variants in 41,426 EUR participants (5,815 from ABCD and 35,611 from UKBB). As the previous EUR-GWAS (28,809 individuals from ENIGMA, CHARGE, and UKBB) for whole brainstem volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e included first released UKBB participants, we excluded these participants and reperformed GWAS for the trait in 30,752 UKBB participants. Based on the GWAS summary data of the prior EUR-GWAS (n\\u0026thinsp;=\\u0026thinsp;28,809), UKBB-GWAS (n\\u0026thinsp;=\\u0026thinsp;30,752), and ABCD-GWAS (n\\u0026thinsp;=\\u0026thinsp;6,060), we conducted the EUR-GWAS meta-analysis for whole brainstem volume in 65,621 participants at 9,755,044 autosomal and 317,092 X-chromosomal variants using METAL\\u003csup\\u003e\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e\\u003c/sup\\u003e, which could output summary statistics for variants included in any GWAS, providing the largest number of EUR-GWAS association statistics. In the EUR-GWAS meta-analyses for the brainstem volumetric traits (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected), we identified 367 independent variant-trait associations (involving 276 autosomal and four X-chromosomal variants), 261 locus-trait associations, and 112 LD-independent loci (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;5\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;8\\u003c/b\\u003e). For example, we found a meaningful association between rs803903 at 9q33.1 and medulla relative volume (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;5.39 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;15\\u003c/sup\\u003e; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ed). The lead variant is mapped to \\u003cem\\u003eASTN2\\u003c/em\\u003e, involving in neuronal migration and linking to migraine without aura\\u003csup\\u003e\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e\\u003c/sup\\u003e, characterized by tractus solitarius impairment in medulla\\u003csup\\u003e\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e\\n\\u003ch3\\u003eGenetic findings in cross-ancestry GWASs\\u003c/h3\\u003e\\n\\u003cp\\u003eWith the same approach as EUR-GWAS meta-analyses, we conducted cross-ancestry GWAS meta-analyses (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected) for brainstem volumetric traits in up to 72,717 participants by further including EAS-GWASs (n\\u0026thinsp;=\\u0026thinsp;7,096). In cross-ancestry GWASs, we found 375 independent variant-trait associations (involving 285 autosomal and four X-chromosomal variants), 260 locus-trait associations, and 108 LD-independent loci (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;6\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;9\\u003c/b\\u003e). The cross-ancestry GWASs identified many associations that were not significant in EAS-GWASs and EUR-GWASs. For example, the association between rs3213141 at 20q11.22 and pons absolute volume (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ee) was significant in cross-ancestry GWAS (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;3.13 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) rather than EAS-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;3.76 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;3\\u003c/sup\\u003e) and EUR-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.70 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e).\\u003c/p\\u003e\\n\\u003ch3\\u003ePooling significant associations and identifying new loci\\u003c/h3\\u003e\\n\\u003cp\\u003eAs most of the prior GWASs for brainstem and substructure volumes were conducted in EUR individuals and EUR samples were at least six times larger than EAS samples in this study, we used the EUR-LD reference to pool GWAS results and identify new associations and loci. We pooled all study-wide significant variant-trait associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) from all GWASs and identified 430 independent variant-trait associations, 296 independent locus-trait associations, and 122 LD-independent loci (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ef, g and \\u003cb\\u003eSupplementary Tables\\u0026nbsp;10 and 11\\u003c/b\\u003e). To identify new variant-trait associations and loci, we integrated all previous GWASs for brainstem and substructure volumes\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR13 CR14 CR15\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e\\u003c/sup\\u003e to define the references (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) of known variant-trait (n\\u0026thinsp;=\\u0026thinsp;211) and locus-trait (n\\u0026thinsp;=\\u0026thinsp;162) associations, and LD-independent loci (n\\u0026thinsp;=\\u0026thinsp;74). Using the EUR-LD reference and the same threshold (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e), we defined a new variant-trait association if the variant was 500 kb away from and not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any variants of the same trait in the list of known variant-trait associations; a new locus-trait association when all lead variants in the locus were 500 kb away from and not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any lead variants in the loci of all known locus-trait associations; and a novel locus when all lead variants in the locus were 500 kb away from and not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any lead variants in known loci. By definition, we found 287 new variant-trait associations, 175 new locus-trait associations, and 63 new LD-independent loci. Even using a lenient threshold of \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e to define the known associations (334 known variant-trait and 236 locus-trait associations, and 104 known LD-independent loci), we still found 241 new variant-trait and 150 new locus-trait associations, and 46 new LD-independent loci (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eg \\u003cb\\u003eand Supplementary Tables\\u0026nbsp;10 and 11\\u003c/b\\u003e). For example, the new locus at 3p14.1 was associated with both absolute and relative volumes of SCP (a white matter tract formed by enormous axons), its lead variant rs2306272 is a missense variant of \\u003cem\\u003eLRIG1\\u003c/em\\u003e, which controls axonal extension, guidance, and branching by regulating growth factor signals during neural development\\u003csup\\u003e\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eHeritability of brainstem and substructure volumes\\u003c/h2\\u003e \\u003cp\\u003eAfter excluding the major histocompatibility complex (MHC) region from the genome, we utilized LDSC\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e\\u003c/sup\\u003e to estimate the single nucleotide polymorphism (SNP)-based heritability of the nine brainstem volumetric traits in EAS, EUR, and cross-ancestry populations based on the EAS, EUR, and cross-ancestry GWAS summary data and LD references, respectively. All brainstem volumetric traits showed significant SNP-based heritability (0.17\\u0026ndash;0.40, all \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;1.17 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;29\\u003c/sup\\u003e; \\u003cb\\u003eSupplementary Table\\u0026nbsp;12\\u003c/b\\u003e) in the EUR and cross-ancestry individuals, although the SNP-based heritability of two traits was not significant (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.05) in EAS individuals.\\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003eCross-ancestry genetic correlations of brainstem volumetric traits\\u003c/h3\\u003e\\n\\u003cp\\u003eAfter excluding the genetic variants in the MHC region, we used Popcorn\\u003csup\\u003e\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e\\u003c/sup\\u003e to calculate the genetic-effect correlation (without considering MAF information) of each brainstem volumetric trait between EAS and EUR, and identified relatively high correlations (0.44\\u0026ndash;1.13; \\u003cb\\u003eSupplementary Table\\u0026nbsp;13\\u003c/b\\u003e). We also calculated genetic-impact correlations (considering MAF information) for these brainstem volumetric traits between EAS and EUR, and also found relatively high correlations (0.49\\u0026ndash;1.12).\\u003c/p\\u003e\\n\\u003ch3\\u003eAncestry-shared and ancestry-specific variant-trait associations\\u003c/h3\\u003e\\n\\u003cp\\u003eTo identify the ancestry-shared and ancestry-specific variant-trait associations for nine brainstem volumetric traits, we used the Cochran's \\u003cem\\u003eQ\\u003c/em\\u003e (CQ) test to quantify effect size differences between EAS and EUR for the pooled variant-trait associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) from EAS-GWASs, EUR-GWASs, and cross-ancestry GWASs. Of the 430 pooled independent variant-trait associations, the variants of 369 associations were included in both CHIMGEN and UKBB datasets, which were used for comparing the effect size differences between EAS and EUR. Based on the CQ-tests, we defined the associations with \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.05 as ancestry-shared associations and \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;1.36 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e (Bonferroni corrected) as ancestry-specific associations. We found three ancestry-specific associations and 313 ancestry-shared associations (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ea, b and \\u003cb\\u003eSupplementary Table\\u0026nbsp;14\\u003c/b\\u003e). The variants of the three ancestry-specific associations and 300 ancestry-shared associations were also included in the ABCD dataset. Using ABCD to replace UKBB data, we replicated all three ancestry-specific associations (CQ-test: \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.0167, Bonferroni corrected) and 292/300 ancestry-shared associations (CQ-test: \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.05) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eb and \\u003cb\\u003eSupplementary Table\\u0026nbsp;15\\u003c/b\\u003e). The three ancestry-specific associations were found between rs111883632 and pons absolute volume, rs606599 and pons relative volume, and rs151057105 and midbrain absolute volume (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ec). As an example of ancestry-shared associations, the association (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ed) between rs16863657 and pons absolute volume was significant in UKBB-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;6.72 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;16\\u003c/sup\\u003e) and nearly genome-wide significant in CHIMGEN-GWAS (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;6.88 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e). The variant is at the upstream of \\u003cem\\u003ePAX3\\u003c/em\\u003e, a transcription factor regulating fetal neural development\\u003csup\\u003e\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eStatistical fine-mapping\\u003c/h2\\u003e \\u003cp\\u003eFor each locus of the pooled locus-trait associations, we used the matched LD reference to perform statistical fine-mapping by estimating the posterior probability (PP) of each variant to be a causal variant (PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.8) using the probabilistic annotation integrator (PAINTOR) tool\\u003csup\\u003e\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e\\u003c/sup\\u003e with the Markov chain Monte Carlo (MCMC) model that allows multiple causal variants. In the 296 pooled locus-trait associations, one was significant in all three categories (EAS, EUR, and cross-ancestry) of GWASs, 36 significant only in EUR-GWASs, 36 significant only in cross-ancestry GWASs, 223 significant in both EUR and cross-ancestry GWASs (\\u003cb\\u003eSupplementary Table\\u0026nbsp;10\\u003c/b\\u003e). We conducted statistical fine-mapping for one locus based on EAS-GWAS using EAS-LD reference, 260 loci based on EUR-GWASs using EUR-LD reference, and 260 loci based on cross-ancestry GWASs using merged LD reference, respectively. We found 250 unique causal variants for brainstem volumetric traits, of which 25 were identified only from EUR analyses, 223 only from cross-ancestry analyses, and two from both EUR and cross-ancestry analyses (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ea \\u003cb\\u003eand Supplementary Table\\u0026nbsp;16\\u003c/b\\u003e). For the 223 locus-trait associations significant in both EUR and cross-ancestry GWASs, we also conducted PAINTOR with one causal variant assumption and reported the results in \\u003cb\\u003eSupplementary Table\\u0026nbsp;17\\u003c/b\\u003e. We then used the Wilcoxon rank-sum test (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) to investigate whether cross-ancestry fine-mapping can reduce the 95% credible sets for these trait-associated loci compared to EUR-specific fine-mapping. We found that the 95% credible sets (median\\u0026thinsp;=\\u0026thinsp;2) from cross-ancestry analyses were significantly smaller (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;2.20 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;16\\u003c/sup\\u003e; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003eb) than those (median\\u0026thinsp;=\\u0026thinsp;13) from EUR analyses. For example, although the locus-trait association between 2p11.2 and pons absolute volume was significant in all GWASs, we failed to identify any causal variants based on the EAS-GWAS and EUR-GWAS, but found two causal variants (rs2090217: PP\\u0026thinsp;=\\u0026thinsp;1.00; rs335124: PP\\u0026thinsp;=\\u0026thinsp;1.00) based on the cross-ancestry GWAS (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ec).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eFunctional annotations\\u003c/h2\\u003e \\u003cp\\u003eTo investigate functional consequences of genetic variants associated with brainstem volumetric traits, we used FUMA\\u003csup\\u003e\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e\\u003c/sup\\u003e to perform functional annotations for the unique variants with PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1 in statistical fine-mapping. Based on the genomic location and functional consequence, we categorized the four variants with PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1 in the only locus-trait association identified by EAS-GWAS, and found one variant in the intronic region, two variants in the intergenic region, and one variant as a missense variant of \\u003cem\\u003eEIF2AK3\\u003c/em\\u003e (\\u003cb\\u003eSupplementary Table\\u0026nbsp;18\\u003c/b\\u003e). We also categorized 2,076 unique variants with PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1 in the 260 locus-trait associations identified by EUR-GWASs, and found that the variants were mainly in the intronic (44.9%) and intergenic (37.7%) regions (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ed and \\u003cb\\u003eSupplementary Table\\u0026nbsp;19\\u003c/b\\u003e). We also found 28 variants in the coding sequence and 48 variants in the untranslated region (UTR), of which three were stop gained and 12 were missense variants. For instance, rs2231142 (4q22.1) associated with brainstem (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;4.37 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;14\\u003c/sup\\u003e, PP\\u0026thinsp;=\\u0026thinsp;0.16) and pons absolute (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;8.70 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;11\\u003c/sup\\u003e, PP\\u0026thinsp;=\\u0026thinsp;0.18) volumes is a stop gained variant of \\u003cem\\u003eABCG2\\u003c/em\\u003e, a regulator of self-renewal of neural stem cells\\u003csup\\u003e\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e\\u003c/sup\\u003e. As for the 3,944 unique genetic variants (PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1) in 260 locus-trait associations identified by cross-ancestry GWASs, they were also mainly in the intronic (46.0%) and intergenic (35.9%) regions (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig3\\\" class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003ed and \\u003cb\\u003eSupplementary Table\\u0026nbsp;20\\u003c/b\\u003e). We found 40 coding and 70 UTR variants, including two splice region variants within 2bp of the splicing junction, one stop-gained variant, and 24 missense variants. For the 5,846 unique variants with PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1, we also used combined annotation-dependent depletion (CADD) score\\u003csup\\u003e\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e\\u003c/sup\\u003e to assess deleteriousness of 5,844 variants included in the CADD database, from which we identified 255 pathogenic variants (CADD score\\u0026thinsp;\\u0026gt;\\u0026thinsp;12.37; \\u003cb\\u003eSupplementary Tables\\u0026nbsp;18\\u0026ndash;20)\\u003c/b\\u003e. We then used RegulomeDB (RDB) score\\u003csup\\u003e\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e\\u003c/sup\\u003e to identify variants with regulatory function, and found 112 most likely regulatory variants (RDB\\u0026thinsp;\\u0026le;\\u0026thinsp;1f; \\u003cb\\u003eSupplementary Tables\\u0026nbsp;18\\u0026ndash;20\\u003c/b\\u003e) from 5,152 fine-mapped variants included in the RDB database.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec13\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eIdentifying genes associated with brainstem volumetric traits\\u003c/h2\\u003e \\u003cp\\u003eWe identified genes associated with brainstem volumetric traits based on the location, gene expression, and chromatin interaction using the gene-based, transcriptome-wide, and chromatin interaction association analyses, respectively.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGene-based association analyses.\\u003c/b\\u003e We conducted the gene-based association analyses to identify genes associated with brainstem volumetric traits (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/17,550/9\\u0026thinsp;=\\u0026thinsp;3.16 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e, Bonferroni correction for 17,550 genes and nine traits) based on the EUR-GWASs using Multivariate Analysis of Genomic Annotation (MAGMA)\\u003csup\\u003e\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e\\u003c/sup\\u003e. We found 189 unique genes associated with brainstem volumetric traits, including 84 with whole brainstem volume, 49 with medulla absolute volume, 29 with medulla relative volume, 81 with pons absolute volume, 51 with pons relative volume, 51 with midbrain absolute volume, 29 with midbrain relative volume, ten with SCP absolute volume, and three with SCP relative volume (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;7\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;21\\u003c/b\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eTranscriptome-wide association analyses.\\u003c/b\\u003e Based on the EUR-GWAS summary data of nine brainstem volumetric traits and eQTL data of 13 brain tissues provided by GTEx\\u003csup\\u003e\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e\\u003c/sup\\u003e, we used S-PrediXcan\\u003csup\\u003e\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e\\u003c/sup\\u003e to perform the transcriptome-wide association studies (TWASs) between predicted gene expression in each tissue and each brainstem volumetric trait (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;8\\u003c/b\\u003e). Based on the gene-trait associations of 13 tissues, we then used S-MultiXcan\\u003csup\\u003e\\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e\\u003c/sup\\u003e to conduct the multi-tissue TWAS to test the joint effects of gene expression on brainstem volumetric traits across 13 brain tissues (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/15,375/9\\u0026thinsp;=\\u0026thinsp;3.61 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e, Bonferroni correction for 15,375 genes and nine traits). We identified 151 unique genes whose brain expression was associated with brainstem volumetric traits, including 75 with whole brainstem volume, 37 with medulla absolute volume, 27 with medulla relative volume, 56 with pons absolute volume, 42 with pons relative volume, 39 with midbrain absolute volume, 19 with midbrain relative volume, 12 with SCP absolute volume, and one with SCP relative volume (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;9 and Supplementary Table\\u0026nbsp;22\\u003c/b\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eChromatin interaction association analyses.\\u003c/b\\u003e Hi-C-coupled MAGMA (H-MAGMA)\\u003csup\\u003e\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e\\u003c/sup\\u003e was used to identify the genes associated with brainstem volumetric traits based on the chromatin interaction profiles in six brain tissues and cells, including adult brain\\u003csup\\u003e\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e\\u003c/sup\\u003e, fetal brain\\u003csup\\u003e\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e\\u003c/sup\\u003e, cortical neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e\\u003c/sup\\u003e, induced pluripotent stem cells (iPSC) derived astrocyte\\u003csup\\u003e\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e\\u003c/sup\\u003e, iPSC derived neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e\\u003c/sup\\u003e, and midbrain DA neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e\\u003c/sup\\u003e. We identified 2,315 significant gene-trait associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/107,575/9\\u0026thinsp;=\\u0026thinsp;5.16 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e, Bonferroni correction for 107,575 genes for six tissues and cells and nine traits) (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;10\\u003c/b\\u003e and \\u003cb\\u003eSupplementary Table\\u0026nbsp;23\\u003c/b\\u003e). These associations contained 465 unique genes, including 231 with whole brainstem volume, 121 with medulla absolute volume, 73 with medulla relative volume, 165 with pons absolute volume, 132 with pons relative volume, 114 with midbrain absolute volume, 60 with midbrain relative volume, 17 with SCP absolute volume, and seven with SCP relative volume.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003ePooling genes from three association analyses.\\u003c/b\\u003e By integrating genes identified by the three approaches, we prioritized 550 unique genes, including 267 associated with whole brainstem volume, 154 with medulla absolute volume, 86 with medulla relative volume, 194 with pons absolute volume, 142 with pons relative volume, 134 with midbrain absolute volume, 75 with midbrain relative volume, 28 with SCP absolute volume, and eight with SCP relative volume (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003ea).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec14\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003ePathways associated with brainstem volumetric traits\\u003c/h2\\u003e \\u003cp\\u003eBased on pathways from Gene Ontology (GO)\\u003csup\\u003e\\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e\\u003c/sup\\u003e biological processes and Reactome\\u003csup\\u003e\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e\\u003c/sup\\u003e, we performed pathway enrichment analyses for the prioritized 550 genes associated with brainstem volumetric traits using g:Profiler\\u003csup\\u003e\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e\\u003c/sup\\u003e (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://biit.cs.ut.ee/gprofiler/gost\\u003c/span\\u003e\\u003cspan address=\\\"https://biit.cs.ut.ee/gprofiler/gost\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). We identified 138 significant enrichment pathways (\\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05, g:SCS (Set Counts and Sizes) corrected; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig4\\\" class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003eb \\u003cb\\u003eand Supplementary Table\\u0026nbsp;24\\u003c/b\\u003e), mainly including metabolic process regulation, animal organ development, activity kinase regulation, neurogenesis, neuron generation, cell proliferation, and neuron differentiation regulation.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec15\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eShared and distinct genetic architectures between brainstem substructures\\u003c/h2\\u003e \\u003cp\\u003eAlthough sharing some white matter tracts, the three brainstem substructures (medulla, pons, and midbrain) are originated from different embryonic structures (medulla and pons from hindbrain and midbrain from mesencephalon) and contain different gray matter nuclei and white matter tracts, indicating the coexistence of shared and distinct genetic architectures. Thus, we investigated the shared and distinct genetic loci, genes, and enriched pathways between each pair of the three brainstem substructures.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct genetic loci between brainstem substructures.\\u003c/b\\u003e Among the 122 LD-independent loci associated with brainstem volumetric traits, 111 were associated with the absolute or relative volume of medulla, pons, or midbrain. From the 111 loci, we searched for loci that were associated with volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume based on the independent locus-trait associations. We identified 62 substructure-specific loci (17 for medulla, 27 for pons, and 18 for midbrain) and 49 substructure-shared loci (20 shared by all substructures, six by medulla and pons, five by medulla and midbrain, and 18 by pons and midbrain) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ea and \\u003cb\\u003eSupplementary Table\\u0026nbsp;25\\u003c/b\\u003e). For example, we found a locus (rs9428966) at 1q43 shared by all brainstem substructures, a medulla-specific locus (rs12448813) at 16q23.2, a pons-specific locus (rs72927168) at 2q31.1, and a midbrain-specific locus (rs2331753) at 1q25.3 (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003eb).\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct genes between brainstem substructures.\\u003c/b\\u003e Among the prioritized 550 genes associated with brainstem volumetric traits, 468 were associated with the absolute or relative volume of medulla, pons, or midbrain. From the 468 prioritized genes, we searched for genes prioritized for one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume. We identified 295 substructure-specific genes (100 for medulla, 123 for pons, and 72 for midbrain) and 173 substructure-shared genes (64 shared by all substructures, 42 shared by medulla and pons, 13 shared by medulla and midbrain, and 54 shared by pons and midbrain) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ec, \\u003cb\\u003eSupplementary Table\\u0026nbsp;26\\u003c/b\\u003e). For example, \\u003cem\\u003eUBE4B\\u003c/em\\u003e was a medulla-specific gene involving axon regrowth\\u003csup\\u003e\\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e\\u003c/sup\\u003e; \\u003cem\\u003eFOXO6\\u003c/em\\u003e was a pons-specific gene involving the regulation of synaptic function\\u003csup\\u003e\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e\\u003c/sup\\u003e; and \\u003cem\\u003eFRAT1\\u003c/em\\u003e and \\u003cem\\u003eFRAT2\\u003c/em\\u003e were midbrain-specific genes affecting midbrain morphogenesis by regulating WNT signaling\\u003csup\\u003e\\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct biological pathways between brainstem substructures.\\u003c/b\\u003e We pooled the prioritized genes for medulla, pons, and midbrain regardless of absolute or relative volume, and obtained 219 genes for medulla volume, 283 for pons volume, and 203 for midbrain volume. We then used g:Profiler\\u003csup\\u003e\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e\\u003c/sup\\u003e to conduct pathway enrichment analyses for the three groups of genes, respectively. We found 90 unique enrichment pathways (\\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05, g:SCS corrected), including 33 pathways for medulla, 85 for pons, and one for midbrain volumes (\\u003cb\\u003eSupplementary Table\\u0026nbsp;27\\u003c/b\\u003e). Then, we searched for the pathways enriched by one, two, or three brainstem substructures, and found 62 substructure-specific pathways (five for medulla and 57 for pons) and 28 substructure-shared pathways (one shared by all substructures and 27 shared by medulla and pons; Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003ed) (\\u003cb\\u003eSupplementary Table\\u0026nbsp;28\\u003c/b\\u003e). The pathway shared by all substructures was the regulation of developmental process. A medulla-specific pathway was the noncanonical activation of NOTCH3, which regulates neurogenesis and neuronal differentiation in hindbrain\\u003csup\\u003e\\u003cspan citationid=\\\"CR54\\\" class=\\\"CitationRef\\\"\\u003e54\\u003c/span\\u003e\\u003c/sup\\u003e. Two pons-specific enrichment pathways contained \\u003cem\\u003eHOX\\u003c/em\\u003e, involving in the regulation of pontine neuronal migration\\u003csup\\u003e\\u003cspan citationid=\\\"CR55\\\" class=\\\"CitationRef\\\"\\u003e55\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec16\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eGenetic architectures shared by brainstem volumetric traits and other phenotypes\\u003c/h2\\u003e \\u003cp\\u003eBased on the available GWAS summary statistics for non-imaging phenotypes that are potentially associated with brainstem functions, we used three complementary methods (genetic correlation, genetic colocalization, and condFDR/conjFDR) to identify shared genetic architectures between brainstem volumetric traits and non-imaging phenotypes. As most GWASs are conducted in EUR population and most approaches require GWAS samples from same ancestry, we included 26 non-imaging phenotypes (\\u003cb\\u003eSupplementary Table\\u0026nbsp;29\\u003c/b\\u003e) with EUR-GWAS summary statistics and conducted the three genetic sharing analyses based on the EUR-GWAS summary data for brainstem volumetric traits.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGenetic correlation analyses\\u003c/b\\u003e. We used LDSC\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e\\u003c/sup\\u003e to calculate the genetic correlations between nine brainstem volumetric traits and 26 non-imaging phenotypes based on the EUR-GWAS summary data. Using a Bonferroni-corrected \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/26/9\\u0026thinsp;=\\u0026thinsp;2.14 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e, we identified six significant genetic correlations between brainstem volumetric traits and non-imaging phenotypes (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ea and \\u003cb\\u003eSupplementary Table\\u0026nbsp;30\\u003c/b\\u003e), including genetic correlations of whole brainstem volume (r = -0.13, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.85 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;6\\u003c/sup\\u003e), midbrain absolute volume (r = -0.18, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.59 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e), and midbrain relative volume (r = -0.13, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;4.58 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;6\\u003c/sup\\u003e) with ADHD, whole brainstem volume (r = -0.08, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.00 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e) and midbrain absolute volume with MDD (r = -0.11, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;1.60 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;5\\u003c/sup\\u003e), and midbrain absolute volume with PD (r\\u0026thinsp;=\\u0026thinsp;0.16, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;2.00 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e).\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGenetic colocalization analyses.\\u003c/b\\u003e We used coloc\\u003csup\\u003e\\u003cspan citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e\\u003c/sup\\u003e to identify the genetic loci shared by brainstem volumetric traits and non-imaging phenotypes based on the EUR-GWAS summary data. Among the 260 loci associated with brainstem volumetric traits in EUR-GWASs (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e), we found genetic colocalization (PP.H4\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.8) between eight brainstem volumetric traits and ten non-imaging phenotypes at 59 genetic loci (19 LD-independent loci) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eb and \\u003cb\\u003eSupplementary Table\\u0026nbsp;31\\u003c/b\\u003e), including the non-imaging phenotypes of brain disorders (MDD, SCZ, and AD), cardiovascular functions (resting heart rate, systolic and diastolic blood pressure, and pulse pressure), circadian rhythms (chronotype and morning person), and subjective well-being. For instance, the two circadian rhythms phenotypes showed colocalizations (PP.H4\\u0026thinsp;=\\u0026thinsp;0.94\\u0026ndash;0.99) at 19p13.11 with whole brainstem volume, medulla absolute volume, medulla relative volume, and pons relative volume. SCZ showed colocalizations at 2q33.1 with whole brainstem volume (PP.H4\\u0026thinsp;=\\u0026thinsp;0.89) and medulla absolute volume (PP.H4\\u0026thinsp;=\\u0026thinsp;0.84). The lead SNPs in the locus are eQTLs of \\u003cem\\u003eTYW5\\u003c/em\\u003e, a regulator of neurodevelopment\\u003csup\\u003e\\u003cspan citationid=\\\"CR58\\\" class=\\\"CitationRef\\\"\\u003e58\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eCondFDR/conjFDR analyses.\\u003c/b\\u003e Based on the EUR-GWAS summary data, we conducted condFDR/conjFDR analyses using pleioFDR (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://github.com/precimed/pleiofdr\\u003c/span\\u003e\\u003cspan address=\\\"https://github.com/precimed/pleiofdr\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e59\\u003c/sup\\u003e to identify genetic variants shared by each pair of nine brainstem volumetric traits and 26 non-imaging phenotypes. We first generated the conditional Q-Q plots to assess the polygenetic enrichment for each brainstem volumetric trait conditioned on \\u003cem\\u003eP\\u003c/em\\u003e-values of the association with each non-imaging phenotype. Among the 234 trait-phenotype pairs, 116 pairs showed polygenetic enrichment with successive leftward shifts from the null distribution (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;11\\u003c/b\\u003e), involving eight brainstem volumetric traits and 23 non-imaging phenotypes. Then, we performed the conjFDR analysis to identify the shared genetic variants (conjFDR\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) between each trait-phenotype pair, in which the conjFDR value of each variant was defined as the maximal FDR value of this variant in the two mutual condFDR analyses. Among the 234 potential trait-phenotype pairs, 201 pairs showed shared genetic variants (conjFDR\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05; \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;12\\u003c/b\\u003e), involving nine brainstem volumetric traits and 25 non-imaging phenotypes. These trait-phenotype pairs shared 7,263 variants and 5,826 loci (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ec and \\u003cb\\u003eSupplementary Table\\u0026nbsp;32\\u003c/b\\u003e). Numerous genetic sharing findings were consistent with genetic correlation and/or colocalization analyses. For example, both conjFDR and colocalization analyses demonstrated that the locus (7q21.2) was shared between brainstem volumetric traits (absolute volumes of medulla and midbrain) and blood pressure phenotypes (systolic blood pressure and pulse pressure) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003ed). Its lead SNP rs42039 is located in the UTR3 region of \\u003cem\\u003eCDK6\\u003c/em\\u003e, a key player in cell cycle progression\\u003csup\\u003e\\u003cspan citationid=\\\"CR60\\\" class=\\\"CitationRef\\\"\\u003e60\\u003c/span\\u003e\\u003c/sup\\u003e. We also found genetic sharing between brainstem volumetric traits and 13 non-imaging phenotypes that were observed in neither genetic correlation nor genetic colocalization analyses.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eBy including all available neuroimaging genetics data, especially those from the ABCD and CHIMGEN studies, we conducted cross-ancestry GWAS meta-analyses for whole brainstem volume in 72,717 individuals and substructure volumes in 48,522 individuals. We identified 122 loci and 550 genes associated with nine brainstem volumetric traits, including 46 new loci. We discovered three (0.8%) ancestry-specific and 292 (79.1%) ancestry-shared associations, consistent with high cross-ancestry genetic correlations between EAS and EUR. We provided new evidence for the merit of cross-ancestry fine-mapping by identifying 225 causal variants (PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.8), which are much greater than 27 from EUR fine-mapping. We revealed shared and distinct loci, genes, and pathways for midbrain, pons, and medulla volumes, and shared genetic architectures of the brainstem volumetric traits with 25 brainstem-related physiological and pathological phenotypes, in line with the importance of the brainstem in the control of heart rate, blood pressure, circadian rhythms, as well as its associations with neuropsychiatric disorders, such us AD, ADHD, MDD, PD, and SCZ.\\u003c/p\\u003e \\u003cp\\u003eThe first contribution of this study is the discovery of nearly two times of variant-trait and locus-trait associations compared to those significant at \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e in the previous GWASs for brainstem volumetric traits\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR13 CR14 CR15\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e\\u003c/sup\\u003e. Even compared to associations significant at \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e in previous GWASs, we still found 241 new variant-trait and 150 new locus-trait associations and 46 new LD-independent loci. The large number of new genetic discoveries may improve the understanding of the genetic architectures of brainstem and substructure volumes. For instance, in the cross-ancestry GWAS meta-analysis for pons absolute volume, we found a new locus-trait association at 20q11.22. Its lead variant rs3213141 is in the upstream of \\u003cem\\u003eE2F1\\u003c/em\\u003e, a cell cycle suppressor regulating neuronal survival and death by interacting with protein product of the retinoblastoma (RB) gene\\u003csup\\u003e\\u003cspan citationid=\\\"CR61\\\" class=\\\"CitationRef\\\"\\u003e61\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR62\\\" class=\\\"CitationRef\\\"\\u003e62\\u003c/span\\u003e\\u003c/sup\\u003e. RB1-deficient mutation can lead to neuronal apoptosis in the hindbrain\\u003csup\\u003e\\u003cspan citationid=\\\"CR63\\\" class=\\\"CitationRef\\\"\\u003e63\\u003c/span\\u003e\\u003c/sup\\u003e including the pons as a main component. Another new locus-trait association was found between 12q13.13 and pons relative volume, its lead variant rs56098072 is an eQTL of \\u003cem\\u003eHOX3\\u003c/em\\u003e, interacting with other \\u003cem\\u003eHOX\\u003c/em\\u003e genes to specify the rhombomere identity in the developing hindbrain\\u003csup\\u003e\\u003cspan citationid=\\\"CR64\\\" class=\\\"CitationRef\\\"\\u003e64\\u003c/span\\u003e\\u003c/sup\\u003e. We also found a locus at Xp11.22, the only new locus on X-chromosome, which was associated with whole brainstem and pons absolute volumes. Its lead variant rs7060542 is an eQTL of the CASK-interacting nucleosome assembly protein (\\u003cem\\u003eCINAP\\u003c/em\\u003e), influencing brain development by binding to \\u003cem\\u003eCASK\\u003c/em\\u003e\\u003csup\\u003e\\u003cspan citationid=\\\"CR65\\\" class=\\\"CitationRef\\\"\\u003e65\\u003c/span\\u003e\\u003c/sup\\u003e, while \\u003cem\\u003eCASK\\u003c/em\\u003e mutation can lead to an X-linked brain malformation with brainstem hypoplasia\\u003csup\\u003e\\u003cspan citationid=\\\"CR66\\\" class=\\\"CitationRef\\\"\\u003e66\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003eThe second contribution is the novel genetic findings derived from cross-ancestry analyses. We revealed high cross-ancestry genetic correlations (0.44\\u0026ndash;1.13) in brainstem volumetric traits between EUR and EAS populations, consistent with 97 times more ancestry-shared associations (n\\u0026thinsp;=\\u0026thinsp;292) than ancestry-specific associations (n\\u0026thinsp;=\\u0026thinsp;3). These results indicate that EAS and EUR individuals have similar genetic architectures for brainstem and substructure volumes. We also conducted statistical fine-mapping for the locus-trait associations of brainstem volumetric traits, and found 250 causal variants, among which 223 (89.2%) were only identified by cross-ancestry fine-mapping. These results further highlight the value of cross-ancestry fine-mapping in detecting causal variants for brain imaging phenotypes\\u003csup\\u003e\\u003cspan citationid=\\\"CR67\\\" class=\\\"CitationRef\\\"\\u003e67\\u003c/span\\u003e\\u003c/sup\\u003e. For instance, rs17010085 was identified as a causal variant (PP\\u0026thinsp;=\\u0026thinsp;0.96-1.00) for whole brainstem volume and pons absolute volume in both EUR and cross-ancestry fine-mapping. The variant is an eQTL of \\u003cem\\u003eRYBP\\u003c/em\\u003e, which is involved in neural differentiation\\u003csup\\u003e\\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003eThe third contribution of this study is the discovery of shared and distinct genetic architectures (loci, genes, and pathways) between brainstem substructure (midbrain, pons, and medulla) volumes. We found 20 loci and 64 genes shared by all substructures, mainly involving neural development processes shared by brain structures. For example, we found a locus at 1q43 shared by all brainstem substructures. Its lead SNP rs9428966 is in the UTR3 region of \\u003cem\\u003eAKT3\\u003c/em\\u003e, an AKT kinase involving a wide variety of biological processes, including brain development\\u003csup\\u003e\\u003cspan citationid=\\\"CR69\\\" class=\\\"CitationRef\\\"\\u003e69\\u003c/span\\u003e\\u003c/sup\\u003e, neuronal survival\\u003csup\\u003e\\u003cspan citationid=\\\"CR70\\\" class=\\\"CitationRef\\\"\\u003e70\\u003c/span\\u003e\\u003c/sup\\u003e, myelination\\u003csup\\u003e\\u003cspan citationid=\\\"CR71\\\" class=\\\"CitationRef\\\"\\u003e71\\u003c/span\\u003e\\u003c/sup\\u003e, and the regulation of blood pressure\\u003csup\\u003e\\u003cspan citationid=\\\"CR72\\\" class=\\\"CitationRef\\\"\\u003e72\\u003c/span\\u003e\\u003c/sup\\u003e and breathing\\u003csup\\u003e\\u003cspan citationid=\\\"CR73\\\" class=\\\"CitationRef\\\"\\u003e73\\u003c/span\\u003e\\u003c/sup\\u003e. We also found 62 loci and 295 genes that were specific to one brainstem substructure, providing more specific insight into the genetic architecture of each brainstem substructure. For instance, a locus at 16q23.2 was a medulla-specific locus and its lead variant rs12448813 is an eQTL of the giant axonal neuropathy (GAN) gene. \\u003cem\\u003eGAN\\u003c/em\\u003e encodes gigaxonin, affecting neuronal survival by regulating the degradation of the light chain of microtubule-associated protein 1B (MAP1B)\\u003csup\\u003e\\u003cspan citationid=\\\"CR74\\\" class=\\\"CitationRef\\\"\\u003e74\\u003c/span\\u003e\\u003c/sup\\u003e. The disorganization of the neurofilament network due to \\u003cem\\u003eGAN\\u003c/em\\u003e mutation is associated GAN\\u003csup\\u003e\\u003cspan citationid=\\\"CR75\\\" class=\\\"CitationRef\\\"\\u003e75\\u003c/span\\u003e\\u003c/sup\\u003e. These findings are consistent with the compact axon arrangement in the medulla. Among the 33 enrichment pathways for medulla volume, 28 (84.9%) were also enriched for pons volume, in line with their shared developmental origin from the hindbrain.\\u003c/p\\u003e \\u003cp\\u003eThe last contribution is the identification of shared genetic architectures between brainstem volumetric traits and brainstem-related non-imaging phenotypes. A previous study has conducted genetic correlation and conjunctional FDR analyses to investigate the shared genetic architectures between brainstem volumetric traits and eight common brain disorders\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u003c/sup\\u003e, but failed to identify any significant genetic correlations after multiple testing correction, although overlapped genetic loci are observed for all eight disorders. We extended the study by including 26 brainstem-related non-imaging phenotypes and by further conducting genetic colocalization analyses, generating many novel findings. We found genetic correlations of midbrain volume with PD, MDD and ADHD, all of which are linked to abnormal DA levels. The findings agree with the role of midbrain DA neurons in the regulation of voluntary movement, reward, salience, motivation, and emotion\\u003csup\\u003e\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e\\u003c/sup\\u003e. In addition to genetic colocalization with brain disorders (MDD, SCZ, and AD), we also identified genetic colocalization between brainstem volumetric traits and physiological phenotypes (heart rate, blood pressure, chronotype, and morning person) associated with cardiovascular function and circadian rhythm, well-known functions of the brainstem\\u003csup\\u003e\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR76\\\" class=\\\"CitationRef\\\"\\u003e76\\u003c/span\\u003e\\u003c/sup\\u003e. The shared genetic architectures with physiological phenotypes were further confirmed by the conjunctional FDR analyses, in which brainstem volumetric traits shared more loci with physiological phenotypes than any other phenotypes.\\u003c/p\\u003e \\u003cp\\u003eSeveral limitations should be noted when interpreting our findings. First, although we included 7,069 Chinese Han individuals from the CHIMGEN study, the sample size of EAS individuals is much smaller than that of EUR individuals, which may bias the cross-ancestry GWASs for brainstem volumetric traits and the effect size comparisons between EAS and EUR populations. More non-EUR individuals from EAS and other ancestral populations should be included in the future cross-ancestry GWASs. Second, despite we controlled for the effect of age in GWASs, we cannot exclude the bias from the age differences among the participants from ABCD (aged 8\\u0026ndash;11), CHIMGEN (aged 18\\u0026ndash;30) and UKBB (aged 40\\u0026ndash;70). Third, we reported the study-wide significant genetic associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) for brainstem volumetric traits, but we cannot completely exclude the false-positive findings due to lacking independent data replication.\\u003c/p\\u003e \\u003c/div\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cdiv id=\\\"Sec18\\\" class=\\\"Section2\\\"\\u003e\\u003cdiv id=\\\"Sec19\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eParticipants and data preparation for GWASs\\u003c/h2\\u003e \\u003cp\\u003eIn GWASs for brainstem volumetric traits, we included three sets of raw neuroimaging genetics datasets (CHIMGEN, ABCD, and UKBB) and one set of GWAS summary data for whole brainstem volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e from ENIGMA, CHARGE, and UKBB datasets. In line with prior GWASs for brainstem substructure volumes\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e\\u003c/sup\\u003e, we used the same Bayesian approach\\u003csup\\u003e\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u003c/sup\\u003e to obtain the absolute volumes of whole brainstem, medulla, pons, midbrain, and superior cerebellar peduncle (SCP) of CHIMGEN, ABCD, and UKBB participants. The obtained whole brainstem (including SCP) volume was included as an additional covariate to perform GWASs for relative volumes of medulla, pons, midbrain, and SCP. As the previous GWAS\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e used the automatic subcortical segmentation method to obtain whole brainstem (not including SCP) volume, we also applied the same approach to calculate whole brainstem volume for each CHIMGEN, UKBB, or ABCD participant, which was used in GWASs for whole brainstem volume throughout our study. As the previous GWAS for whole brainstem volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e included first released UKBB data, we excluded these participants and reperformed the UKBB-GWAS for whole brainstem volume, and utilized the obtained GWAS summary statistics in the meta-analyses with the prior study\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e. The number of participants included in each GWAS are shown in \\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003e\\u003cb\\u003eCHMGEN participants and data preparation\\u003c/b\\u003e. All the EAS participants were recruited from the CHIMGEN study (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://chimgen.tmu.edu.cn/\\u003c/span\\u003e\\u003cspan address=\\\"http://chimgen.tmu.edu.cn/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), which collected genomic and neuroimaging data from 7,306 healthy Chinese Han participants aged 18\\u0026ndash;30 years from 32 centers using the predefined inclusion and exclusion criteria (\\u003cb\\u003eSupplementary Table\\u0026nbsp;33\\u003c/b\\u003e). The CHIMGEN study was approved by the Medical Research Ethics Committees of Tianjin Medical University General Hospital and all other institutions, and written informed consent was obtained from each participant. Among the 7,306 participants, 7,195 participants with DNA samples were genotyped by Illumina ASA-750K (Asian Screening Array) that was specially designed for Asian individuals. PLINK v2.084\\u003csup\\u003e77\\u003c/sup\\u003e (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttp://www.cog-genomics.org/plink2\\u003c/span\\u003e\\u003cspan address=\\\"http://www.cog-genomics.org/plink2\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) was applied for quality control of genetic data. Details for sample-level and variant-level quality control, principal component analysis (PCA), and genetic data imputation are provided in our prior studies\\u003csup\\u003e\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR67\\\" class=\\\"CitationRef\\\"\\u003e67\\u003c/span\\u003e\\u003c/sup\\u003e. After quality control, we included 7,163 participants and 8,790,144 imputed autosomal and 227,168 X-chromosomal bi-allelic variants (MAF\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5%, info\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.6, and \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026ge;\\u0026thinsp;1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e in Hardy\\u0026ndash;Weinberg equilibrium (HWE)) in EAS-GWASs. The brain structural MRI data were acquired by ten types of 3.0-Tesla MRI scanners and 12 sets of scanning parameters (\\u003cb\\u003eSupplementary Table\\u0026nbsp;34\\u003c/b\\u003e). After excluding 61 participants without qualified structural MRI data, we calculated the volumes of whole brainstem, medulla, pons, midbrain, and SCP for the remaining 7,102 participants using the Bayesian segmentation algorithm\\u003csup\\u003e\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u003c/sup\\u003e implemented in FreeSurfer v7.0 (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://surfer.nmr.mgh.harvard.edu\\u003c/span\\u003e\\u003cspan address=\\\"https://surfer.nmr.mgh.harvard.edu\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). With FreeSurfer v7.0, we also used the automatic subcortical segmentation method to obtain the whole brainstem volume (not including SCP). For each of the five brainstem volumetric traits, we removed the participants with volumes greater than five times the median absolute deviation (MAD) from the median value, and finally included 7,094\\u0026thinsp;\\u0026minus;\\u0026thinsp;7,096 participants in EAS-GWASs. The quality control procedures for CHIMGEN data are presented in \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;13\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003e\\u003cb\\u003eABCD participants and data preparation.\\u003c/b\\u003e The ABCD study (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://abcdstudy.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://abcdstudy.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e) is a publicly available longitudinal dataset containing over 10,000 participants aged 9\\u0026ndash;10 years at their baseline assessment from 21 research centers. Procedures of the study in most research centers were approved by a central Institutional Review Board (IRB) at the University of California, San Diego, and by local IRB in a few research centers\\u003csup\\u003e\\u003cspan citationid=\\\"CR78\\\" class=\\\"CitationRef\\\"\\u003e78\\u003c/span\\u003e\\u003c/sup\\u003e. All parents or caregivers provided the written informed consent and children provided the written assent. We accessed to the data under application ID 17607. From the 11,099 participants with qualified imputed genotype data, we included 6,605 EUR participants whose genetic ancestry was largely (\\u0026gt;\\u0026thinsp;80%, a previously recommended threshold\\u003csup\\u003e\\u003cspan citationid=\\\"CR79\\\" class=\\\"CitationRef\\\"\\u003e79\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR80\\\" class=\\\"CitationRef\\\"\\u003e80\\u003c/span\\u003e\\u003c/sup\\u003e) European estimated by SNPweights v2.1\\u003csup\\u003e81\\u003c/sup\\u003e based on the SNP weights for European, West African, East Asian, and Native American populations. We applied MAF\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5%, imputation quality r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.3, and \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003eHWE\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026ge; 1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e to filter the variants and transformed them from GRCh38/hg38 to GRCh37/hg19 to be consistent with the genetic data from CHIMGEN and UKBB. In the ABCD-GWASs, we included 9,064,819 autosomal and 278,985 X chromosomal bi-allelic variants. Among the 6,605 participants, 6,060 had qualified whole brainstem (not including SCP) volume data obtained by automatic subcortical segmentation using FreeSurfer v7.0. Of the 6,060 participants, 5,815 had raw brain structural MRI data, from which we calculated volumes of whole brainstem, medulla, pons, midbrain, and SCP using the Bayesian segmentation algorithm\\u003csup\\u003e\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e\\u003c/sup\\u003e in FreeSurfer v7.0. After removing the participants with volumes greater than five times MAD from median, we included 5,804-6,060 participants (\\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e) in ABCD-GWASs. The detailed quality control procedures for ABCD data are presented in \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;14\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003e\\u003cb\\u003eUKBB participants and data preparation.\\u003c/b\\u003e Most of the included EUR participants were recruited from the UKBB study (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://www.ukbiobank.ac.uk/\\u003c/span\\u003e\\u003cspan address=\\\"https://www.ukbiobank.ac.uk/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e82\\u003c/sup\\u003e, which collected approximately 500,000 participants aged 40\\u0026ndash;69 years at recruitment from 22 research centers across the United Kingdom. The UKBB study was approved by the National Health Service (NHS) Research Ethics Service (21/NW/0157), and written informed consent was obtained from each participant. We accessed to the data under application number 75556. After initial genetic data quality control and imputation, the remaining 487,207 participants were included in the further sample-level quality control. After excluding 651 participants with sex chromosome aneuploidy, 186 with sex mismatch, and 78,257 non-Caucasians, we then included 408,113 EUR participants with qualified genomic data. Using the filtering criteria of MAF\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.5%, info\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.6, and \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003eHWE\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026ge; 1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e, we finally included 10,042,001 autosomal and 390,192 X-chromosomal bi-allelic variants in UKBB-GWASs. Among the 408,113 participants, 36,533 had the volumetric data of brainstem and substructures obtained by the two segmentation approaches using FreeSurfer 7.0. We then visually checked raw brain structural images and the brainstem segmentation images, and further excluded 914 participants with brain tumors, imaging artifacts, incomplete brainstem coverage, or incorrected brainstem segmentation. After further removing participants with volumes greater than five times MAD from median, we finally included 35,521\\u0026thinsp;\\u0026minus;\\u0026thinsp;35,611 participants (\\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e) in UKBB-GWASs. The quality control procedures for UKBB data are shown in \\u003cb\\u003eSupplementary Fig.\\u0026nbsp;15\\u003c/b\\u003e.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec20\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eReproducibility of brainstem segmentation\\u003c/h2\\u003e \\u003cp\\u003eIn 7,096 CHIMGEN, 5,815 ABCD, and 35,611 UKBB participants, we calculated the intraclass correlation coefficient (ICC) of whole brainstem volumes obtained from the two brainstem segmentation methods, respectively. Although the automatic subcortical segmentation generated whole brainstem volume not including SCP and the Bayesian segmentation generated whole brainstem volume including SCP, we found high ICCs (0.956\\u0026ndash;0.997; \\u003cb\\u003eSupplementary Table\\u0026nbsp;1\\u003c/b\\u003e) in whole brainstem volumes obtained by the two brainstem segmentation methods. Among these participants, 24 CHIMGEN, 4,077 ABCD, and 2,698 UKBB participants had brain structural MRI data acquired at two time points. For each participant, we calculated the brainstem and substructure volumes using the Bayesian brainstem segmentation based on the MRI data acquired at the two time points. For each dataset, we calculated the ICC of each volumetric trait obtained from the two time points to assess the test-retest reliability of the Bayesian brainstem segmentation, and found high ICCs (0.868\\u0026ndash;0.992; \\u003cb\\u003eSupplementary Table\\u0026nbsp;1\\u003c/b\\u003e).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec21\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eHarmonization and normalization of brainstem and substructure volumes\\u003c/h2\\u003e \\u003cp\\u003eThe brain structural MRI data of the CHIMGEN, ABCD, and UKBB participants were acquired by different MRI scanners, which may bring bias to the integrated analyses of brainstem and substructure volumetric data from multiple centers. To remove the bias, for each brainstem volumetric trait from each dataset, the Combat method was used to harmonize the volume data calculated based on MRI data acquired by different scanners, which can remove between-scanner variation and preserve biological variability\\u003csup\\u003e\\u003cspan citationid=\\\"CR83\\\" class=\\\"CitationRef\\\"\\u003e83\\u003c/span\\u003e\\u003c/sup\\u003e. We tested the effect of ComBat harmonization in two participants who traveled to different centers and were scanned at 28 MRI scanners. In each participant, we segmented and calculated the five brainstem and substructure volumes based on the MRI data acquired from each scanner and used the coefficient of variation (CV) to assess between-scanner variations of these traits. In the two participants, we found that CVs of these volumetric traits before harmonization significantly reduced (Wilcoxon rank-sum test: \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;=\\u0026thinsp;0.018) after harmonization (\\u003cb\\u003eSupplementary Fig.\\u0026nbsp;16\\u003c/b\\u003e). As the skewed data distribution would violate the assumption of normal distribution when using linear regression model to perform GWASs, quantile normalization was applied to the harmonized brainstem and substructure volumetric data.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec22\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eCovariates for GWASs\\u003c/h2\\u003e \\u003cp\\u003eWe controlled for age at imaging, genetic-determined sex, age \\u0026times; sex, total intracranial volume (TIV), and first genetic principal components (PCs) in CHIMGEN-GWASs, ABCD-GWASs, and UKBB-GWASs, and further controlled for genotyping batches in ABCD-GWASs and UKBB-GWASs. For each participant, TIV was also estimated by FreeSurfer 7.0 and followed ComBat harmonization and quantile normalization. We controlled for the first ten genetic PCs in CHIMGEN-GWASs, 32 in ABCD-GWASs, and 40 in UKBB-GWASs, which were selected based on the population complexity.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec23\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eGWASs for brainstem and substructure volumes\\u003c/h2\\u003e \\u003cp\\u003e \\u003cb\\u003eGWASs for single dataset.\\u003c/b\\u003e We used the mixed linear model (MLM) from fastGWA\\u003csup\\u003e\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e\\u003c/sup\\u003e to conduct GWASs (additive effect) with the predefined covariates for absolute volumes of brainstem, medulla, pons, midbrain, and SCP at both autosomal and X-chromosomal variants in CHIMGEN, ABCD, and UKBB participants, respectively. Considering that brainstem substructure volumes were highly correlated with whole brainstem volume (r\\u0026thinsp;=\\u0026thinsp;0.556\\u0026ndash;0.982; \\u003cb\\u003eSupplementary Table\\u0026nbsp;35\\u003c/b\\u003e), we also performed GWASs for relative volumes of medulla, pons, midbrain, and SCP by additionally accounting for the whole brainstem volume, which can reveal genetic signals beyond commonality in volume\\u003csup\\u003e\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR84\\\" class=\\\"CitationRef\\\"\\u003e84\\u003c/span\\u003e\\u003c/sup\\u003e. The participants and genetic variants on both autosomes and X-chromosome included in each GWAS are presented in \\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e. As the fastGWA cannot output summary statistics for all input variants, only variants with output summary statistics (\\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e) were included in the subsequent GWAS meta-analyses. Due to lacking independent replication, we reported the study-wide significant associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, Bonferroni corrected for the nine traits) for all GWASs.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eEUR-GWAS meta-analyses.\\u003c/b\\u003e Based on the summary statistics of ABCD-GWASs and UKBB-GWASs for eight brainstem substructure volumes, we used the inverse variance weighted (IVW) fixed effect model in METAL\\u003csup\\u003e\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e\\u003c/sup\\u003e to conduct EUR-GWAS meta-analyses for the eight brainstem volumetric traits. For each trait, in addition to the identification of new genetic associations by increasing the sample size, the meta-analysis could also output summary statistics for variants included in either of the two GWASs, providing the largest number of EUR-GWAS association statistics. In EUR-GWAS meta-analysis for whole brainstem volume, to make full use of the available data resources, we also included the GWAS summary data for whole brainstem volume (n\\u0026thinsp;=\\u0026thinsp;28,809) from ENIGMA, CHARGE, and UKBB (first release) datasets\\u003csup\\u003e\\u003cspan citationid=\\\"CR14\\\" class=\\\"CitationRef\\\"\\u003e14\\u003c/span\\u003e\\u003c/sup\\u003e. As the GWAS included first released UKBB data, we then excluded these participants and reperformed GWAS for whole brainstem volume in 30,752 UKBB participants. Based on the GWAS summary data of UKBB-GWAS (n\\u0026thinsp;=\\u0026thinsp;30,752), ABCD-GWAS (n\\u0026thinsp;=\\u0026thinsp;6,060), and previous EUR-GWAS (n\\u0026thinsp;=\\u0026thinsp;28,809)\\u003csup\\u003e14\\u003c/sup\\u003e, we conducted EUR-GWAS meta-analysis for whole brainstem volume. The participants and genetic variants included in EUR-GWAS meta-analyses for brainstem volumetric traits are presented in \\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eCross-ancestry GWAS meta-analyses.\\u003c/b\\u003e Based on the obtained summary statistics from CHIMGEN-GWASs and EUR-GWAS meta-analyses, we conducted the cross-ancestry GWAS meta-analyses for these nine brainstem volumetric traits using the IVW fixed effect model in METAL\\u003csup\\u003e\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e\\u003c/sup\\u003e. The participants and genetic variants included in the cross-ancestry GWAS meta-analyses are also presented in \\u003cb\\u003eSupplementary Table\\u0026nbsp;3\\u003c/b\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003ePopulation stratification estimation.\\u003c/b\\u003e For each GWAS, we used the genomic control inflation factor (λ\\u003csub\\u003eGC\\u003c/sub\\u003e) and linkage disequilibrium score regression (LDSC) intercepts\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u003c/sup\\u003e to estimate population stratification. λ\\u003csub\\u003eGC\\u003c/sub\\u003e was calculated as the median of the resulting chi-squared (χ\\u003csup\\u003e\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e\\u003c/sup\\u003e) test statistics (z scores) divided by 0.4549, the expected median of the χ\\u003csup\\u003e\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e\\u003c/sup\\u003e distribution with one degree of freedom. As high λ\\u003csub\\u003eGC\\u003c/sub\\u003e indicates either genomic inflation or polygenicity, we used LDSC intercept to identify genomic inflation based on LD scores.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec24\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eDefining LD references and independent associations and loci\\u003c/h2\\u003e \\u003cp\\u003eWe used imputed genotype data from 7,096 CHIMGEN and 35,611 UKBB participants with qualified genetic and brainstem volumetric data to construct EAS-LD and EUR-LD references, respectively. We also constructed a cross-ancestry LD reference using the sample-weighted method based on the two datasets. For each brainstem volumetric trait, based on the summary statistics of CHIMGEN, ABCD, UKBB, EUR, or cross-ancestry GWAS, we used the matched LD reference to identify independent variant-trait associations by PLINK clumping\\u003csup\\u003e\\u003cspan citationid=\\\"CR85\\\" class=\\\"CitationRef\\\"\\u003e85\\u003c/span\\u003e\\u003c/sup\\u003e with the following steps: (1) all significant variants were included in a list of candidate variants; (2) the most significant variant was defined as the first lead variant (independent variant), and variants within 500 kb from and in LD with (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1) the lead variant were clumped; (3) the remaining variants formed a new list of candidate variants, and then step (2) was repeated; and (4) the iterative process stopped until the list was empty. We identified independent locus-trait associations by: (1) creating loci for independent variants by adding 500 kb to both sides of each variant; (2) merging loci within 500kb; (3) merging loci if any independent variant of one locus was in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1) with any independent variant of another locus; and (4) merging loci overlapped with the major histocompatibility complex (MHC) or 8p23.1 region into one locus. The associations of the remaining loci with this trait were defined as independent locus-trait associations.\\u003c/p\\u003e \\u003cp\\u003eAs EUR samples were at least six times larger than EAS samples in this study, for each trait, we used the EUR-LD reference to pool GWAS results from the five different categories of GWASs. We used the above-mentioned strategies to identify independent variant-trait and locus-trait associations for each trait. We identified LD-independent loci by merging the loci of the pooled locus-trait associations within 500kb or their lead variants with LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1).\\u003c/p\\u003e \\u003cp\\u003eAs the prior GWASs for brainstem and substructure volumes\\u003csup\\u003e\\u003cspan additionalcitationids=\\\"CR13 CR14 CR15\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e\\u003c/sup\\u003e were conducted mainly in EUR participants, we used the same strategies and EUR-LD reference to pool these GWASs to identify known independent variant-trait and locus-trait associations and LD-independent loci using two significance thresholds (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e and \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e), respectively. Using EUR-LD reference and each known list of associations and loci, we defined a new variant-trait association if the variant was 500 kb away from and not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any variants of the same trait in the list of known variant-trait associations; a new locus-trait association when the locus was 500 kb away from loci of all known locus-trait associations and all lead variants in the locus were not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any lead variants in the loci of all known locus-trait associations; and a novel locus when the locus was 500 kb away from all known loci and all lead variants in the locus were not in LD (r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.1) with any lead variants in known loci.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec25\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eSNP-based heritability and cross-ancestry genetic correlation\\u003c/h2\\u003e \\u003cp\\u003eWe used LDSC\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u003c/sup\\u003e to estimate the SNP-based heritability for brainstem volumetric traits in EAS and EUR based on the ancestry-specific LD reference and the GWAS summary statistics for autosomal variants. We applied the covariate-adjusted LDSC\\u003csup\\u003e\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e\\u003c/sup\\u003e method to estimate the cross-ancestry SNP-based heritability for these traits based on the cross-ancestry LD reference and GWAS summary statistics for autosomal variants, while adjusting for 40 genetic PCs derived from both CHIMGEN and UKBB genetic data. After excluding variants in the MHC region, based on EAS-LD and EUR-LD references and cross-ancestry GWAS summary statistics for autosomal variants, we used Popcorn (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://github.com/brielin/Popcorn\\u003c/span\\u003e\\u003cspan address=\\\"https://github.com/brielin/Popcorn\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e30\\u003c/sup\\u003e to estimate the cross-ancestry genetic-effect (not considering MAF information) and genetic-impact (considering MAF information) correlations between EAS and EUR for each brainstem volumetric trait.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec26\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eAllele effect heterogeneity between EAS and EUR\\u003c/h2\\u003e \\u003cp\\u003eWe used Cochran\\u0026rsquo;s \\u003cem\\u003eQ\\u003c/em\\u003e test (CQ-test) to assess the allelic effect heterogeneity between EAS and EUR by comparing the differences in effect sizes of candidate variants derived from EAS-GWASs and EUR-GWASs. Cochran\\u0026rsquo;s \\u003cem\\u003eQ\\u003c/em\\u003e was estimated based on the χ\\u003csup\\u003e\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e\\u003c/sup\\u003e test with a null hypothesis that homogeneity existed between the two ancestral populations and an alternative hypothesis that heterogeneity existed between ancestries. We defined the associations with CQ-test \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.05 as ancestry-shared associations and those with \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05 (Bonferroni correction for the total number of associations tested) as ancestry-specific associations. The candidate variants were those included in the pooled variant-trait associations (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) from EAS, EUR, and cross-ancestry GWASs. Using UKBB-GWASs to represent EUR-GWASs, we conducted CQ-test for the variants of the pooled variant-trait associations included in GWAS summary statistics from both CHIMGEN and UKBB. The ancestry-shared and ancestry-specific associations were validated by replacing UKBB-GWASs by ABCD-GWASs. A successful verification for ancestry-specific associations was defined as \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05 (Bonferroni correction for the total number of the discovered ancestry-specific associations).\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec27\\\" class=\\\"Section3\\\"\\u003e \\u003ch2\\u003eStatistical fine-mapping\\u003c/h2\\u003e \\u003cp\\u003eFor each locus of the pooled locus-trait association, we used the ancestry-matched LD references and GWAS summary data to perform statistical fine-mapping by estimating the posterior probability (PP) of each variant to be a causal variant (PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.8) using the probabilistic annotation integrator (PAINTOR) tool\\u003csup\\u003e\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e\\u003c/sup\\u003e with the Markov chain Monte Carlo (MCMC) model that allows multiple causal variants. We conducted fine-mapping using the ancestry-specific LD reference for pooled locus-trait associations significant only in EAS-GWASs or EUR-GWASs, and the cross-ancestry LD reference for pooled locus-trait associations significant only in cross-ancestry GWASs. When a locus-trait association was significant in two or more categories of GWASs, we performed fine-mapping using matched LD references, respectively. For pooled locus-trait associations significant in both EUR and cross-ancestry GWASs, we also conducted PAINTOR with one causal variant assumption and used the Wilcoxon rank-sum test to assess whether cross-ancestry fine mapping can reduce the 95% credible sets for these loci compared to EUR-specific fine mapping.\\u003c/p\\u003e \\u003c/div\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec28\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eFunctional annotations\\u003c/h2\\u003e \\u003cp\\u003eWe used FUMA\\u003csup\\u003e\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e\\u003c/sup\\u003e to perform functional annotations for unique variants with PP\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1 in statistical fine-mapping for the pooled locus-trait associations identified by EAS-GWASs, EUR-GWASs, and cross-ancestry GWASs, respectively. We categorized these variants based on genic position, such as exon, intron, untranslated region (UTR), and intergenic region. We used the CADD score to prioritize deleterious and pathogenic variants, and a variant was considered as pathogenic when the CADD score was above 12.37\\u003csup\\u003e36\\u003c/sup\\u003e. We also used the RegulomeDB score to prioritize the variants in non-coding regions with classification scheme based on known and predicted regulatory elements\\u003csup\\u003e\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec29\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eIdentifying genes associated with brainstem volumetric traits\\u003c/h2\\u003e \\u003cp\\u003eBased on the EUR-GWAS summary data for brainstem volumetric traits, we identified genes associated with these traits using gene-based, transcriptome-wide, and chromatin interaction association analyses, respectively.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGene-based association analyses.\\u003c/b\\u003e We mapped the variants included in the EUR-GWAS summary statistics for brainstem volumetric traits to 17,550 protein-coding genes based on location. We then conducted gene-based association analyses to identify the genes associated with brainstem volumetric traits (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/17,550/9\\u0026thinsp;=\\u0026thinsp;3.16 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e, Bonferroni correction for 17,550 genes and nine traits) based on the EUR-GWAS summary data and the EUR-LD reference from 1000 Genomes Project using Multivariate Analysis of Genomic Annotation (MAGMA)\\u003csup\\u003e\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e\\u003cb\\u003eTranscriptome-wide association analyses.\\u003c/b\\u003e Based on the EUR-GWAS summary data of nine brainstem volumetric traits and eQTL data of 13 brain tissues provided by GTEx (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://predictdb.org/\\u003c/span\\u003e\\u003cspan address=\\\"https://predictdb.org/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e39,86,87\\u003c/sup\\u003e, we used S-PrediXcan\\u003csup\\u003e\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e\\u003c/sup\\u003e to perform the transcriptome-wide association study (TWAS) between the predicted gene expression in each tissue and each brainstem volumetric trait. Based on the identified gene-trait associations of 13 brain tissues, we then used S-MultiXcan\\u003csup\\u003e\\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e\\u003c/sup\\u003e to conduct multi-tissue TWAS to test the joint effects of gene expression on these brainstem volumetric traits across the 13 brain tissues (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/15,375/9\\u0026thinsp;=\\u0026thinsp;3.61 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;7\\u003c/sup\\u003e, Bonferroni correction for 15,375 genes under consideration and nine traits).\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eChromatin interaction association analyses.\\u003c/b\\u003e We used the Hi-C-coupled MAGMA (H-MAGMA)\\u003csup\\u003e\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR88\\\" class=\\\"CitationRef\\\"\\u003e88\\u003c/span\\u003e\\u003c/sup\\u003e to identify genes associated with brainstem volumetric traits based on the chromatin interaction profiles in six brain tissues and cells, including adult brain\\u003csup\\u003e\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e\\u003c/sup\\u003e, fetal brain\\u003csup\\u003e\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e\\u003c/sup\\u003e, cortical neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e\\u003c/sup\\u003e, induced pluripotent stem cells (iPSC) derived astrocyte\\u003csup\\u003e\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e\\u003c/sup\\u003e, iPSC derived neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e\\u003c/sup\\u003e, and midbrain dopamine (DA) neuron\\u003csup\\u003e\\u003cspan citationid=\\\"CR47\\\" class=\\\"CitationRef\\\"\\u003e47\\u003c/span\\u003e\\u003c/sup\\u003e. By mapping variants of EUR-GWASs for nine brainstem volumetric traits to protein-coding genes included in the variant-gene annotation files of the six brain tissues and cells based on the EUR-LD reference derived from the 1000 Genomes Project, we obtained 17,938 genes for adult brain, 17,955 for fetal brain, 17,952 for cortical neuron, 17,926 for astrocyte and 17,931 for neuron derived from iPSC, and 17,873 for midbrain DA neuron. We conducted H-MAGMA to identify the genes associated with these brainstem volumetric traits at \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/107,575/9\\u0026thinsp;=\\u0026thinsp;5.16 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e (Bonferroni correction for the nine brainstem volumetric traits and the total number of protein-coding genes for six brain tissues and cells).\\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003ePrioritized genes and pathways\\u003c/h3\\u003e\\n\\u003cp\\u003eWe pooled significant genes in MAGMA, S-MultiXcan, and H-MAGMA analyses, and considered them as prioritized genes associated with brainstem volumetric traits, which were then submitted to g:Profiler\\u003csup\\u003e\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e\\u003c/sup\\u003e (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://biit.cs.ut.ee/gprofiler/gost\\u003c/span\\u003e\\u003cspan address=\\\"https://biit.cs.ut.ee/gprofiler/gost\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e), a web server for functional enrichment analysis, to perform pathway enrichment analyses based on pre-specified pathways from GO\\u003csup\\u003e\\u003cspan citationid=\\\"CR48\\\" class=\\\"CitationRef\\\"\\u003e48\\u003c/span\\u003e\\u003c/sup\\u003e (15,472 biological processes) and Reactome\\u003csup\\u003e\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e\\u003c/sup\\u003e (2,562 terms) databases. We corrected for multiple testing using the method (\\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003ec\\u003c/em\\u003e\\u003c/sub\\u003e \\u0026lt; 0.05, g:SCS corrected) recommended by the g:Profiler tool. The significant GO biological processes and Reactome terms were visualized as a graph using Cytoscape\\u003csup\\u003e\\u003cspan citationid=\\\"CR89\\\" class=\\\"CitationRef\\\"\\u003e89\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR90\\\" class=\\\"CitationRef\\\"\\u003e90\\u003c/span\\u003e\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec31\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eShared and distinct genetic architectures between brainstem substructures\\u003c/h2\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct genetic loci between brainstem substructures.\\u003c/b\\u003e From the pooled LD-independent loci associated with brainstem volumetric traits, we searched for loci that were associated with volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume based on the independent locus-trait associations. By this way, we identified medulla-, pons-, and midbrain-specific loci, as well as loci shared by any two substructures or all three substructures.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct genes between brainstem substructures.\\u003c/b\\u003e From the prioritized genes for brainstem volumetric traits, we searched for genes prioritized for volume(s) of one, two, or three substructures (medulla, pons, and midbrain) regardless of absolute or relative volume. Thus, we identified medulla-, pons-, and midbrain-specific genes, as well as genes shared by any two substructures or all three substructures.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eShared and distinct enrichment pathways between brainstem substructures.\\u003c/b\\u003e Based on the prioritized genes for medulla, pons, and midbrain regardless of absolute or relative volume, we used the above-mentioned method to perform pathway enrichment analyses for genes associated with medulla, pons, and midbrain volumes, respectively. Then, we searched for pathways enriched by volume(s) of one, two, or three substructures to identify the medulla-, pons-, and midbrain-specific pathways, and pathways shared by any two substructures or all three substructures.\\u003c/p\\u003e \\u003c/div\\u003e \\u003cdiv id=\\\"Sec32\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003eGenetic architectures shared by brainstem volumetric traits and other phenotypes\\u003c/h2\\u003e \\u003cp\\u003eBased on the available GWAS summary statistics for non-imaging phenotypes that are potentially associated with brainstem functions, we used three complementary methods (genetic correlation, genetic colocalization, and condFDR/conjFDR) to identify shared genetic architectures between brainstem volumetric traits and non-imaging phenotypes. As GWASs for both traits and phenotypes are mainly conducted in the EUR population and most approaches require GWAS samples from the same ancestry, we only included 26 non-imaging phenotypes (\\u003cb\\u003eSupplementary Table\\u0026nbsp;29\\u003c/b\\u003e) with EUR-GWAS summary statistics and conducted the three genetic sharing analyses based on the EUR-GWAS summary data and the EUR-LD reference.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGenetic correlation analyses\\u003c/b\\u003e. We used LDSC\\u003csup\\u003e\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e,\\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e\\u003c/sup\\u003e to calculate the genetic correlations between nine brainstem volumetric traits and 26 non-imaging phenotypes based on the EUR-GWAS summary statistics and the EUR-LD reference. We corrected for the nine brainstem volumetric traits and 26 non-imaging phenotypes, generating a Bonferroni-corrected threshold of \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05/26/9\\u0026thinsp;=\\u0026thinsp;2.14 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eGenetic colocalization analyses.\\u003c/b\\u003e We used a Bayesian colocalization method named coloc (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://chr1swallace.github.io/coloc/\\u003c/span\\u003e\\u003cspan address=\\\"https://chr1swallace.github.io/coloc/\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e57\\u003c/sup\\u003e to identify the loci shared by each pair of nine brainstem volumetric traits (locus-trait associations with \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e) and 26 non-imaging phenotypes (locus-phenotype associations with \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;8\\u003c/sup\\u003e) based on the EUR-GWAS summary statistics. With the default priors (\\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003e1\\u003c/em\\u003e\\u003c/sub\\u003e\\u0026thinsp;=\\u0026thinsp;1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e, \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003e2\\u003c/em\\u003e\\u003c/sub\\u003e\\u0026thinsp;=\\u0026thinsp;1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;4\\u003c/sup\\u003e, and \\u003cem\\u003eP\\u003c/em\\u003e\\u003csub\\u003e\\u003cem\\u003e12\\u003c/em\\u003e\\u003c/sub\\u003e\\u0026thinsp;=\\u0026thinsp;1 \\u0026times; 10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;5\\u003c/sup\\u003e), we considered evidence for colocalization if PP.H4 (the posterior probability of shared causal variant) was greater than 0.8.\\u003c/p\\u003e \\u003cp\\u003e \\u003cb\\u003eCondFDR/conjFDR analyses.\\u003c/b\\u003e Based on the EUR-GWAS summary data and EUR-LD reference from 1000 Genomes Project, we conducted the condFDR/conjFDR analyses using pleioFDR (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://github.com/precimed/pleiofdr\\u003c/span\\u003e\\u003cspan address=\\\"https://github.com/precimed/pleiofdr\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e)\\u003csup\\u003e59\\u003c/sup\\u003e to identify the variants shared by each pair of nine brainstem volumetric traits and 26 non-imaging phenotypes. We generated conditional Q-Q plots to assess polygenetic enrichment for each brainstem volumetric trait conditioned on \\u003cem\\u003eP\\u003c/em\\u003e-values (\\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.10, \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01, and \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001) of the association with each non-imaging phenotype. We defined polygenetic enrichment as the curves with more significant \\u003cem\\u003eP\\u003c/em\\u003e-values for the phenotype showed successive leftward shifts from the null distribution. Then, we performed the conjFDR analysis to identify shared genetic variants (conjFDR\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.05) between each trait-phenotype pair, in which the conjFDR value of each variant was defined as the maximal FDR value of this variant in the two mutual conditional FDR analyses. In the analyses, we excluded variants in the MHC and 8p23.1 regions. We pooled the shared variants for each trait-phenotype pair with the same standard (\\u0026lt;\\u0026thinsp;500kb and LD r\\u003csup\\u003e2\\u003c/sup\\u003e\\u0026thinsp;\\u0026gt;\\u0026thinsp;0.1) based on the EUR-LD reference. We also annotated the shared independent variants for each trait-phenotype pair using the same strategies.\\u003c/p\\u003e \"},{\"header\":\"Declarations\",\"content\":\"\\u003cdiv id=\\\"Sec33\\\" class=\\\"Section3\\\"\\u003e\\n \\u003ch2\\u003eData availability\\u003c/h2\\u003e\\n \\u003cp\\u003eThe GWAS summary statistics used in this work from following publicly available dataset: the ENIGMA study (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://enigma.ini.usc.edu/research/download-enigma-gwas-results/\\u003c/span\\u003e\\u003c/span\\u003e). All GWAS summary statistics from EAS, EUR and cross-ancestry meta-analyses of the brainstem and substructures volumes are publicly available at Zendo (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://zenodo.org/records/13382122?preview=1\\u0026amp;token=eyJhbGciOiJIUzUxMiJ9.eyJpZCI6IjhkNTgwNjM0LTFhNzUtNDFiZC1iNjk5LTM5M\\u003c/span\\u003e\\u003c/span\\u003e\\u003cbr\\u003e\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003eWE3YWMwMmQyYyIsImRhdGEiOnt9LCJyYW5kb20iOiI4NzRkYjM1MzY4OTc2NjRiN2ZhZWI1Njk\\u003c/span\\u003e\\u003c/span\\u003e\\u003cbr\\u003e\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003e5MWNiNmIxMiJ9.TiWysH_4b6FO4fReeYT2pzQ-SOyjDZ1wPnsl07KAd5K0pFhVm3Z-Mh7LA_CMkd1fX888VR2_20IyierI9zoSP-g\\u003c/span\\u003e\\u003c/span\\u003e).\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003cdiv id=\\\"Sec34\\\" class=\\\"Section3\\\"\\u003e\\n \\u003ch2\\u003eCode availability\\u003c/h2\\u003e\\n \\u003cp\\u003eWe made use of publicly available software and tools. All codes used to generate results reported in this paper are publicly available (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://github.com/xuehui2014/The-genetic-architecture-of-brainstem-structures\\u003c/span\\u003e\\u003c/span\\u003e).\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003ch2\\u003eCompeting interests\\u003c/h2\\u003e\\n\\u003cp\\u003eThe authors declare no competing interests.\\u003c/p\\u003e\\n\\u003ch2\\u003eAuthor contributions\\u003c/h2\\u003e\\n\\u003cp\\u003eC.Y., H.X. and J.F. designed the study. C.Y. and H.X. wrote the article. H.X. analyzed the data. C.Y., Z.G., S.Q. and W.Z. supervised this work. C.Y., Z.G., J.C., M.W., L.Z., G.C., Y.Y., W.L., H.Z., B.G., X.X., T.H., Z.Y., Q.Z., W.Q., F.L., M.L., S.W., Q.X., J.X., C.W., N.L., Y.J., H.X., P.Z., W.L., W.W., D.S., S.L., Z.Y., F.C., J.Z., W.S., Y.M., D.W., J.-H.G., Y.Y., K.X., J.X., B.Z., X.Z., X.-N.Z., M.J.L., Z.Y., S.Q., W.Z. acquired the data. All authors critically reviewed the manuscript.\\u003c/p\\u003e\\n\\u003ch2\\u003eAcknowledgements\\u003c/h2\\u003e\\n\\u003cp\\u003eWe are grateful to all participants and researchers from CHIMGEN, UKBB, and ABCD, who generously donated their time to make these resources available. We acknowledge funding from the National Natural Science Foundation of China (82430063, 82030053, 81425013 to C.Y.). We are grateful to the ENIGMA for providing the GWAS summary statistics of whole brainstem volume.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\u003cli\\u003e\\u003cspan\\u003eDel Negro CA, Funk GD, Feldman JL (2018) Breathing matters. Nat Rev Neurosci 19:351\\u0026ndash;367\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGuyenet PG (2006) The sympathetic control of blood pressure. Nat Rev Neurosci 7:335\\u0026ndash;346\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHerring N, Kalla M, Paterson DJ (2019) The autonomic nervous system and cardiac arrhythmias: current concepts and emerging therapies. Nat Reviews Cardiol 16:707\\u0026ndash;726\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePoe GR et al (2020) Locus coeruleus: a new look at the blue spot. Nat Rev Neurosci 21:644\\u0026ndash;659\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSmidt MP, Burbach JPH (2007) How to make a mesodiencephalic dopaminergic neuron. Nat Rev Neurosci 8:21\\u0026ndash;32\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCastellanos FX, Tannock R (2002) Neuroscience of attention-deficit/hyperactivity disorder: The search for endophenotypes. Nat Rev Neurosci 3:617\\u0026ndash;628\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSteriade M (1996) Arousal: revisiting the reticular activating system. Science 272:225\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eIglesias JE et al (2015) Bayesian segmentation of brainstem structures in MRI. NeuroImage 113:184\\u0026ndash;195\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMilella G et al (2022) Medulla oblongata volume as a promising predictor of survival in amyotrophic lateral sclerosis. Neuroimage-Clinical 34\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChong CD, Plasencia JD, Frakes DH, Schwedt TJ (2017) Structural alterations of the brainstem in migraine. Neuroimage-Clinical 13:223\\u0026ndash;227\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBosco P et al (2019) Brainstem enlargement in preschool children with autism: Results from an intermethod agreement study of segmentation algorithms. Hum Brain Mapp 40:7\\u0026ndash;19\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eElvs\\u0026aring;shagen T et al (2020) The genetic architecture of human brainstem structures and their involvement in common brain disorders. Nat Commun 11:4016\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSmith SM et al (2021) An expanded set of genome-wide association studies of brain imaging phenotypes in UK Biobank. Nat Neurosci 24:737\\u0026ndash;745\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSatizabal CL et al (2019) Genetic architecture of subcortical brain structures in 38,851 individuals. Nat Genet 51:1624\\u0026ndash;1636\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZhao B et al (2019) Genome-wide association analysis of 19,629 individuals identifies variants influencing regional brain volumes and refines their genetic co-architecture with cognitive and mental health traits. Nat Genet 51:1637\\u0026ndash;1644\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFu JL et al (2024) Cross-ancestry genome-wide association studies of brain imaging phenotypes. Nat Genet 56:1110\\u0026ndash;1120\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eXu Q et al (2020) CHIMGEN: a Chinese imaging genetics cohort to enhance cross-ethnic and cross-geographic brain research. Mol Psychiatry 25:517\\u0026ndash;529\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMiller KL et al (2016) Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat Neurosci 19:1523\\u0026ndash;1536\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eCasey BJ et al (2018) The Adolescent Brain Cognitive Development (ABCD) study: Imaging acquisition across 21 sites. Dev Cogn Neurosci 32:43\\u0026ndash;54\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePsaty BM et al (2009) Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium Design of Prospective Meta-Analyses of Genome-Wide Association Studies From 5 Cohorts. Circulation-Cardiovascular Genet 2:73\\u0026ndash;80\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eThompson PM et al (2014) The ENIGMA Consortium: large-scale collaborative analyses of neuroimaging and genetic data. Brain Imaging Behav 8:153\\u0026ndash;182\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJiang LD et al (2019) A resource-efficient tool for mixed model association analysis of large-scale data. Nat Genet 51:1749\\u0026ndash;1755\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWiller CJ, Li Y, Abecasis GR (2010) METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics 26:2190\\u0026ndash;2191\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBulik-Sullivan BK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47:291\\u0026ndash;295\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eStutzbach LD et al (2013) The unfolded protein response is activated in disease-affected brain regions in progressive supranuclear palsy and Alzheimer's disease. Acta Neuropathol Commun 1:31\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFreilinger T et al (2012) Genome-wide association analysis identifies susceptibility loci for migraine without aura. Nat Genet 44:777\\u0026ndash;782\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNyholt DR, Borsook D, Griffiths LR (2017) Migrainomics - identifying brain and genetic markers of migraine. Nat Reviews Neurol 13:725\\u0026ndash;741\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMandai K et al (2009) LIG Family Receptor Tyrosine Kinase-Associated Proteins Modulate Growth Factor Signals during Neural Development. Neuron 63:614\\u0026ndash;627\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLuo Y et al (2021) Estimating heritability and its enrichment in tissue-specific gene sets in admixed populations. Hum Mol Genet 30:1521\\u0026ndash;1534\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBrown BC, Ye CJ, Price AL, Zaitlen N, Network AGE (2016) Transethnic Genetic-Correlation Estimates from Summary Statistics. Am J Hum Genet 99:76\\u0026ndash;88\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMonsoro-Burq AH (2015) PAX transcription factors in neural crest development. Semin Cell Dev Biol 44:87\\u0026ndash;96\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKichaev G et al (2014) Integrating Functional Data to Prioritize Causal Variants in Statistical Fine-Mapping Studies. PLoS Genet 10\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKichaev G et al (2017) Improved methods for multi-trait fine mapping of pleiotropic risk loci. Bioinformatics 33:248\\u0026ndash;255\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWatanabe K, Taskesen E, van Bochoven A, Posthuma D (2017) Functional mapping and annotation of genetic associations with FUMA. Nat Commun 8:1826\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLin TT, Islam O, Heese K (2006) ABC transporters, neural stem cells and neurogenesis - a different perspective. Cell Res 16:857\\u0026ndash;871\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRentzsch P, Witten D, Cooper GM, Shendure J, Kircher M (2019) CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res 47:D886\\u0026ndash;D894\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDong SC et al (2023) Annotating and prioritizing human non-coding variants with RegulomeDB v.2. Nat Genet 55:724\\u0026ndash;726\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ede Leeuw CA, Mooij JM, Heskes T, Posthuma D (2015) MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput Biol 11:e1004219\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAguet F et al (2020) The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369:1318\\u0026ndash;1330\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBarbeira AN et al (2018) Exploring the phenotypic consequences of tissue specific gene expression variation inferred from GWAS summary statistics. Nat Commun 9:1825\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBarbeira AN et al (2019) Integrating predicted transcriptome from multiple tissues improves association detection. PLoS Genet 15:e1007889\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSey NYA et al (2020) A computational tool (H-MAGMA) for improved prediction of brain-disorder risk genes by incorporating brain chromatin interaction profiles. Nat Neurosci 23:583\\u0026ndash;589\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang D et al (2018) Comprehensive functional genomic resource and integrative model for the human brain. Science 362:1266\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWon HJ et al (2016) Chromosome conformation elucidates regulatory relationships in developing human brain. Nature 538:523\\u0026ndash;527\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHu B et al (2021) Neuronal and glial 3D chromatin architecture informs the cellular etiology of brain disorders. Nat Commun 12:3968\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eRajarajan P et al (2018) Neuron-specific signatures in the chromosomal connectome associated with schizophrenia risk. Science 362:1269\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSey NYA et al (2022) Chromatin architecture in addiction circuitry identifies risk genes and potential biological mechanisms underlying cigarette smoking and alcohol use traits. Mol Psychiatry 27:3085\\u0026ndash;3094\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAshburner M et al (2000) Gene Ontology: tool for the unification of biology. Nat Genet 25:25\\u0026ndash;29\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGillespie M et al (2022) The reactome pathway knowledgebase 2022. Nucleic Acids Res 50:D687\\u0026ndash;D692\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eKolberg L et al (2023) g:Profiler-interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Res 51:W207\\u0026ndash;W212\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eJin S et al (2023) Downregulation of UBE4B promotes CNS axon regrowth and functional recovery after stroke. iScience 26:105885\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSalih DAM et al (2012) FoxO6 regulates memory consolidation and synaptic function. Genes Dev 26:2780\\u0026ndash;2801\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNordstrom U, Jessell TM, Edlund T (2002) Progressive induction of caudal neural character by graded Wnt signaling. Nat Neurosci 5:525\\u0026ndash;532\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHevia CF, Engel-Pizcueta C, Udina F, Pujades C (2022) The neurogenic fate of the hindbrain boundaries relies on Notch3-dependent asymmetric cell divisions. Cell Rep 39:110915\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGeisen MJ et al (2008) Hox paralog group 2 genes control the migration of mouse pontine neurons through slit-robo signaling. PLoS Biol 6:e142\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBulik-Sullivan B et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47:1236\\u0026ndash;1241\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWallace C (2020) Eliciting priors and relaxing the single causal variant assumption in colocalisation analyses. PLoS Genet 16:e1008720\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLi S et al (2022) Regulatory variants at 2q33.1 confer schizophrenia risk by modulating distal gene TYW5 expression. Brain 145:770\\u0026ndash;786\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAndreassen OA et al (2013) Improved Detection of Common Variants Associated with Schizophrenia and Bipolar Disorder Using Pleiotropy-Informed Conditional False Discovery Rate. PLoS Genet 9:e1003455\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFassl A, Geng Y, Sicinski P (2022) CDK4 and CDK6 kinases: From basic science to cancer therapy. Science 375:6577\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLiu DX, Greene LA (2001) Regulation of neuronal survival and death by E2F-dependent gene repression and derepression. Neuron 32:425\\u0026ndash;438\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHarbour JW, Dean DC (2000) Rb function in cell-cycle regulation and apoptosis. Nat Cell Biol 2:E65\\u0026ndash;E67\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eZhao S et al (2024) Role of RB1 in neurodegenerative diseases: inhibition of post-mitotic neuronal apoptosis via Kmt5b. Cell Death Discovery 10:182\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGaufo GO, Thomas KR, Capecchi MR (2003) Hox3 genes coordinate mechanisms of genetic suppression and activation in the generation of branchial and somatic motoneurons. Development 130:5191\\u0026ndash;5201\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eWang GS et al (2004) Transcriptional modification by a CASK-interacting nucleosome assembly protein. Neuron 42:113\\u0026ndash;128\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eNajm J et al (2008) Mutations of CASK cause an X-linked brain malformation phenotype with microcephaly and hypoplasia of the brainstem and cerebellum. Nat Genet 40:1065\\u0026ndash;1067\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLiu N et al (2023) Cross-ancestry genome-wide association meta-analyses of hippocampal and subfield volumes. Nat Genet 55:1126\\u0026ndash;1137\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eLiu YJ et al (2023) Functional dissection of PRC1 subunits RYBP and YAF2 during neural differentiation of embryonic stem cells. Nat Commun 14:7164\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBaek ST et al (2015) An AKT3-FOXG1-reelin network underlies defective migration in human focal malformations of cortical development. Nat Med 21:1445\\u0026ndash;1454\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eDudek H et al (1997) Regulation of neuronal survival by the serine-threonine protein kinase Akt. Science 275:661\\u0026ndash;665\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFlores AI et al (2008) Constitutively active Akt induces enhanced myelination in the CNS. J Neurosci 28:7174\\u0026ndash;7183\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHuang HN et al (2004) In situ Akt phosphorylation in the nucleus tractus solitarii is involved in central control of blood pressure and heart rate. Circulation 110:2476\\u0026ndash;2483\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFelix MS et al (2014) Activation of Akt/FKHR in the medulla oblongata contributes to spontaneous respiratory recovery after incomplete spinal cord injury in adult rats. Neurobiol Dis 69:93\\u0026ndash;107\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAllen E et al (2005) Gigaxonin-controlled degradation of MAP1B light chain is critical to neuronal survival. Nature 438:224\\u0026ndash;228\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMahammad S et al (2013) Giant axonal neuropathy-associated gigaxonin mutations impair intermediate filament protein degradation. J Clin Invest 123:1964\\u0026ndash;1975\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eHastings MH, Maywood ES, Brancaccio M (2018) Generation of circadian rhythms in the suprachiasmatic nucleus. Nat Rev Neurosci 19:453\\u0026ndash;469\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChang CC et al (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4:7\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAuchter AM et al (2018) A description of the ABCD organizational structure and communication framework. Dev Cogn Neurosci 32:8\\u0026ndash;15\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eMukamel RE et al (2023) Repeat polymorphisms underlie top genetic risk loci for glaucoma and colorectal cancer. Cell 186:3659\\u0026ndash;3673\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eAhern J, Thompson W, Fan CC, Loughnan R (2023) Comparing Pruning and Thresholding with Continuous Shrinkage Polygenic Score Methods in a Large Sample of Ancestrally Diverse Adolescents from the ABCD Study. Behav Genet 53:292\\u0026ndash;309\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eChen CY et al (2013) Improved ancestry inference using weights from external reference panels. Bioinformatics 29:1399\\u0026ndash;1406\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBycroft C et al (2018) The UK Biobank resource with deep phenotyping and genomic data. Nature 562:203\\u0026ndash;209\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eFortin JP et al (2018) Harmonization of cortical thickness measurements across scanners and sites. NeuroImage 167:104\\u0026ndash;120\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003evan der Meer D et al (2020) Brain scans from 21,297 individuals reveal the genetic architecture of hippocampal subfield volumes. Mol Psychiatry 25:3053\\u0026ndash;3065\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003ePurcell S et al (2007) A tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 81 PLINK:559\\u0026ndash;575\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eGamazon ER et al (2015) A gene-based association method for mapping traits using reference transcriptome data. Nat Genet 47:1091\\u0026ndash;1098\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eBarbeira AN et al (2021) Exploiting the GTEx resources to decipher the mechanisms at GWAS loci. Genome Biol 22:49\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eSey NYA, Pratt BM, Won H (2023) Annotating genetic variants to target genes using H-MAGMA. Nat Protoc 18:22\\u0026ndash;35\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eShannon P et al (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498\\u0026ndash;2504\\u003c/span\\u003e\\u003c/li\\u003e \\u003cli\\u003e\\u003cspan\\u003eReimand J et al (2019) Pathway enrichment analysis and visualization of omics data using g:Profiler, GSEA, Cytoscape and EnrichmentMap. Nat Protoc 14:482\\u0026ndash;517\\u003c/span\\u003e\\u003c/li\\u003e\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-portfolio\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Nature Portfolio\",\"twitterHandle\":\"\",\"acdcEnabled\":false,\"dfaEnabled\":false,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false},\"keywords\":\"\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-5060768/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-5060768/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003eThe brainstem contains numerous nuclei and tracts with vital functions. Genome-wide associations with brainstem substructure volumes are explored in European individuals, however other ancestries are under-represented. Here, we conducted the cross-ancestry genome-wide association meta-analyses in 72,717 individuals for brainstem and 48,522 for eight substructure volumes, including 7,096 Chinese Han individuals. We identified 122 genetic loci associated with brainstem and substructure volumes at \\u003cem\\u003eP\\u003c/em\\u003e\\u0026thinsp;\\u0026lt;\\u0026thinsp;5.56 \\u0026times;10\\u003csup\\u003e\\u0026minus;\\u0026thinsp;9\\u003c/sup\\u003e, including 46 new loci. Three associations had different effect sizes and 292 associations had similar effect sizes between ancestries. We prioritized 550 genes for these brainstem volumetric traits, primarily enriching for neural development. We identified the shared and distinct genetic loci, genes, and pathways for midbrain, pons, and medulla volumes, and the shared genetic architectures with brainstem-related neuropsychiatric disorders and physiological functions. The results provide new insight into genetic architectures of brainstem and substructure volumes and their genetic associations with brainstem physiologies and pathologies.\\u003c/p\\u003e\",\"manuscriptTitle\":\"The genetic architecture of brainstem structures\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2024-10-09 06:33:51\",\"doi\":\"10.21203/rs.3.rs-5060768/v1\",\"editorialEvents\":[],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-communications\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"NCOMMS\",\"sideBox\":\"Learn more about [Nature Communications](http://www.nature.com/ncomms/)\",\"snPcode\":\"\",\"submissionUrl\":\"https://mts-ncomms.nature.com/\",\"title\":\"Nature Communications\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"Nature Communications\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false}}],\"origin\":\"\",\"ownerIdentity\":\"11ff3e29-8076-408f-972a-ba16b8a6459d\",\"owner\":[],\"postedDate\":\"October 9th, 2024\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[{\"id\":37627665,\"name\":\"Biological sciences/Genetics/Genetic association study/Genome-wide association studies\"},{\"id\":37627666,\"name\":\"Biological sciences/Neuroscience/Neurogenesis\"}],\"tags\":[],\"updatedAt\":\"2026-01-15T08:05:48+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-5060768\",\"link\":\"https://doi.org/10.1038/s41467-025-67221-6\",\"journal\":{\"identity\":\"nature-communications\",\"isVorOnly\":false,\"title\":\"Nature Communications\"},\"publishedOn\":\"2025-12-10 05:00:00\",\"publishedOnDateReadable\":\"December 10th, 2025\"},\"versionCreatedAt\":\"2024-10-09 06:33:51\",\"video\":\"\",\"vorDoi\":\"10.1038/s41467-025-67221-6\",\"vorDoiUrl\":\"https://doi.org/10.1038/s41467-025-67221-6\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-5060768\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-5060768\",\"identity\":\"rs-5060768\",\"version\":[\"v1\"]},\"buildId\":\"qtupq5eGEP_6zYnWcrvyt\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}