Vestibular migraine as a vestibulo-trigeminal interface phenotype: a triangulation study across genetics, peripheral multiomics and human cell atlases

Vestibular migraine as a vestibulo-trigeminal interface phenotype: a triangulation study across genetics, peripheral multiomics and human cell atlases | 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 Research Article Vestibular migraine as a vestibulo-trigeminal interface phenotype: a triangulation study across genetics, peripheral multiomics and human cell atlases Hanchen Rui, Guimei Fan, Guangcong Li, Lan Zhu, Chaoping Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9347560/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Vestibular migraine (VM) is clinically established, but the biological problem is narrower and harder: whether VM has its own causal architecture or instead reflects migraine liability that is preferentially expressed through vestibular systems. We tested the latter possibility by asking whether migraine-vertigo overlap converges on vestibular and trigeminal programs. Methods We used four evidence layers: population-scale migraine-related and vertigo-related GWAS summary statistics, with FinnGen as the primary backbone; an independent external vertigo GWAS meta-analysis for supportive anchoring; disease-labeled peripheral blood transcriptomic and multiomic datasets spanning VM, migraine, Meniere disease, and healthy controls; and human trigeminal ganglion and vestibular or inner-ear single-cell atlases for biological localization. We quantified genome-wide and local migraine-vertigo overlap, prioritized shared-liability loci and genes, and then asked whether downstream layers supported the same signal. Results Migraine-related and vertigo-related phenotypes showed strong genome-wide genetic correlation, including overall migraine versus vertigo (rg = 0.5277, SE = 0.0525, p = 9.18 x 10^-24), migraine with aura versus vertigo (rg = 0.5698, SE = 0.0734, p = 8.61 x 10^-15), and migraine without aura versus vertigo (rg = 0.4710, SE = 0.0615, p = 1.90 x 10^-14). Local analyses identified 8 shared blocks, and shared-liability prioritization yielded 204 candidate loci. Of these, 133 were matched in an independent external vertigo GWAS, 19 showed nominal support, and 71.3% were directionally concordant. Cross-layer integration converged on six prioritized genes, including five higher-confidence candidates. Representative locus-level reinforcement highlighted ARMC9 and TECTA, with ARMC9 showing the more stable cross-layer profile through nominal external support and vestibulo-trigeminal localization. Cell-atlas summaries supported a vestibulo-trigeminal landing pattern, whereas peripheral blood datasets were only partially informative and did not provide exclusive support. Conclusions The data do not justify claiming a VM-specific causal architecture. However, they do support a narrower interpretation: VM is more plausibly read as a vestibulo-trigeminal interface phenotype arising from shared migraine liability than as a wholly separate disease entity. That framework is useful precisely because clinically adjudicated population-scale VM data remain limited. vestibular migraine migraine genetics vertigo shared liability trigeminal ganglion vestibular system single-cell atlas multiomics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Vestibular migraine (VM) is one of the most common causes of recurrent episodic vertigo in neurological and vestibular practice [ 1 – 2 ]. Updated diagnostic criteria have made the syndrome more reproducible across studies and clinics [ 1 , 3 ]. Recent clinical work has described meaningful heterogeneity within that framework [ 4 ]. The central question is therefore no longer whether VM can be recognized clinically, but what kind of biological entity it represents: a distinct disorder, or a vestibularly weighted expression of migraine susceptibility [ 5 – 6 ]. That distinction matters because it determines what should count as informative evidence. If VM mainly reflects migraine biology expressed through vestibular systems, then shared liability and biologic localization should carry more weight than any single peripheral signature [ 7 – 8 ]. If VM is biologically separable from migraine, one would expect clearer divergence. The current evidence base is uneven: migraine is now increasingly tractable as a disorder of sensory-network susceptibility [ 9 – 10 ], with population-scale genetics and downstream interpretation advancing rapidly [ 11 – 14 ], whereas rigorously adjudicated vestibular phenotypes remain relatively scarce and are often absorbed into broader vertigo or comorbidity categories [ 5 , 13 ]. That imbalance makes a direct disease-specific inference difficult, but not impossible. Rather than asking for a population GWAS of strictly adjudicated VM that does not yet exist, one can ask whether the component shared by migraine-related and vertigo-related phenotypes carries a biologic signature relevant to VM. That is a narrower claim, but it is also the claim the currently available data can actually test. Open-data resources now make that narrower question tractable. Population-scale migraine and vertigo GWAS allow formal cross-trait analysis [ 15 – 17 ], and established methods can quantify overlap both genome-wide and regionally [ 18 – 19 ]. Disease-labeled peripheral datasets address a different issue: whether programs derived from shared liability show a directionally compatible pattern in clinically annotated VM, migraine, and Meniere disease samples [ 20 – 22 ]. Trigeminal ganglion and vestibular or inner-ear cell atlases answer yet another question:where such signals land biologically, and are therefore more informative for localization than blood alone [ 23 – 24 ]. We therefore did not use this study to mimic a de novo GWAS of clinically adjudicated VM [ 25 – 26 ]. Instead, we asked whether the overlap between migraine-related and vertigo-related population phenotypes defines a signal that remains coherent across downstream layers and localizes to biologically plausible vestibular and trigeminal programs. That is a deliberately constrained question. Under this framework, support for a VM-relevant model depends on convergence across evidence layers: population-scale overlap, regional concordance, locus prioritization, atlas localization, and only limited but directionally compatible support in peripheral disease-labeled data, rather than on any single dataset. Methods Study design and inferential framework The study combined four evidence layers that answer different parts of the same question: population-scale genetics, external vertigo data, disease-labeled peripheral blood datasets, and atlas-based localization. The point was not redundancy. Each layer is informative for a different reason and vulnerable to a different bias. Following triangulation principles, we treated agreement across these layers as more persuasive than a signal seen in only one of them [25-26]. The study was therefore designed for biological interpretation of VM, not for direct genome-wide discovery of clinically adjudicated VM. Data sources and prespecified analytic roles FinnGen provided the main discovery layer for migraine-related and vertigo-related phenotypes [15]. Independent vertigo GWAS datasets were used only as an external anchoring layer [16-17]. Disease-labeled support came from publicly available PBMC multiomic datasets spanning VM, migraine, Meniere disease, and healthy controls, supplemented by bulk PBMC transcriptomic context from Meniere disease [20,22]. Biological localization relied on trigeminal ganglion and vestibular or inner-ear single-cell atlases [23-24], with GTEx used only for tissue-expression context [27]. These roles were fixed in advance as discovery, anchoring, disease-relevance support, tissue-context support, or localization (Supplementary Tables S1-S3). Phenotype definition and VM-like shared liability Because no widely available population-scale GWAS yet captures rigorously adjudicated VM, we did not treat VM as a directly measured discovery phenotype. Instead, we defined VM-like shared liability as the latent component jointly indexed by migraine-related and vertigo-related population phenotypes. That definition is intentionally narrower than disease identity, but it matches what the available data can support. GWAS harmonization and quality control Before analysis, all summary statistics were brought into a common format. This involved harmonizing column structure, aligning genome builds where needed, applying quality-control filters, excluding problematic strand-ambiguous variants, and deriving a shared high-quality SNP set for cross-trait analyses. Genome-wide and local cross-trait analyses The genetic layer included SNP-based heritability, genome-wide cross-trait genetic correlation, local overlap analysis, and shared-liability modeling [18-19]. We treated both genome-wide and regional overlap as evidence of shared liability, not as proof that migraine and vertigo define the same disease. The inferential boundaries for those claims are listed in Supplementary Table S2, and the compact discovery-layer summary is given in Supplementary Table S4. Cross-layer gene prioritization and evidence integration We prioritized loci by asking which signals survived contact with more than one layer: discovery strength, external lookup support, localization, and peripheral disease relevance. A prespecified evidence matrix then separated higher-confidence from lower-confidence candidates, and manuscript-level claims were restricted to genes supported across discovery, anchoring, and localization layers (Supplementary Tables S2 and S5). Representative locus-level reinforcement analyses To see how the shared-liability framework reads at the locus level, we examined two representative higher-confidence loci: TECTA (SHARED_L1/rs11172113) and ARMC9 (SHARED_L2/rs56304645). They were chosen as exemplars, not as the only loci of interest. For each locus, we considered the shared-liability statistics, external lookup results, approximate Wakefield ABF credible sets [28], integrated evidence matrices, and atlas-based localization summaries. When a locus was absent from the precomputed local-overlap block table, local cross-trait coherence was recalculated directly within the prespecified locus window. Peripheral disease-labeled validation Peripheral blood bulk summaries and donor-level single-cell summaries were used to ask a limited question: whether liability-derived candidate programs showed the same directional pattern in disease-labeled samples [20,22]. We did not treat blood as a localization layer, because neither blood expression nor blood-based regulatory signal can establish vestibular tissue origin. Cell-atlas localization analyses Trigeminal ganglion and vestibular or inner-ear single-cell atlases were used to place candidate programs in plausible cell compartments [23-24,29]. We treated these atlas signals as localization, not as proof that any single cell type is uniquely causal. Interpretive boundaries Three boundaries were set in advance. First, the study addresses biological interpretation of VM rather than disease-specific discovery. Second, the external vertigo dataset functions as anchoring support, not formal replication. Third, peripheral blood data are supportive and non-exclusive, whereas the main inferential weight rests on population-genetic overlap and atlas-based localization. The overall design and evidentiary hierarchy are summarized in Fig. 1 and Table 1. Figure 1. Study design and inferential framework. The analysis integrated four evidence layers: population-scale genetics, external vertigo anchoring, disease-labeled peripheral blood data, and atlas-based localization. These layers were interpreted jointly rather than treated as interchangeable evidence for vestibular migraine. Table 1. Overview of datasets and analytical layers used in the study.. Data layer Dataset / source Disease or trait domain Sample type Ancestry / population Sample size Case definition / phenotype granularity Primary role in this study Used in primary analysis Notes Population genetics FinnGen migraine / vertigo GWAS outputs Migraine + vertigo-related traits GWAS summary statistics European ancestry 3 trait pairs; 8 shared blocks; 204 loci; 3 formal rg estimates Population-scale registry endpoints Primary shared-liability discovery backbone Yes Formal LDSC rg successfully estimated for migraine vs vertigo and aura-defined migraine subtypes; observed-scale h2 values are interpreted cautiously and emphasized in supplementary context. External anchoring Independent vertigo GWAS meta-analysis Vertigo GWAS summary statistics European ancestry 204 queried loci Meta-analytic broad vertigo phenotype External replication / anchoring Yes 133 matched loci; 19 nominal replications; 71.3% direction concordance among evaluable loci Candidate prioritization candidate_genes_final.csv Final candidate genes Gene list Not applicable 6 Final downstream candidate set Downstream prioritization summary Yes Six finalized genes Expression support summary_gtex_expression.csv Candidate gene expression GTEx summary table Public reference 6 Per-gene max/mean median TPM summary Auxiliary tissue-expression support Yes No vestibular-specific inference implied Disease-labeled validation summary_bulk_module_scores.csv VM / MD / HC PBMC bulk module summary Clinically labeled samples 41 labeled samples Group-labeled summary file Bulk validation layer Yes All candidate-module-score values missing Disease-labeled validation summary_scrna_module_scores.csv VM / MI / MD / HC PBMC donor-level scRNA module summary Clinically labeled samples 23 Donor-level mean module score + cell counts Single-cell disease relevance validation Yes 5 VM, 5 MI, 8 MD, 5 HC donors Cell atlas localization summary_trigeminal_localization.csv Trigeminal compartment Module-level atlas summary Human atlas summary 38028 Neuron vs non-neuron compartments Trigeminal landing-zone localization Yes 3873 neuronal cells Cell atlas localization summary_vestibular_localization.csv Vestibular compartment Module-level atlas summary Human atlas summary 23792 Adult vs fetal compartments Vestibular landing-zone localization Yes 3348 adult cells Abbreviations: VM, vestibular migraine; MI, migraine; MD, Meniere disease; HC, healthy controls; PBMC, peripheral blood mononuclear cell; GWAS, genome-wide association study. Footnote: Each dataset was used for a distinct analytic purpose and was not interpreted as interchangeable evidence. Large language model assistance was used only during manuscript preparation for language editing, structural revision, and formatting support under author supervision. It was not used for data analysis, result generation, or scientific decision-making. The authors reviewed all outputs and take full responsibility for the manuscript. Results Migraine and vertigo share robust but incomplete genetic architecture Formal LDSC analyses [18-19] showed strong genome-wide genetic correlation between migraine-related and vertigo-related phenotypes. The signal was present for overall migraine versus vertigo (rg = 0.5277, SE = 0.0525, z = 10.0501, p = 9.18 × 10^-24), migraine with aura versus vertigo (rg = 0.5698, SE = 0.0734, z = 7.7582, p = 8.61 × 10^-15), and migraine without aura versus vertigo (rg = 0.4710, SE = 0.0615, z = 7.6574, p = 1.90 × 10^-14). The overlap is therefore substantial but not complete, and it is not confined to the aura subtype. Local analyses identified 8 shared migraine-vertigo blocks, and shared-liability prioritization advanced 204 candidate loci for downstream review (Supplementary Table S4). Shared-liability prioritization defines a compact candidate space Cross-layer prioritization contracted the signal to six genes - OTOG, OTOGL, TECTA, OTOP1, ARMC9, and ZNF91. That small final set matters: the signal did not dissolve into a long tail of weak candidates, but remained compact enough to read against external and localization evidence. Final confidence classes are listed in Supplementary Table S5. External anchoring supports generalizability without implying strict replication External lookup recovered a meaningful subset of the discovery signal. Of the 204 shared candidate loci, 133 were matched in the independent external vertigo GWAS, 19 showed nominal support, and 92 of 129 evaluable loci were directionally concordant. This is not one-to-one replication, nor is it meant to be. However, it argue against the shared signal being unique to the FinnGen discovery layer (Supplementary Table S4). Genome-wide and local shared architecture are shown in Fig. 2, and the top shared-liability loci are listed in Table 2. Figure 2. Shared genetic architecture of migraine and vertigo. (A) Forest plot of genome-wide rg estimates. (B) Local shared blocks; bubble size reflects the number of overlapping variants and color indicates sign concordance. (C) Top shared-liability loci with external lookup support. Table 2. Top shared-liability loci with external vertigo support. Locus ID Lead variant Chr:Pos Shared P Support n Sign conc. Ext. P External support SHARED_L1 rs11172113 12:57133500 6.31e-11 32 1.00 3.95e-01 Matched, concordant SHARED_L2 rs56304645 1:3168622 1.86e-10 48 1.00 3.75e-05 Nominal, concordant SHARED_L3 rs6601512 8:10728086 9.20e-10 358 1.00 2.66e-01 Matched, concordant SHARED_L4 rs146245458 1:184357670 3.36e-09 7 1.00 3.69e-01 Matched, concordant SHARED_L5 rs12642146 4:130747169 2.09e-08 4 1.00 NA Not found SHARED_L6 rs11190975 10:101376997 2.58e-08 42 1.00 2.92e-03 Nominal, concordant SHARED_L7 rs73576816 13:113022566 3.14e-08 3 1.00 9.67e-01 Matched, concordant SHARED_L8 rs9653353 2:220237061 3.25e-08 160 1.00 7.24e-02 Matched, concordant SHARED_L9 rs10929971 2:160121855 4.16e-08 2 1.00 5.75e-01 Matched, discordant SHARED_L10 rs72829857 6:16965821 4.32e-08 25 1.00 7.02e-01 Matched, concordant Footnote: Shared P values are Stouffer-combined statistics from the shared-liability table. External support refers to lookup in the independent external vertigo GWAS and is used here as anchoring rather than formal replication. Representative loci show differential cross-layer stability Both TECTA and ARMC9 remained credible once local coherence, external lookup, approximate fine-mapping, and atlas-based localization were considered together. For TECTA, shared-liability support was strong (Stouffer p = 6.31 × 10^-11; 32 supporting variants), with 15,093 overlapping variants, a positive block z-correlation of 14.102, and sign concordance of 0.554 after recomputation within the locus window. External support was directionally concordant but not nominally significant (p = 0.395). ARMC9 showed similarly strong shared-liability support (Stouffer p = 1.86 × 10^-10; 48 supporting variants), with 18,561 overlapping variants, a block z-correlation of 5.825, and sign concordance of 0.537. Unlike TECTA, ARMC9 also showed nominal external support (p = 3.75 × 10^-5) together with adult-weighted vestibular localization and stronger trigeminal neuronal localization within the atlas framework [24,29]. The credible sets remained broad at both loci, so the signal is regional rather than fine-mapped to a single causal variant [28]. Representative locus-level follow-up is shown in Fig. 3. Figure 3. Representative loci in cross-layer follow-up. (A) Shared-liability signal versus external anchoring. (B) Local coherence within each locus window. (C) Credible-set sizes on a log scale. (D) Localization across trigeminal and vestibular contexts. Cell-atlas localization supports a distributed vestibulo-trigeminal landing pattern At module level, the trigeminal atlas shifted toward the neuronal compartment (module z = 0.471) relative to the non-neuronal compartment (module z = -0.471), based on 3,873 neuronal and 34,155 non-neuronal cells [24,29]. The vestibular atlas likewise favored the adult compartment (module z = 0.236; 3,348 cells) over the fetal compartment (module z = -0.236; 20,444 cells) [23]. These summaries do not resolve fine subclusters. They place the shared signal on a distributed vestibulo-trigeminal axis rather than in a single exclusive cell state. Localization summaries are presented in Table 3, and cross-layer gene-level integration is illustrated in Fig. 4. Table 3. Localization summary of the prioritized candidate module in trigeminal and vestibular atlases. Compartment Atlas / dataset Cell type / cell state Enrichment statistic Adjusted P / FDR Leading genes Linked biological interpretation Cross-atlas consistency Trigeminal summary_trigeminal_localization.csv Neuron 0.471 Not provided OTOG, OTOGL, TECTA, OTOP1 Positive neuronal shift of candidate module High Trigeminal summary_trigeminal_localization.csv Non-neuron -0.471 Not provided ARMC9, ZNF91 Reference negative compartment shift Supportive Vestibular summary_vestibular_localization.csv Adult 0.236 Not provided OTOG, OTOGL, TECTA, OTOP1 Adult-weighted vestibular localization signal High Vestibular summary_vestibular_localization.csv Fetal -0.236 Not provided ARMC9, ZNF91 Relative negative developmental compartment shift Supportive Footnote: These enrichment summaries indicate where the candidate program tends to localize; they do not identify an exclusive disease cell type. Figure 4. Cross-layer evidence for the six prioritized genes. (A) Integrated evidence matrix with raw values overlaid. (B) Trigeminal-versus-vestibular weighting. (C) GTEx expression support plotted against the stronger localization value. Peripheral blood datasets provide limited but directionally compatible disease-relevance support The bulk PBMC summary, derived from a disease-labeled PBMC expression dataset, was uninformative in the current version because all candidate module-score values were missing [22]. In donor-level PBMC single-cell summaries, VM donors (n = 5) showed the least negative mean module score (-0.0195, SD 0.0374), followed by healthy controls (n = 5, mean -0.0336, SD 0.0384), Meniere disease (n = 8, mean -0.0501, SD 0.0205), and migraine (n = 5, mean -0.0597, SD 0.0060) [20]. The largest numerical contrast was VM versus migraine. Smaller differences separated VM from Meniere disease and healthy controls. None of the exploratory Welch tests reached conventional significance, and cell-count-weighted means preserved the same rank order. This layer therefore provides directional context, not decisive evidence, which is why blood signals are interpreted here as disease relevance rather than tissue localization [20,30]. Peripheral-layer summaries are shown in Fig. 5. Figure 5. Peripheral disease-labeled validation. (A) Unweighted donor means with SD bars. (B) Cell-count weighting preserves the same rank order. The peripheral layer was directionally consistent but limited. Integrated evidence supports a biologically interpretable VM framework Cross-layer integration left six prioritized genes. Five met higher-confidence criteria (ARMC9, OTOG, OTOGL, TECTA, and ZNF91), and one met moderate-confidence criteria (OTOP1). Several of these genes, especially OTOG, OTOGL, and OTOP1, already have links to inner-ear support or interface biology [31-33]. TECTA remains more closely anchored to cochlear extracellular-matrix biology than to VM-specific biology [34]. Read together, the set fits preferential vestibular landing of shared migraine liability better than a purely peripheral vestibular disorder [8,35]. Across layers, the discovery-to-integration chain comprised three formal migraine-vertigo genetic correlation estimates, eight local overlap blocks, 204 shared candidate loci, external lookup support for 133 loci, and a final six-gene prioritized set. The final integrated evidence matrix is presented in Table 4. Table 4. Integrated evidence matrix for the six prioritized vestibular migraine genes. Gene GTEx max TPM Trigeminal localization Vestibular localization scRNA donor means Support score Level Interpretation ZNF91 12.879 neuron (1.152) Adult (1.404) VM -0.020; HC -0.034; MI -0.060; MD -0.050 4 High Prioritized VM interface candidate ARMC9 7.927 neuron (0.436) Adult (0.852) VM -0.020; HC -0.034; MI -0.060; MD -0.050 4 High Prioritized VM interface candidate TECTA 3.401 neuron (0.072) Fetal (0.400) VM -0.020; HC -0.034; MI -0.060; MD -0.050 4 High Prioritized VM interface candidate OTOGL 1.236 neuron (0.018) Adult (4.414) VM -0.020; HC -0.034; MI -0.060; MD -0.050 4 High Prioritized VM interface candidate OTOG 0.056 non.neuron (0.005) Adult (4.982) VM -0.020; HC -0.034; MI -0.060; MD -0.050 4 High Prioritized VM interface candidate OTOP1 0.000 neuron (0.002) Fetal (0.028) VM -0.020; HC -0.034; MI -0.060; MD -0.050 2 Moderate Needs added functional support Footnote: Blood-based findings were used as disease-relevance context, not as evidence of tissue origin or a vestibular migraine-specific biomarker. Confidence levels summarize cross-layer integration and are reported in the main text as higher or moderate confidence. Discussion The main result is not that vestibular migraine (VM) has now been genetically isolated as a separate disorder. It is that VM becomes biologically more intelligible when read as migraine liability expressed through vestibular and trigeminal systems. That interpretation fits the present data better than either a loose coexistence of migraine and dizziness or a sharply separate disease category [5,8,38]. Clinically, VM is now defined more consistently [1], increasingly characterized [5-6,36], and still marked by care gaps [37]. However, it remains under-resolved biologically. That distinction matters because it changes the evidentiary hierarchy. Once VM is framed as a vestibularly weighted form of migraine biology, shared liability and localization become more informative than any single peripheral signature. This is compatible with current headache research, which places migraine within distributed sensory processing and network-level susceptibility rather than within an isolated vascular or otologic mechanism [9-10,39]. Experimental and human VM data likewise point to trigemino-vestibular interactions [40-41], an interpretation also reflected in recent pathophysiologic reviews [8,35]. The genetic results fit that model, but they should be read with discipline. The rg estimates are too strong to dismiss as incidental comorbidity, and the shared local blocks argue against pure background polygenicity [15,18-19]. At the same time, the discovery layer was built from broad vertigo phenotypes rather than rigorously adjudicated VM [5-6,8]. What this supports is VM-relevant shared architecture; what it does not support is disease-specific genetic identification of VM. These analyses also sit within a broader shift in migraine genetics from locus cataloguing toward biologic interpretation [11-12,14]. The representative loci make that boundary tangible. OTOG, OTOGL, and OTOP1 already point toward inner-ear or vestibular-facing biology [31-33], which makes them plausible candidates within a preferential vestibular-landing model. TECTA remains biologically credible, but its strongest prior anchor is cochlear extracellular-matrix biology rather than VM itself [34]. ARMC9, in contrast, remained the more stable locus once external support and atlas-based localization were considered together. The broad credible sets at both loci therefore matter: they strengthen regional convergence, but they still stop short of a single causal variant [28]. The atlas layer helps narrow where the shared signal is likely to act. Trigeminal ganglion atlases place migraine-relevant programs across both neuronal and non-neuronal compartments [24,29], while inner-ear single-cell resources distinguish adult vestibular localization from broader developmental signal [23]. Read together, our summaries fit a distributed vestibulo-trigeminal landing pattern rather than a single exclusive cell state. That reading also aligns with current VM syntheses [35] and with recent neuroimaging work suggesting altered multisensory network organization rather than a solitary focal lesion [41]. The blood-based results are weaker and should remain weaker in the argument. Peripheral transcriptomic and multiomic studies can register disease-relevant immune signals in VM, Meniere disease, and migraine [20-22,30], but they cannot localize biology to the tissue or circuit level [30]. In our data, VM showed numerically higher module scores than migraine and Meniere disease in donor-level single-cell summaries, yet donor numbers were small and the bulk PBMC layer was uninformative in the current version [20]. We therefore treat the peripheral layer as supportive context, not as a defining signal. The dominant limitation is the phenotype available at discovery scale. We inferred a VM-relevant signal from migraine-related and broad vertigo-related phenotypes because no widely available population GWAS yet captures strictly adjudicated VM. That makes the external vertigo dataset an anchoring layer rather than definitive replication [16-17]. It also means that the atlas and blood-based layers, though useful, remain thinner than the genetic discovery layer [20,23-24]. Finally, several prioritized genes point toward vestibular-facing biology, but the present design cannot resolve whether those programs operate peripherally, centrally, or at the interface between the two [8,35,40]. Colocalization, SMR, and finer causal mapping would substantially sharpen that question [11,14,28]. The conclusion should therefore stay narrow. We are not arguing that VM has already been isolated as a separate molecular entity. We are arguing that the available data fit better with a vestibulo-trigeminal interface phenotype, in which shared migraine liability is preferentially expressed through vestibularly relevant programs and only partly echoed in peripheral disease-labeled data [8,35,38]. The conceptual model is summarized in Fig. 6 Conclusion These data do not isolate vestibular migraine as a separate molecular disorder. They instead support treating VM as a biologically coherent interface phenotype, in which shared migraine liability is preferentially expressed through vestibulo-trigeminal programs. Integrating population genetics, external vertigo data, disease-labeled peripheral datasets, and cell-atlas localization sharpens that interpretation while keeping clear limits on phenotype breadth, validation depth, and causal resolution. Abbreviations ABF, approximate Bayes factor; GTEx, Genotype-Tissue Expression; GWAS, genome-wide association study; HC, healthy controls; LD, linkage disequilibrium; LDSC, LD score regression; MD, Meniere disease; MI, migraine; PBMC, peripheral blood mononuclear cell; SNP, single nucleotide polymorphism; VM, vestibular migraine. Declarations Ethics approval and consent to participate This study used only publicly available, de-identified summary statistics and publicly accessible transcriptomic, multiomic, and atlas-level datasets. No new human participants were recruited and no identifiable individual-level data were collected. Institutional ethics approval was therefore not required under local policy. Consent for publication Not applicable. Availability of data and materials All public datasets analysed in this study are identified with repositories, accession identifiers, releases, and access dates in Supplementary Table S1. These include FinnGen release 12 summary statistics and endpoint-definition resources, GWAS Catalog study GCST90085927, GEO series GSE109558, GSE269117, GSE269114, GSE197289, and GSE213796, and GTEx portal resources. Processed summary outputs generated during the current study are available from the corresponding author on reasonable request. Code availability Scripts used for summary-statistic harmonization, shared-liability prioritization, locus-level reinforcement, cross-layer integration, and figure generation are available from the corresponding author on reasonable request during peer review and after publication. Competing interests The authors declare that they have no competing interests. Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Authors’ contributions Blinded for peer review. Acknowledgements Blinded for peer review. Authors’ information Blinded for peer review. References Lempert T, Olesen J, Furman J et al (2022) Vestibular migraine: Diagnostic criteria1. J Vestib Res 32(1):1–6. 10.3233/VES-201644 Villar-Martinez MD, Abdalla A, Goadsby PJ (2026) Vestibular migraine. Clinical and diagnostic challenges, and emerging therapeutic approaches. Curr Opin Neurol 39(1):42–47. 10.1097/WCO.0000000000001447 Lempert T, Olesen J, Furman J et al (2012) Vestibular migraine: diagnostic criteria. J Vestib Res 22(4):167–172. 10.3233/VES-2012-0453 Teggi R, Colombo B, Cugnata F et al (2024) Phenotypes and clinical subgroups in vestibular migraine: a cross-sectional study with cluster analysis. Neurol Sci 45(3):1209–1216. 10.1007/s10072-023-07116-w Beh SC (2022) Vestibular Migraine. Curr Neurol Neurosci Rep 22(10):601–609. 10.1007/s11910-022-01222-6 Villar-Martinez MD, Goadsby PJ (2024) Vestibular migraine: an update. Curr Opin Neurol 37(3):252–263. 10.1097/WCO.0000000000001257 Huang TC, Arshad Q, Kheradmand A (2024) Focused Update on Migraine and Vertigo Comorbidity. Curr Pain Headache Rep 28(7):613–620. 10.1007/s11916-024-01256-0 Ceriani CEJ (2024) Vestibular Migraine Pathophysiology and Treatment: a Narrative Review. Curr Pain Headache Rep 28(2):47–54. 10.1007/s11916-023-01182-7 Raggi A, Leonardi M, Arruda M et al (2024) Hallmarks of primary headache: part 1 - migraine. J Headache Pain 25(1):189 Published 2024 Oct 31. 10.1186/s10194-024-01889-x Ferrari MD, Goadsby PJ, Burstein R et al (2022) Migraine. Nat Rev Dis Primers. ;8(1):2. Published 2022 Jan 13. 10.1038/s41572-021-00328-4 Grangeon L, Lange KS, Waliszewska-Prosół M et al (2023) Genetics of migraine: where are we now? J Headache Pain 24(1):12 Published 2023 Feb 20. 10.1186/s10194-023-01547-8 Hautakangas H, Winsvold BS, Ruotsalainen SE et al (2022) Genome-wide analysis of 102,084 migraine cases identifies 123 risk loci and subtype-specific risk alleles. Nat Genet 54(2):152–160. 10.1038/s41588-021-00990-0 Ma YM, Zhang DP, Zhang HL et al (2024) Why is vestibular migraine associated with many comorbidities? J Neurol 271(12):7422–7433. 10.1007/s00415-024-12692-8 Li SJ, Shi JJ, Mao CY et al (2023) Identifying causal genes for migraine by integrating the proteome and transcriptome. J Headache Pain 24(1):111 Published 2023 Aug 17. 10.1186/s10194-023-01649-3 Kurki MI, Karjalainen J, Palta P et al (2023) FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613(7944):508–518. 10.1038/s41586-022-05473-8 Skuladottir AT, Bjornsdottir G, Nawaz MS et al (2021) A genome-wide meta-analysis uncovers six sequence variants conferring risk of vertigo. Commun Biol 4(1):1148 Published 2021 Oct 7. 10.1038/s42003-021-02673-2 Chen SP, Hsu CL, Chen TH et al (2024) A genome-wide association study identifies novel loci of vertigo in an Asian population-based cohort. Commun Biol. ;7(1):1034. Published 2024 Aug 22. 10.1038/s42003-024-06603-w Bulik-Sullivan BK, Loh PR, Finucane HK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47(3):291–295. 10.1038/ng.3211 Bulik-Sullivan B, Finucane HK, Anttila V et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47(11):1236–1241. 10.1038/ng.3406 Cruz-Granados P, Frejo L, Perez-Carpena P et al (2024) Multiomic-based immune response profiling in migraine, vestibular migraine and Meniere's disease. Immunology 173(4):768–779. 10.1111/imm.13863 Aczél T, Körtési T, Kun J et al (2021) Identification of disease- and headache-specific mediators and pathways in migraine using blood transcriptomic and metabolomic analysis. J Headache Pain. ;22(1):117. Published 2021 Oct 6. 10.1186/s10194-021-01285-9 Flook M, Leong AC, Hamid M et al (2024) Cytokine profiling and transcriptomics in mononuclear cells define immune variants in Meniere disease. Genes Immun 25(2):124–131. 10.1038/s41435-024-00260-z van der Valk WH, van Beelen ESA, Steinhart MR et al (2023) A single-cell level comparison of human inner ear organoids with the human cochlea and vestibular organs. Cell Rep 42(6):112623. 10.1016/j.celrep.2023.112623 Yang L, Xu M, Bhuiyan SA et al (2022) Human and mouse trigeminal ganglia cell atlas implicates multiple cell types in migraine. Neuron 110(11):1806–1821e8. 10.1016/j.neuron.2022.03.003 Lawlor DA, Tilling K, Davey Smith G (2016) Triangulation in aetiological epidemiology. Int J Epidemiol 45(6):1866–1886. 10.1093/ije/dyw314 Treur JL, Lukas E, Sallis HM, Wootton RE (2024) A guide for planning triangulation studies to investigate complex causal questions in behavioural and psychiatric research. Epidemiol Psychiatr Sci 33:e61 Published 2024 Nov 7. 10.1017/S2045796024000623 GTEx Consortium (2020) The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369(6509):1318–1330. 10.1126/science.aaz1776 Wakefield J (2009) Bayes factors for genome-wide association studies: comparison with P-values. Genet Epidemiol 33(1):79–86. 10.1002/gepi.20359 Bhuiyan SA, Xu M, Yang L et al (2024) Harmonized cross-species cell atlases of trigeminal and dorsal root ganglia. Sci Adv 10(25):eadj9173. 10.1126/sciadv.adj9173 Ha WS, Chu MK (2024) Altered immunity in migraine: a comprehensive scoping review. J Headache Pain 25(1):95 Published 2024 Jun 7. 10.1186/s10194-024-01800-8 Simmler MC, Cohen-Salmon M, El-Amraoui A et al (2000) Targeted disruption of otog results in deafness and severe imbalance. Nat Genet 24(2):139–143. 10.1038/72793 Yariz KO, Duman D, Zazo Seco C et al (2012) Mutations in OTOGL, encoding the inner ear protein otogelin-like, cause moderate sensorineural hearing loss. Am J Hum Genet 91(5):872–882. 10.1016/j.ajhg.2012.09.011 Kim E, Hyrc KL, Speck J et al (2011) Missense mutations in Otopetrin 1 affect subcellular localization and inhibition of purinergic signaling in vestibular supporting cells. Mol Cell Neurosci 46(3):655–661. 10.1016/j.mcn.2011.01.005 Verhoeven K, Van Laer L, Kirschhofer K et al (1998) Mutations in the human alpha-tectorin gene cause autosomal dominant non-syndromic hearing impairment. Nat Genet 19(1):60–62. 10.1038/ng0598-60 Arshad Q, Moreno-Ajona D, Goadsby PJ, Kheradmand A (2024) What visuospatial perception has taught us about the pathophysiology of vestibular migraine. Curr Opin Neurol 37(1):32–39. 10.1097/WCO.0000000000001232 Özçelik P, Kocoglu K, Ozturk V et al (2022) Characteristic differences between vestibular migraine and migraine-only patients. J Neurol 269(1):336–341. 10.1007/s00415-021-10636-0 Mallampalli MP, Rizk HG, Kheradmand A et al (2022) Care Gaps and Recommendations in Vestibular Migraine: An Expert Panel Summit. Front Neurol 12:812678 Published 2022 Jan 3. 10.3389/fneur.2021.812678 Eggers SDZ, Staab JP (2024) Vestibular migraine and persistent postural perceptual dizziness. Handb Clin Neurol 199:389–411. 10.1016/B978-0-12-823357-3.00028-8 Eigenbrodt AK, Ashina H, Khan S et al (2021) Diagnosis and management of migraine in ten steps. Nat Rev Neurol 17(8):501–514. 10.1038/s41582-021-00509-5 Zhang Y, Zhang Y, Wang Y et al (2023) Inhibition of glutamatergic trigeminal nucleus caudalis- vestibular nucleus projection neurons attenuates vestibular dysfunction in the chronic-NTG model of migraine. J Headache Pain 24(1):77 Published 2023 Jun 30. 10.1186/s10194-023-01607-z Xiong X, Dai L, Chen W et al (2024) Dynamics and concordance alterations of regional brain function indices in vestibular migraine: a resting-state fMRI study. J Headache Pain. ;25(1):1. Published 2024 Jan 5. 10.1186/s10194-023-01705-y Additional Declarations No competing interests reported. Supplementary Files Supplementary.docx Cite Share Download PDF Status: Posted 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-9347560","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":625735254,"identity":"30d8a700-825e-44e5-b8d9-ae9afdad8b39","order_by":0,"name":"Hanchen Rui","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIiWNgGAWjYFACxjYQmcDAwHyAZC1sCURbwwbVwmNAnHrzGcltD37usMvjl+75+PHHn7rEDTcSGD98zMGtReZGYrth75nkYsk5ZzdLSLaxGRvcSGCWnLkNtxYJicQ2Cd62A0DDc7cxGDbwyAG1sDHzEtAi+ReoZf+NnGcMCX8keIjSIg22RSKHjeEAmwERtvA8bJOWbUtOnHEjzViysS3BWPLMw2b8fmFPfyb5ts0usX9G8kNwiPUdTz744SMeLZhA4UJiAynqgUC+/wCJOkbBKBgFo2C4AwBvbFNpuU4/2AAAAABJRU5ErkJggg==","orcid":"","institution":"Cangzhou Central Hospital","correspondingAuthor":true,"prefix":"","firstName":"Hanchen","middleName":"","lastName":"Rui","suffix":""},{"id":625735256,"identity":"6432b4ea-6a3d-489b-a308-1760779a65ad","order_by":1,"name":"Guimei Fan","email":"","orcid":"","institution":"Cangzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Guimei","middleName":"","lastName":"Fan","suffix":""},{"id":625735258,"identity":"0ab81966-562d-4333-aa69-9c93146117cd","order_by":2,"name":"Guangcong Li","email":"","orcid":"","institution":"Cangzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Guangcong","middleName":"","lastName":"Li","suffix":""},{"id":625735260,"identity":"1f1f92cf-3495-45c1-9fd0-d713d8a1c709","order_by":3,"name":"Lan Zhu","email":"","orcid":"","institution":"Cangzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lan","middleName":"","lastName":"Zhu","suffix":""},{"id":625735262,"identity":"828e2428-6e24-40b9-9ae8-7fcba7b24a71","order_by":4,"name":"Chaoping Yang","email":"","orcid":"","institution":"Cangzhou Central Hospital","correspondingAuthor":false,"prefix":"","firstName":"Chaoping","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2026-04-07 16:11:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9347560/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9347560/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107366294,"identity":"647ddcc8-215b-42f8-a7e6-8bfec9702cf1","added_by":"auto","created_at":"2026-04-20 20:05:09","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":127759,"visible":true,"origin":"","legend":"\u003cp\u003eStudy design and inferential framework. The analysis integrated four evidence layers: population-scale genetics, external vertigo anchoring, disease-labeled peripheral blood data, and atlas-based localization. These layers were interpreted jointly rather than treated as interchangeable evidence for vestibular migraine.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/679d81cf2b8a03359c87f248.png"},{"id":107488249,"identity":"bcb08dc6-89c1-4c5e-a3ae-65177f499d23","added_by":"auto","created_at":"2026-04-22 02:43:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":180528,"visible":true,"origin":"","legend":"\u003cp\u003eShared genetic architecture of migraine and vertigo. (A) Forest plot of genome-wide rg estimates. (B) Local shared blocks; bubble size reflects the number of overlapping variants and color indicates sign concordance. (C) Top shared-liability loci with external lookup support.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/9118100c51afe81f35487956.png"},{"id":107366296,"identity":"da0cad65-f996-404c-9652-ce2f5ecd8668","added_by":"auto","created_at":"2026-04-20 20:05:09","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":169726,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative loci in cross-layer follow-up. (A) Shared-liability signal versus external anchoring. (B) Local coherence within each locus window. (C) Credible-set sizes on a log scale. (D) Localization across trigeminal and vestibular contexts.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/fbd731f631c0bf0a2c1d9786.png"},{"id":107486921,"identity":"637dcaa0-a615-408e-a09c-25184eca0154","added_by":"auto","created_at":"2026-04-22 02:39:17","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":155429,"visible":true,"origin":"","legend":"\u003cp\u003eCross-layer evidence for the six prioritized genes. (A) Integrated evidence matrix with raw values overlaid. (B) Trigeminal-versus-vestibular weighting. (C) GTEx expression support plotted against the stronger localization value.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/66c58323096ddd2681a5f7ab.png"},{"id":107366299,"identity":"f94cd3da-1c51-4b38-9ff4-39329e8918d5","added_by":"auto","created_at":"2026-04-20 20:05:09","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":83537,"visible":true,"origin":"","legend":"\u003cp\u003ePeripheral disease-labeled validation. (A) Unweighted donor means with SD bars. (B) Cell-count weighting preserves the same rank order. The peripheral layer was directionally consistent but limited.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/08e697ecd315cdd022059580.png"},{"id":107366300,"identity":"81997284-d29f-4700-8954-7c221ca200a1","added_by":"auto","created_at":"2026-04-20 20:05:09","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":147715,"visible":true,"origin":"","legend":"\u003cp\u003eConceptual synthesis. Shared migraine liability is proposed to express preferentially through vestibular and trigeminal programs, giving rise to a clinically recognizable vestibular migraine phenotype.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/e58197fd90419ebfb2090e2d.png"},{"id":107489487,"identity":"e5d62c40-31fa-42eb-afed-8cb1efabf487","added_by":"auto","created_at":"2026-04-22 02:47:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1290838,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/bfd69407-abfe-4dce-acc3-17fe2dd02023.pdf"},{"id":107487823,"identity":"f2fcfb05-6fc4-4bea-a70c-289b6f68cd07","added_by":"auto","created_at":"2026-04-22 02:42:53","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":41536,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-9347560/v1/6a9650362c2fadfb0ed40998.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Vestibular migraine as a vestibulo-trigeminal interface phenotype: a triangulation study across genetics, peripheral multiomics and human cell atlases","fulltext":[{"header":"Background","content":"\u003cp\u003eVestibular migraine (VM) is one of the most common causes of recurrent episodic vertigo in neurological and vestibular practice [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Updated diagnostic criteria have made the syndrome more reproducible across studies and clinics [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Recent clinical work has described meaningful heterogeneity within that framework [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The central question is therefore no longer whether VM can be recognized clinically, but what kind of biological entity it represents: a distinct disorder, or a vestibularly weighted expression of migraine susceptibility [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThat distinction matters because it determines what should count as informative evidence. If VM mainly reflects migraine biology expressed through vestibular systems, then shared liability and biologic localization should carry more weight than any single peripheral signature [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. If VM is biologically separable from migraine, one would expect clearer divergence. The current evidence base is uneven: migraine is now increasingly tractable as a disorder of sensory-network susceptibility [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], with population-scale genetics and downstream interpretation advancing rapidly [\u003cspan additionalcitationids=\"CR12 CR13\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], whereas rigorously adjudicated vestibular phenotypes remain relatively scarce and are often absorbed into broader vertigo or comorbidity categories [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThat imbalance makes a direct disease-specific inference difficult, but not impossible. Rather than asking for a population GWAS of strictly adjudicated VM that does not yet exist, one can ask whether the component shared by migraine-related and vertigo-related phenotypes carries a biologic signature relevant to VM. That is a narrower claim, but it is also the claim the currently available data can actually test.\u003c/p\u003e \u003cp\u003eOpen-data resources now make that narrower question tractable. Population-scale migraine and vertigo GWAS allow formal cross-trait analysis [\u003cspan additionalcitationids=\"CR16\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], and established methods can quantify overlap both genome-wide and regionally [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Disease-labeled peripheral datasets address a different issue: whether programs derived from shared liability show a directionally compatible pattern in clinically annotated VM, migraine, and Meniere disease samples [\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Trigeminal ganglion and vestibular or inner-ear cell atlases answer yet another question:where such signals land biologically, and are therefore more informative for localization than blood alone [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe therefore did not use this study to mimic a de novo GWAS of clinically adjudicated VM [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Instead, we asked whether the overlap between migraine-related and vertigo-related population phenotypes defines a signal that remains coherent across downstream layers and localizes to biologically plausible vestibular and trigeminal programs.\u003c/p\u003e \u003cp\u003eThat is a deliberately constrained question. Under this framework, support for a VM-relevant model depends on convergence across evidence layers: population-scale overlap, regional concordance, locus prioritization, atlas localization, and only limited but directionally compatible support in peripheral disease-labeled data, rather than on any single dataset.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eStudy design and inferential framework\u003c/h2\u003e\n\u003cp\u003eThe study combined four evidence layers that answer different parts of the same question: population-scale genetics, external vertigo data, disease-labeled peripheral blood datasets, and atlas-based localization. The point was not redundancy. Each layer is informative for a different reason and vulnerable to a different bias. Following triangulation principles, we treated agreement across these layers as more persuasive than a signal seen in only one of them [25-26]. The study was therefore designed for biological interpretation of VM, not for direct genome-wide discovery of clinically adjudicated VM.\u003c/p\u003e\n\u003ch2\u003eData sources and prespecified analytic roles\u003c/h2\u003e\n\u003cp\u003eFinnGen provided the main discovery layer for migraine-related and vertigo-related phenotypes [15]. Independent vertigo GWAS datasets were used only as an external anchoring layer [16-17]. Disease-labeled support came from publicly available PBMC multiomic datasets spanning VM, migraine, Meniere disease, and healthy controls, supplemented by bulk PBMC transcriptomic context from Meniere disease [20,22]. Biological localization relied on trigeminal ganglion and vestibular or inner-ear single-cell atlases [23-24], with GTEx used only for tissue-expression context [27]. These roles were fixed in advance as discovery, anchoring, disease-relevance support, tissue-context support, or localization (Supplementary Tables S1-S3).\u003c/p\u003e\n\u003ch2\u003ePhenotype definition and VM-like shared liability\u003c/h2\u003e\n\u003cp\u003eBecause no widely available population-scale GWAS yet captures rigorously adjudicated VM, we did not treat VM as a directly measured discovery phenotype. Instead, we defined VM-like shared liability as the latent component jointly indexed by migraine-related and vertigo-related population phenotypes. That definition is intentionally narrower than disease identity, but it matches what the available data can support.\u003c/p\u003e\n\u003ch2\u003eGWAS harmonization and quality control\u003c/h2\u003e\n\u003cp\u003eBefore analysis, all summary statistics were brought into a common format. This involved harmonizing column structure, aligning genome builds where needed, applying quality-control filters, excluding problematic strand-ambiguous variants, and deriving a shared high-quality SNP set for cross-trait analyses.\u003c/p\u003e\n\u003ch2\u003eGenome-wide and local cross-trait analyses\u003c/h2\u003e\n\u003cp\u003eThe genetic layer included SNP-based heritability, genome-wide cross-trait genetic correlation, local overlap analysis, and shared-liability modeling [18-19]. We treated both genome-wide and regional overlap as evidence of shared liability, not as proof that migraine and vertigo define the same disease. The inferential boundaries for those claims are listed in Supplementary Table S2, and the compact discovery-layer summary is given in Supplementary Table S4.\u003c/p\u003e\n\u003ch2\u003eCross-layer gene prioritization and evidence integration\u003c/h2\u003e\n\u003cp\u003eWe prioritized loci by asking which signals survived contact with more than one layer: discovery strength, external lookup support, localization, and peripheral disease relevance. A prespecified evidence matrix then separated higher-confidence from lower-confidence candidates, and manuscript-level claims were restricted to genes supported across discovery, anchoring, and localization layers (Supplementary Tables S2 and S5).\u003c/p\u003e\n\u003ch2\u003eRepresentative locus-level reinforcement analyses\u003c/h2\u003e\n\u003cp\u003eTo see how the shared-liability framework reads at the locus level, we examined two representative higher-confidence loci: TECTA (SHARED_L1/rs11172113) and ARMC9 (SHARED_L2/rs56304645). They were chosen as exemplars, not as the only loci of interest. For each locus, we considered the shared-liability statistics, external lookup results, approximate Wakefield ABF credible sets [28], integrated evidence matrices, and atlas-based localization summaries. When a locus was absent from the precomputed local-overlap block table, local cross-trait coherence was recalculated directly within the prespecified locus window.\u003c/p\u003e\n\u003ch2\u003ePeripheral disease-labeled validation\u003c/h2\u003e\n\u003cp\u003ePeripheral blood bulk summaries and donor-level single-cell summaries were used to ask a limited question: whether liability-derived candidate programs showed the same directional pattern in disease-labeled samples [20,22]. We did not treat blood as a localization layer, because neither blood expression nor blood-based regulatory signal can establish vestibular tissue origin.\u003c/p\u003e\n\u003ch2\u003eCell-atlas localization analyses\u003c/h2\u003e\n\u003cp\u003eTrigeminal ganglion and vestibular or inner-ear single-cell atlases were used to place candidate programs in plausible cell compartments [23-24,29]. We treated these atlas signals as localization, not as proof that any single cell type is uniquely causal.\u003c/p\u003e\n\u003ch2\u003eInterpretive boundaries\u003c/h2\u003e\n\u003cp\u003eThree boundaries were set in advance. First, the study addresses biological interpretation of VM rather than disease-specific discovery. Second, the external vertigo dataset functions as anchoring support, not formal replication. Third, peripheral blood data are supportive and non-exclusive, whereas the main inferential weight rests on population-genetic overlap and atlas-based localization. The overall design and evidentiary hierarchy are summarized in Fig. 1 and Table 1.\u003c/p\u003e\n\u003cp\u003eFigure 1. Study design and inferential framework. The analysis integrated four evidence layers: population-scale genetics, external vertigo anchoring, disease-labeled peripheral blood data, and atlas-based localization. These layers were interpreted jointly rather than treated as interchangeable evidence for vestibular migraine.\u003c/p\u003e\n\u003cp\u003eTable 1. Overview of datasets and analytical layers used in the study..\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"1001\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eData layer\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDataset / source\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDisease or trait domain\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSample type\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAncestry / population\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSample size\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCase definition / phenotype granularity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePrimary role in this study\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUsed in primary analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 100px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNotes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePopulation genetics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eFinnGen migraine / vertigo GWAS outputs\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eMigraine + vertigo-related traits\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eGWAS summary statistics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eEuropean ancestry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e3 trait pairs;\u003cbr\u003e\u0026nbsp;8 shared blocks;\u003cbr\u003e\u0026nbsp;204 loci;\u003cbr\u003e\u0026nbsp;3 formal rg estimates\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePopulation-scale registry endpoints\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePrimary shared-liability discovery backbone\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eFormal LDSC rg successfully estimated for migraine vs vertigo and aura-defined migraine subtypes; observed-scale h2 values are interpreted cautiously and emphasized in supplementary context.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eExternal anchoring\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eIndependent vertigo GWAS meta-analysis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eVertigo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eGWAS summary statistics\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eEuropean ancestry\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e204 queried loci\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eMeta-analytic broad vertigo phenotype\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eExternal replication / anchoring\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e133 matched loci; 19 nominal replications; 71.3% direction concordance among evaluable loci\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eCandidate prioritization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ecandidate_genes_final.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eFinal candidate genes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eGene list\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eNot applicable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eFinal downstream candidate set\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eDownstream prioritization summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eSix finalized genes\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eExpression support\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003esummary_gtex_expression.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eCandidate gene expression\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eGTEx summary table\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePublic reference\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePer-gene max/mean median TPM summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eAuxiliary tissue-expression support\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eNo vestibular-specific inference implied\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eDisease-labeled validation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003esummary_bulk_module_scores.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eVM / MD / HC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePBMC bulk module summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eClinically labeled samples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e41 labeled samples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eGroup-labeled summary file\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eBulk validation layer\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eAll candidate-module-score values missing\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eDisease-labeled validation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003esummary_scrna_module_scores.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eVM / MI / MD / HC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003ePBMC donor-level scRNA module summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eClinically labeled samples\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eDonor-level mean module score + cell counts\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eSingle-cell disease relevance validation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e5 VM, 5 MI, 8 MD, 5 HC donors\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eCell atlas localization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003esummary_trigeminal_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eTrigeminal compartment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eModule-level atlas summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eHuman atlas summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e38028\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eNeuron vs non-neuron compartments\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eTrigeminal landing-zone localization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e3873 neuronal cells\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eCell atlas localization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003esummary_vestibular_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eVestibular compartment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eModule-level atlas summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eHuman atlas summary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e23792\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eAdult vs fetal compartments\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eVestibular landing-zone localization\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 100px;\"\u003e\n \u003cp\u003e3348 adult cells\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eAbbreviations:\u0026nbsp;\u003c/strong\u003eVM, vestibular migraine; MI, migraine; MD, Meniere disease; HC, healthy controls; PBMC, peripheral blood mononuclear cell; GWAS, genome-wide association study.\u003c/p\u003e\n\u003cp\u003eFootnote: Each dataset was used for a distinct analytic purpose and was not interpreted as interchangeable evidence.\u003c/p\u003e\n\u003cp\u003eLarge language model assistance was used only during manuscript preparation for language editing, structural revision, and formatting support under author supervision. It was not used for data analysis, result generation, or scientific decision-making. The authors reviewed all outputs and take full responsibility for the manuscript.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eMigraine and vertigo share robust but incomplete genetic architecture\u003c/h2\u003e\n\u003cp\u003eFormal LDSC analyses [18-19] showed strong genome-wide genetic correlation between migraine-related and vertigo-related phenotypes. The signal was present for overall migraine versus vertigo (rg = 0.5277, SE = 0.0525, z = 10.0501, p = 9.18 \u0026times; 10^-24), migraine with aura versus vertigo (rg = 0.5698, SE = 0.0734, z = 7.7582, p = 8.61 \u0026times; 10^-15), and migraine without aura versus vertigo (rg = 0.4710, SE = 0.0615, z = 7.6574, p = 1.90 \u0026times; 10^-14). The overlap is therefore substantial but not complete, and it is not confined to the aura subtype. Local analyses identified 8 shared migraine-vertigo blocks, and shared-liability prioritization advanced 204 candidate loci for downstream review (Supplementary Table S4).\u003c/p\u003e\n\u003ch2\u003eShared-liability prioritization defines a compact candidate space\u003c/h2\u003e\n\u003cp\u003eCross-layer prioritization contracted the signal to six genes - OTOG, OTOGL, TECTA, OTOP1, ARMC9, and ZNF91. That small final set matters: the signal did not dissolve into a long tail of weak candidates, but remained compact enough to read against external and localization evidence. Final confidence classes are listed in Supplementary Table S5.\u003c/p\u003e\n\u003ch2\u003eExternal anchoring supports generalizability without implying strict replication\u003c/h2\u003e\n\u003cp\u003eExternal lookup recovered a meaningful subset of the discovery signal. Of the 204 shared candidate loci, 133 were matched in the independent external vertigo GWAS, 19 showed nominal support, and 92 of 129 evaluable loci were directionally concordant. This is not one-to-one replication, nor is it meant to be. However, it argue against the shared signal being unique to the FinnGen discovery layer (Supplementary Table S4). Genome-wide and local shared architecture are shown in Fig. 2, and the top shared-liability loci are listed in Table 2.\u003c/p\u003e\n\u003cp\u003eFigure 2. Shared genetic architecture of migraine and vertigo. (A) Forest plot of genome-wide rg estimates. (B) Local shared blocks; bubble size reflects the number of overlapping variants and color indicates sign concordance. (C) Top shared-liability loci with external lookup support.\u003c/p\u003e\n\u003cp\u003eTable 2. Top shared-liability loci with external vertigo support.\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLocus ID\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLead variant\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eChr:Pos\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eShared P\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSupport n\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSign conc.\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eExt. P\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eExternal support\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers11172113\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e12:57133500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e6.31e-11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.95e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers56304645\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1:3168622\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.86e-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.75e-05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eNominal, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers6601512\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e8:10728086\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e9.20e-10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e358\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2.66e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers146245458\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1:184357670\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.36e-09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.69e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers12642146\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4:130747169\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2.09e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eNA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eNot found\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers11190975\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e10:101376997\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2.58e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2.92e-03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eNominal, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers73576816\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e13:113022566\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.14e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e9.67e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers9653353\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2:220237061\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.25e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e160\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e7.24e-02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers10929971\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2:160121855\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4.16e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e5.75e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, discordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eSHARED_L10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ers72829857\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e6:16965821\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4.32e-08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e7.02e-01\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eMatched, concordant\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eFootnote: Shared P values are Stouffer-combined statistics from the shared-liability table. External support refers to lookup in the independent external vertigo GWAS and is used here as anchoring rather than formal replication.\u003c/p\u003e\n\u003ch2\u003eRepresentative loci show differential cross-layer stability\u003c/h2\u003e\n\u003cp\u003eBoth TECTA and ARMC9 remained credible once local coherence, external lookup, approximate fine-mapping, and atlas-based localization were considered together. For TECTA, shared-liability support was strong (Stouffer p = 6.31 \u0026times; 10^-11; 32 supporting variants), with 15,093 overlapping variants, a positive block z-correlation of 14.102, and sign concordance of 0.554 after recomputation within the locus window. External support was directionally concordant but not nominally significant (p = 0.395). ARMC9 showed similarly strong shared-liability support (Stouffer p = 1.86 \u0026times; 10^-10; 48 supporting variants), with 18,561 overlapping variants, a block z-correlation of 5.825, and sign concordance of 0.537. Unlike TECTA, ARMC9 also showed nominal external support (p = 3.75 \u0026times; 10^-5) together with adult-weighted vestibular localization and stronger trigeminal neuronal localization within the atlas framework [24,29]. The credible sets remained broad at both loci, so the signal is regional rather than fine-mapped to a single causal variant [28]. Representative locus-level follow-up is shown in Fig. 3.\u003c/p\u003e\n\u003cp\u003eFigure 3. Representative loci in cross-layer follow-up. (A) Shared-liability signal versus external anchoring. (B) Local coherence within each locus window. (C) Credible-set sizes on a log scale. (D) Localization across trigeminal and vestibular contexts.\u003c/p\u003e\n\u003ch2\u003eCell-atlas localization supports a distributed vestibulo-trigeminal landing pattern\u003c/h2\u003e\n\u003cp\u003eAt module level, the trigeminal atlas shifted toward the neuronal compartment (module z = 0.471) relative to the non-neuronal compartment (module z = -0.471), based on 3,873 neuronal and 34,155 non-neuronal cells [24,29]. The vestibular atlas likewise favored the adult compartment (module z = 0.236; 3,348 cells) over the fetal compartment (module z = -0.236; 20,444 cells) [23]. These summaries do not resolve fine subclusters. They place the shared signal on a distributed vestibulo-trigeminal axis rather than in a single exclusive cell state. Localization summaries are presented in Table 3, and cross-layer gene-level integration is illustrated in Fig. 4.\u003c/p\u003e\n\u003cp\u003eTable 3. Localization summary of the prioritized candidate module in trigeminal and vestibular atlases.\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCompartment\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAtlas / dataset\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCell type / cell state\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 68px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEnrichment statistic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 60px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAdjusted P / FDR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 102px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLeading genes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 144px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLinked biological interpretation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCross-atlas consistency\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eTrigeminal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003esummary_trigeminal_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNeuron\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.471\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNot provided\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eOTOG, OTOGL, TECTA, OTOP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003ePositive neuronal shift of candidate module\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eTrigeminal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003esummary_trigeminal_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNon-neuron\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e-0.471\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNot provided\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eARMC9, ZNF91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eReference negative compartment shift\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eSupportive\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eVestibular\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003esummary_vestibular_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eAdult\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e0.236\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNot provided\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eOTOG, OTOGL, TECTA, OTOP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eAdult-weighted vestibular localization signal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eVestibular\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003esummary_vestibular_localization.csv\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eFetal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003e-0.236\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eNot provided\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eARMC9, ZNF91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eRelative negative developmental compartment shift\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 125px;\"\u003e\n \u003cp\u003eSupportive\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eFootnote: These enrichment summaries indicate where the candidate program tends to localize; they do not identify an exclusive disease cell type.\u003c/p\u003e\n\u003cp\u003eFigure 4. Cross-layer evidence for the six prioritized genes. (A) Integrated evidence matrix with raw values overlaid. (B) Trigeminal-versus-vestibular weighting. (C) GTEx expression support plotted against the stronger localization value.\u003c/p\u003e\n\u003ch2\u003ePeripheral blood datasets provide limited but directionally compatible disease-relevance support\u003c/h2\u003e\n\u003cp\u003eThe bulk PBMC summary, derived from a disease-labeled PBMC expression dataset, was uninformative in the current version because all candidate module-score values were missing [22]. In donor-level PBMC single-cell summaries, VM donors (n = 5) showed the least negative mean module score (-0.0195, SD 0.0374), followed by healthy controls (n = 5, mean -0.0336, SD 0.0384), Meniere disease (n = 8, mean -0.0501, SD 0.0205), and migraine (n = 5, mean -0.0597, SD 0.0060) [20]. The largest numerical contrast was VM versus migraine. Smaller differences separated VM from Meniere disease and healthy controls. None of the exploratory Welch tests reached conventional significance, and cell-count-weighted means preserved the same rank order. This layer therefore provides directional context, not decisive evidence, which is why blood signals are interpreted here as disease relevance rather than tissue localization [20,30]. Peripheral-layer summaries are shown in Fig. 5.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFigure 5. Peripheral disease-labeled validation. (A) Unweighted donor means with SD bars. (B) Cell-count weighting preserves the same rank order. The peripheral layer was directionally consistent but limited.\u003c/p\u003e\n\u003ch2\u003eIntegrated evidence supports a biologically interpretable VM framework\u003c/h2\u003e\n\u003cp\u003eCross-layer integration left six prioritized genes. Five met higher-confidence criteria (ARMC9, OTOG, OTOGL, TECTA, and ZNF91), and one met moderate-confidence criteria (OTOP1). Several of these genes, especially OTOG, OTOGL, and OTOP1, already have links to inner-ear support or interface biology [31-33]. TECTA remains more closely anchored to cochlear extracellular-matrix biology than to VM-specific biology [34]. Read together, the set fits preferential vestibular landing of shared migraine liability better than a purely peripheral vestibular disorder [8,35]. Across layers, the discovery-to-integration chain comprised three formal migraine-vertigo genetic correlation estimates, eight local overlap blocks, 204 shared candidate loci, external lookup support for 133 loci, and a final six-gene prioritized set. The final integrated evidence matrix is presented in Table 4.\u003c/p\u003e\n\u003cp\u003eTable 4. Integrated evidence matrix for the six prioritized vestibular migraine genes.\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGene\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGTEx max\u003cbr\u003e\u0026nbsp;TPM\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrigeminal\u003cbr\u003e\u0026nbsp;localization\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVestibular\u003cbr\u003e\u0026nbsp;localization\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003escRNA\u003cbr\u003e\u0026nbsp;donor means\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSupport\u003cbr\u003e\u0026nbsp;score\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLevel\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInterpretation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eZNF91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e12.879\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eneuron (1.152)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eAdult (1.404)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ePrioritized VM interface candidate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eARMC9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e7.927\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eneuron (0.436)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eAdult (0.852)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ePrioritized VM interface candidate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eTECTA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e3.401\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eneuron (0.072)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eFetal (0.400)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ePrioritized VM interface candidate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eOTOGL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e1.236\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eneuron (0.018)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eAdult (4.414)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ePrioritized VM interface candidate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eOTOG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e0.056\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003enon.neuron (0.005)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eAdult (4.982)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003ePrioritized VM interface candidate\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eOTOP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e0.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eneuron (0.002)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eFetal (0.028)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eVM -0.020; HC -0.034; MI -0.060; MD -0.050\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eModerate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 78px;\"\u003e\n \u003cp\u003eNeeds added functional support\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eFootnote: Blood-based findings were used as disease-relevance context, not as evidence of tissue origin or a vestibular migraine-specific biomarker. Confidence levels summarize cross-layer integration and are reported in the main text as higher or moderate confidence.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe main result is not that vestibular migraine (VM) has now been genetically isolated as a separate disorder. It is that VM becomes biologically more intelligible when read as migraine liability expressed through vestibular and trigeminal systems. That interpretation fits the present data better than either a loose coexistence of migraine and dizziness or a sharply separate disease category [5,8,38]. Clinically, VM is now defined more consistently [1], increasingly characterized [5-6,36], and still marked by care gaps [37].\u0026nbsp;However, it remains under-resolved\u0026nbsp;biologically.\u003c/p\u003e\n\u003cp\u003eThat distinction matters because it changes the evidentiary hierarchy. Once VM is framed as a vestibularly weighted form of migraine biology, shared liability and localization become more informative than any single peripheral signature. This is compatible with current headache research, which places migraine within distributed sensory processing and network-level susceptibility rather than within an isolated vascular or otologic mechanism [9-10,39]. Experimental and human VM data likewise point to trigemino-vestibular interactions [40-41], an interpretation also reflected in recent pathophysiologic reviews [8,35].\u003c/p\u003e\n\u003cp\u003eThe genetic results fit that model, but they should be read with discipline. The rg estimates are too strong to dismiss as incidental comorbidity, and the shared local blocks argue against pure background polygenicity [15,18-19]. At the same time, the discovery layer was built from broad vertigo phenotypes rather than rigorously adjudicated VM [5-6,8]. What this supports is VM-relevant shared architecture; what it does not support is disease-specific genetic identification of VM. These analyses also sit within a broader shift in migraine genetics from locus cataloguing toward biologic interpretation [11-12,14].\u003c/p\u003e\n\u003cp\u003eThe representative loci make that boundary tangible. OTOG, OTOGL, and OTOP1 already point toward inner-ear or vestibular-facing biology [31-33], which makes them plausible candidates within a preferential vestibular-landing model. TECTA remains biologically credible, but its strongest prior anchor is cochlear extracellular-matrix biology rather than VM itself [34]. ARMC9, in contrast, remained the more stable locus once external support and atlas-based localization were considered together. The broad credible sets at both loci therefore matter: they strengthen regional convergence, but they still stop short of a single causal variant [28].\u003c/p\u003e\n\u003cp\u003eThe atlas layer helps narrow where the shared signal is likely to act. Trigeminal ganglion atlases place migraine-relevant programs across both neuronal and non-neuronal compartments [24,29], while inner-ear single-cell resources distinguish adult vestibular localization from broader developmental signal [23]. Read together, our summaries fit a distributed vestibulo-trigeminal landing pattern rather than a single exclusive cell state. That reading also aligns with current VM syntheses [35] and with recent neuroimaging work suggesting altered multisensory network organization rather than a solitary focal lesion [41].\u003c/p\u003e\n\u003cp\u003eThe blood-based results are weaker and should remain weaker in the argument. Peripheral transcriptomic and multiomic studies can register disease-relevant immune signals in VM, Meniere disease, and migraine [20-22,30], but they cannot localize biology to the tissue or circuit level [30]. In our data, VM showed numerically higher module scores than migraine and Meniere disease in donor-level single-cell summaries, yet donor numbers were small and the bulk PBMC layer was uninformative in the current version [20]. We therefore treat the peripheral layer as supportive context, not as a defining signal.\u003c/p\u003e\n\u003cp\u003eThe dominant limitation is the phenotype available at discovery scale. We inferred a VM-relevant signal from migraine-related and broad vertigo-related phenotypes because no widely available population GWAS yet captures strictly adjudicated VM. That makes the external vertigo dataset an anchoring layer rather than definitive replication [16-17]. It also means that the atlas and blood-based layers, though useful, remain thinner than the genetic discovery layer [20,23-24]. Finally, several prioritized genes point toward vestibular-facing biology, but the present design cannot resolve whether those programs operate peripherally, centrally, or at the interface between the two [8,35,40]. Colocalization, SMR, and finer causal mapping would substantially sharpen that question [11,14,28].\u003c/p\u003e\n\u003cp\u003eThe conclusion should therefore stay narrow. We are not arguing that VM has already been isolated as a separate molecular entity. We are arguing that the available data fit better with a vestibulo-trigeminal interface phenotype, in which shared migraine liability is preferentially expressed through vestibularly relevant programs and only partly echoed in peripheral disease-labeled data [8,35,38]. The conceptual model is summarized in Fig. 6\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThese data do not isolate vestibular migraine as a separate molecular disorder. They instead support treating VM as a biologically coherent interface phenotype, in which shared migraine liability is preferentially expressed through vestibulo-trigeminal programs. Integrating population genetics, external vertigo data, disease-labeled peripheral datasets, and cell-atlas localization sharpens that interpretation while keeping clear limits on phenotype breadth, validation depth, and causal resolution.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eABF, approximate Bayes factor; GTEx, Genotype-Tissue Expression; GWAS, genome-wide association study; HC, healthy controls; LD, linkage disequilibrium; LDSC, LD score regression; MD, Meniere disease; MI, migraine; PBMC, peripheral blood mononuclear cell; SNP, single nucleotide polymorphism; VM, vestibular migraine.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e\n\u003cp\u003eThis study used only publicly available, de-identified summary statistics and publicly accessible transcriptomic, multiomic, and atlas-level datasets. No new human participants were recruited and no identifiable individual-level data were collected. Institutional ethics approval was therefore not required under local policy.\u003c/p\u003e\n\u003ch2\u003eConsent for publication\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e\n\u003cp\u003eAll public datasets analysed in this study are identified with repositories, accession identifiers, releases, and access dates in Supplementary Table S1. These include FinnGen release 12 summary statistics and endpoint-definition resources, GWAS Catalog study GCST90085927, GEO series GSE109558, GSE269117, GSE269114, GSE197289, and GSE213796, and GTEx portal resources. Processed summary outputs generated during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003ch2\u003eCode availability\u003c/h2\u003e\n\u003cp\u003eScripts used for summary-statistic harmonization, shared-liability prioritization, locus-level reinforcement, cross-layer integration, and figure generation are available from the corresponding author on reasonable request during peer review and after publication.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\n\u003ch2\u003eAuthors’ contributions\u003c/h2\u003e\n\u003cp\u003eBlinded for peer review.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eBlinded for peer review.\u003c/p\u003e\n\u003ch2\u003eAuthors’ information\u003c/h2\u003e\n\u003cp\u003eBlinded for peer review.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLempert T, Olesen J, Furman J et al (2022) Vestibular migraine: Diagnostic criteria1. J Vestib Res 32(1):1\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3233/VES-201644\u003c/span\u003e\u003cspan address=\"10.3233/VES-201644\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillar-Martinez MD, Abdalla A, Goadsby PJ (2026) Vestibular migraine. Clinical and diagnostic challenges, and emerging therapeutic approaches. Curr Opin Neurol 39(1):42\u0026ndash;47. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/WCO.0000000000001447\u003c/span\u003e\u003cspan address=\"10.1097/WCO.0000000000001447\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLempert T, Olesen J, Furman J et al (2012) Vestibular migraine: diagnostic criteria. J Vestib Res 22(4):167\u0026ndash;172. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3233/VES-2012-0453\u003c/span\u003e\u003cspan address=\"10.3233/VES-2012-0453\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTeggi R, Colombo B, Cugnata F et al (2024) Phenotypes and clinical subgroups in vestibular migraine: a cross-sectional study with cluster analysis. Neurol Sci 45(3):1209\u0026ndash;1216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10072-023-07116-w\u003c/span\u003e\u003cspan address=\"10.1007/s10072-023-07116-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeh SC (2022) Vestibular Migraine. Curr Neurol Neurosci Rep 22(10):601\u0026ndash;609. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11910-022-01222-6\u003c/span\u003e\u003cspan address=\"10.1007/s11910-022-01222-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVillar-Martinez MD, Goadsby PJ (2024) Vestibular migraine: an update. Curr Opin Neurol 37(3):252\u0026ndash;263. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/WCO.0000000000001257\u003c/span\u003e\u003cspan address=\"10.1097/WCO.0000000000001257\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang TC, Arshad Q, Kheradmand A (2024) Focused Update on Migraine and Vertigo Comorbidity. Curr Pain Headache Rep 28(7):613\u0026ndash;620. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11916-024-01256-0\u003c/span\u003e\u003cspan address=\"10.1007/s11916-024-01256-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCeriani CEJ (2024) Vestibular Migraine Pathophysiology and Treatment: a Narrative Review. Curr Pain Headache Rep 28(2):47\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11916-023-01182-7\u003c/span\u003e\u003cspan address=\"10.1007/s11916-023-01182-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaggi A, Leonardi M, Arruda M et al (2024) Hallmarks of primary headache: part 1 - migraine. J Headache Pain 25(1):189 Published 2024 Oct 31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-024-01889-x\u003c/span\u003e\u003cspan address=\"10.1186/s10194-024-01889-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFerrari MD, Goadsby PJ, Burstein R et al (2022) Migraine. Nat Rev Dis Primers. ;8(1):2. Published 2022 Jan 13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41572-021-00328-4\u003c/span\u003e\u003cspan address=\"10.1038/s41572-021-00328-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrangeon L, Lange KS, Waliszewska-Pros\u0026oacute;ł M et al (2023) Genetics of migraine: where are we now? J Headache Pain 24(1):12 Published 2023 Feb 20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-023-01547-8\u003c/span\u003e\u003cspan address=\"10.1186/s10194-023-01547-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHautakangas H, Winsvold BS, Ruotsalainen SE et al (2022) Genome-wide analysis of 102,084 migraine cases identifies 123 risk loci and subtype-specific risk alleles. Nat Genet 54(2):152\u0026ndash;160. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41588-021-00990-0\u003c/span\u003e\u003cspan address=\"10.1038/s41588-021-00990-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa YM, Zhang DP, Zhang HL et al (2024) Why is vestibular migraine associated with many comorbidities? J Neurol 271(12):7422\u0026ndash;7433. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00415-024-12692-8\u003c/span\u003e\u003cspan address=\"10.1007/s00415-024-12692-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi SJ, Shi JJ, Mao CY et al (2023) Identifying causal genes for migraine by integrating the proteome and transcriptome. J Headache Pain 24(1):111 Published 2023 Aug 17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-023-01649-3\u003c/span\u003e\u003cspan address=\"10.1186/s10194-023-01649-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKurki MI, Karjalainen J, Palta P et al (2023) FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613(7944):508\u0026ndash;518. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-022-05473-8\u003c/span\u003e\u003cspan address=\"10.1038/s41586-022-05473-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSkuladottir AT, Bjornsdottir G, Nawaz MS et al (2021) A genome-wide meta-analysis uncovers six sequence variants conferring risk of vertigo. Commun Biol 4(1):1148 Published 2021 Oct 7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s42003-021-02673-2\u003c/span\u003e\u003cspan address=\"10.1038/s42003-021-02673-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen SP, Hsu CL, Chen TH et al (2024) A genome-wide association study identifies novel loci of vertigo in an Asian population-based cohort. Commun Biol. ;7(1):1034. Published 2024 Aug 22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s42003-024-06603-w\u003c/span\u003e\u003cspan address=\"10.1038/s42003-024-06603-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulik-Sullivan BK, Loh PR, Finucane HK et al (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet 47(3):291\u0026ndash;295. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3211\u003c/span\u003e\u003cspan address=\"10.1038/ng.3211\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBulik-Sullivan B, Finucane HK, Anttila V et al (2015) An atlas of genetic correlations across human diseases and traits. Nat Genet 47(11):1236\u0026ndash;1241. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3406\u003c/span\u003e\u003cspan address=\"10.1038/ng.3406\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCruz-Granados P, Frejo L, Perez-Carpena P et al (2024) Multiomic-based immune response profiling in migraine, vestibular migraine and Meniere's disease. Immunology 173(4):768\u0026ndash;779. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/imm.13863\u003c/span\u003e\u003cspan address=\"10.1111/imm.13863\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAcz\u0026eacute;l T, K\u0026ouml;rt\u0026eacute;si T, Kun J et al (2021) Identification of disease- and headache-specific mediators and pathways in migraine using blood transcriptomic and metabolomic analysis. J Headache Pain. ;22(1):117. Published 2021 Oct 6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-021-01285-9\u003c/span\u003e\u003cspan address=\"10.1186/s10194-021-01285-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFlook M, Leong AC, Hamid M et al (2024) Cytokine profiling and transcriptomics in mononuclear cells define immune variants in Meniere disease. Genes Immun 25(2):124\u0026ndash;131. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41435-024-00260-z\u003c/span\u003e\u003cspan address=\"10.1038/s41435-024-00260-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Valk WH, van Beelen ESA, Steinhart MR et al (2023) A single-cell level comparison of human inner ear organoids with the human cochlea and vestibular organs. Cell Rep 42(6):112623. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.celrep.2023.112623\u003c/span\u003e\u003cspan address=\"10.1016/j.celrep.2023.112623\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang L, Xu M, Bhuiyan SA et al (2022) Human and mouse trigeminal ganglia cell atlas implicates multiple cell types in migraine. Neuron 110(11):1806\u0026ndash;1821e8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuron.2022.03.003\u003c/span\u003e\u003cspan address=\"10.1016/j.neuron.2022.03.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLawlor DA, Tilling K, Davey Smith G (2016) Triangulation in aetiological epidemiology. Int J Epidemiol 45(6):1866\u0026ndash;1886. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/ije/dyw314\u003c/span\u003e\u003cspan address=\"10.1093/ije/dyw314\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTreur JL, Lukas E, Sallis HM, Wootton RE (2024) A guide for planning triangulation studies to investigate complex causal questions in behavioural and psychiatric research. Epidemiol Psychiatr Sci 33:e61 Published 2024 Nov 7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/S2045796024000623\u003c/span\u003e\u003cspan address=\"10.1017/S2045796024000623\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGTEx Consortium (2020) The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369(6509):1318\u0026ndash;1330. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.aaz1776\u003c/span\u003e\u003cspan address=\"10.1126/science.aaz1776\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWakefield J (2009) Bayes factors for genome-wide association studies: comparison with P-values. Genet Epidemiol 33(1):79\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/gepi.20359\u003c/span\u003e\u003cspan address=\"10.1002/gepi.20359\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhuiyan SA, Xu M, Yang L et al (2024) Harmonized cross-species cell atlases of trigeminal and dorsal root ganglia. Sci Adv 10(25):eadj9173. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/sciadv.adj9173\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.adj9173\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHa WS, Chu MK (2024) Altered immunity in migraine: a comprehensive scoping review. J Headache Pain 25(1):95 Published 2024 Jun 7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-024-01800-8\u003c/span\u003e\u003cspan address=\"10.1186/s10194-024-01800-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimmler MC, Cohen-Salmon M, El-Amraoui A et al (2000) Targeted disruption of otog results in deafness and severe imbalance. Nat Genet 24(2):139\u0026ndash;143. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/72793\u003c/span\u003e\u003cspan address=\"10.1038/72793\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYariz KO, Duman D, Zazo Seco C et al (2012) Mutations in OTOGL, encoding the inner ear protein otogelin-like, cause moderate sensorineural hearing loss. Am J Hum Genet 91(5):872\u0026ndash;882. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajhg.2012.09.011\u003c/span\u003e\u003cspan address=\"10.1016/j.ajhg.2012.09.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim E, Hyrc KL, Speck J et al (2011) Missense mutations in Otopetrin 1 affect subcellular localization and inhibition of purinergic signaling in vestibular supporting cells. Mol Cell Neurosci 46(3):655\u0026ndash;661. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.mcn.2011.01.005\u003c/span\u003e\u003cspan address=\"10.1016/j.mcn.2011.01.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerhoeven K, Van Laer L, Kirschhofer K et al (1998) Mutations in the human alpha-tectorin gene cause autosomal dominant non-syndromic hearing impairment. Nat Genet 19(1):60\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng0598-60\u003c/span\u003e\u003cspan address=\"10.1038/ng0598-60\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArshad Q, Moreno-Ajona D, Goadsby PJ, Kheradmand A (2024) What visuospatial perception has taught us about the pathophysiology of vestibular migraine. Curr Opin Neurol 37(1):32\u0026ndash;39. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/WCO.0000000000001232\u003c/span\u003e\u003cspan address=\"10.1097/WCO.0000000000001232\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026Ouml;z\u0026ccedil;elik P, Kocoglu K, Ozturk V et al (2022) Characteristic differences between vestibular migraine and migraine-only patients. J Neurol 269(1):336\u0026ndash;341. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00415-021-10636-0\u003c/span\u003e\u003cspan address=\"10.1007/s00415-021-10636-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMallampalli MP, Rizk HG, Kheradmand A et al (2022) Care Gaps and Recommendations in Vestibular Migraine: An Expert Panel Summit. Front Neurol 12:812678 Published 2022 Jan 3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fneur.2021.812678\u003c/span\u003e\u003cspan address=\"10.3389/fneur.2021.812678\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEggers SDZ, Staab JP (2024) Vestibular migraine and persistent postural perceptual dizziness. Handb Clin Neurol 199:389\u0026ndash;411. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/B978-0-12-823357-3.00028-8\u003c/span\u003e\u003cspan address=\"10.1016/B978-0-12-823357-3.00028-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEigenbrodt AK, Ashina H, Khan S et al (2021) Diagnosis and management of migraine in ten steps. Nat Rev Neurol 17(8):501\u0026ndash;514. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41582-021-00509-5\u003c/span\u003e\u003cspan address=\"10.1038/s41582-021-00509-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Zhang Y, Wang Y et al (2023) Inhibition of glutamatergic trigeminal nucleus caudalis- vestibular nucleus projection neurons attenuates vestibular dysfunction in the chronic-NTG model of migraine. J Headache Pain 24(1):77 Published 2023 Jun 30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-023-01607-z\u003c/span\u003e\u003cspan address=\"10.1186/s10194-023-01607-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong X, Dai L, Chen W et al (2024) Dynamics and concordance alterations of regional brain function indices in vestibular migraine: a resting-state fMRI study. J Headache Pain. ;25(1):1. Published 2024 Jan 5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s10194-023-01705-y\u003c/span\u003e\u003cspan address=\"10.1186/s10194-023-01705-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"vestibular migraine, migraine genetics, vertigo, shared liability, trigeminal ganglion, vestibular system, single-cell atlas, multiomics","lastPublishedDoi":"10.21203/rs.3.rs-9347560/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9347560/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eVestibular migraine (VM) is clinically established, but the biological problem is narrower and harder: whether VM has its own causal architecture or instead reflects migraine liability that is preferentially expressed through vestibular systems. We tested the latter possibility by asking whether migraine-vertigo overlap converges on vestibular and trigeminal programs.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe used four evidence layers: population-scale migraine-related and vertigo-related GWAS summary statistics, with FinnGen as the primary backbone; an independent external vertigo GWAS meta-analysis for supportive anchoring; disease-labeled peripheral blood transcriptomic and multiomic datasets spanning VM, migraine, Meniere disease, and healthy controls; and human trigeminal ganglion and vestibular or inner-ear single-cell atlases for biological localization. We quantified genome-wide and local migraine-vertigo overlap, prioritized shared-liability loci and genes, and then asked whether downstream layers supported the same signal.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eMigraine-related and vertigo-related phenotypes showed strong genome-wide genetic correlation, including overall migraine versus vertigo (rg\u0026thinsp;=\u0026thinsp;0.5277, SE\u0026thinsp;=\u0026thinsp;0.0525, p\u0026thinsp;=\u0026thinsp;9.18 x 10^-24), migraine with aura versus vertigo (rg\u0026thinsp;=\u0026thinsp;0.5698, SE\u0026thinsp;=\u0026thinsp;0.0734, p\u0026thinsp;=\u0026thinsp;8.61 x 10^-15), and migraine without aura versus vertigo (rg\u0026thinsp;=\u0026thinsp;0.4710, SE\u0026thinsp;=\u0026thinsp;0.0615, p\u0026thinsp;=\u0026thinsp;1.90 x 10^-14). Local analyses identified 8 shared blocks, and shared-liability prioritization yielded 204 candidate loci. Of these, 133 were matched in an independent external vertigo GWAS, 19 showed nominal support, and 71.3% were directionally concordant. Cross-layer integration converged on six prioritized genes, including five higher-confidence candidates. Representative locus-level reinforcement highlighted ARMC9 and TECTA, with ARMC9 showing the more stable cross-layer profile through nominal external support and vestibulo-trigeminal localization. Cell-atlas summaries supported a vestibulo-trigeminal landing pattern, whereas peripheral blood datasets were only partially informative and did not provide exclusive support.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe data do not justify claiming a VM-specific causal architecture. However, they do support a narrower interpretation: VM is more plausibly read as a vestibulo-trigeminal interface phenotype arising from shared migraine liability than as a wholly separate disease entity. That framework is useful precisely because clinically adjudicated population-scale VM data remain limited.\u003c/p\u003e","manuscriptTitle":"Vestibular migraine as a vestibulo-trigeminal interface phenotype: a triangulation study across genetics, peripheral multiomics and human cell atlases","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-20 20:05:04","doi":"10.21203/rs.3.rs-9347560/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"33802ff6-4bc4-4496-bf4b-32323f8d5a9b","owner":[],"postedDate":"April 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-21T18:24:20+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-20 20:05:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9347560","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9347560","identity":"rs-9347560","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}