Skip to main content
NCBI home page
BMC Psychiatry logoLink to BMC Psychiatry
. 2026 Jan 19;26:153. doi: 10.1186/s12888-026-07816-5

Unraveling the genetic interplay and therapeutic potentials between major depressive disorder and metabolic syndrome: multi-ancestry and multi-trait genome-wide association analyses

Yiheng Feng 1,#, Yihao Li 2,#, Hong Zhang 2, Huilin Xia 2, Yuan Lu 2, Hang Zhou 2, Peng Huang 2,✉
PMCID: PMC12895643  PMID: 41555281

This investigation elucidates the genetic connection between major depressive disorder (MDD) and metabolic syndrome (MetS), suggesting bidirectional genetic correlations and shared pleiotropic genes. Leveraging a comprehensive genome-wide association study (GWAS) dataset from European and East Asian populations, we discovered new genetic markers linked to MDD and enhanced the robustness of genetic associations via cross-trait analysis. Moreover, the study harnessed computational strategies for drug repurposing, identifying Cytochrome P450 and HDAC inhibitors as candidate drugs for further investigation in MDD and MetS. Employing BLISS technology, we pinpointed proteins significantly linked to both conditions, advancing our comprehension of their molecular underpinnings. Through Mendelian randomization, we investigated how diverse dietary patterns across populations influence MDD and MetS, shedding light on the relationship between diet and disease susceptibility. This research not only enriches our understanding of the intersecting biological pathways of MDD and MetS but also opens avenues for innovative preventive and therapeutic measures.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-026-07816-5.

Keywords: Major depressive disorder (MDD), Genome-Wide association studies (GWAS), Multi-ancestry meta-analysis, Genetic correlation, Mendelian randomization

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12888-026-07816-5.

Introduction

Major Depressive Disorder (MDD) ranks among the most prevalent mental health conditions globally, impacting over 350 million individuals [1, 2]. Characterized by enduring sadness, diminished interest or pleasure, and reduced energy, MDD significantly hampers everyday functionality and is associated with an increased risk of self-harm and suicide. The disorder manifests in various forms, including seasonal affective disorder, postpartum depression, and depression linked to chronic illnesses [3, 4].

Increasing research into mental health has unveiled a bidirectional link between metabolic syndrome (MetS) and psychiatric conditions, noting a 32.6% MetS comorbidity rate among patients with severe mental disorders [5]. Specifically, MetS prevalence in MDD patients has been reported to reach 30% [6, 7]. Several biological mechanisms, such as inflammation, oxidative stress, and disruptions in neuroendocrine regulation (notably in the HPA axis functioning), have been proposed to underpin the connection between these disorders [8–10].

Genome-wide association studies (GWAS) have emerged as pivotal in identifying numerous genetic variations linking these diseases by extensively analyzing genomic data across large populations [11–14]. However, the full scope of genetic correlation remains elusive, underscoring the need to pinpoint the genetic variants or loci responsible for this correlation [15]. Such elucidation is crucial for comprehending the shared genetic underpinnings of these conditions [16], potentially involving pleiotropy and often serving as genetic confounders in trait associations [17, 18]. Employing cross-trait analysis to utilize GWAS signal correlations offers insightful revelations into the genetic interactions across different traits, thereby enriching our understanding of the genetics behind complex diseases [19, 20].

To date, genetic linkages between these conditions have predominantly been investigated within European ancestry populations. To capitalize on available data and optimize both sample size and effectiveness, we compiled a GWAS dataset combining MDD and MetS-related metrics across European and East Asian ancestries. Through multi-ancestry and multi-trait meta-analyses, followed by detailed risk loci examination and molecular association studies, this research delves into the intricate biological interplay between MDD and MetS.

Method

Meta-analysis of genome-wide association studies

This study integrates the most comprehensive MDD GWAS datasets available, encompassing individuals of both European and East Asian ancestries. For the European cohort (MA EUR), we included data from 170,756 MDD patients and 329,443 controls from the Psychiatric Genomics Consortium (PGC), excluding data from 23andMe [21], alongside 47,696 patients and 359,290 controls from FinnGen’s R10 release [22], and 27,507 patients with 36,368 controls from iPSYCH as cataloged in the GWAS Catalog [23, 24]. For East Asian ancestry (MA EAS), we drew on the latest meta-analysis data, comprising 21,980 MDD patients and 360,956 controls [25].

Moreover, we incorporated GWAS summary data for 10 MetS-related metrics from both European and East Asian ancestries as detailed in Supplementary Table 1 [26–32]. Rigorous quality control measures were applied to all GWAS summary data, including genomic control value assessments.

To amalgamate the European MDD GWAS datasets, we first ascertained minimal to no sample overlap and consistent effect sizes across studies by evaluating λmeta. A λmeta significantly greater than 1 suggests heterogeneity between cohorts, while a value significantly lower indicates sample overlap. Our analysis ensured λmeta values were closely aligned with 1 [33, 34]. Subsequently, using the METAL software, we employed the inverse variance weighted (IVW) fixed-effects method to amalgamate the 3 EUR MDD GWAS datasets [35, 36].

Through linkage disequilibrium score regression (LDSC) [37], genetic correlations between MDD and 10 metabolic indices were independently assessed across European and East Asian ancestries, leveraging linkage disequilibrium scores derived from the 1000 Genomes Project for both groups [38]. A significant genetic correlation was acknowledged when p-values, adjusted for a false discovery rate (FDR) of 0.05, were attained.

For multi-trait analyses (MT), the MTAG tool was employed to examine MDD and its correlated traits across ancestries, yielding separate results for European (MT EUR) and East Asian (MT EAS) ancestries. MTAG enhances statistical power by leveraging trait correlations, facilitating the identification of novel genetic associations [39, 40]. For multi-ancestry multi-trait (MAMT) analyses, we continued with the IVW fixed-effects method, amalgamating multi-trait results across ancestries, and assessing heterogeneity with METAL [35, 36].

These meta-analyses excluded the MHC region (chromosome 6: 25–35 MB) due to its disproportionately large effect sizes, which could skew model assumptions [34]. All GWAS meta-analyses conducted in this study calculated the λ statistic and the LDSC intercept, including for the 8 multi-ancestry dietary habits GWAS summary data subsequently utilized in Mendelian randomization analyses (Supplementary Table 3).

Identification of independent risk loci

To delineate independent genetic variants, we employed the Functional Mapping and Annotation of Genetic Associations (FUMA) platform, adopting a significance threshold of 5 × 10− 8 for our analysis [41]. FUMA was utilized to define genomic risk loci and annotate variant functions using LD data from the 1000 Genomes Project Phase 3 for populations of all ancestries. To classify MD-associated loci as novel, we compared our lead SNPs against previously reported MDD lead SNPs from two reference databases: the Psychiatric Genomics Consortium (PGC) MDD GWAS results and the NHGRI-EBI GWAS Catalog. A locus was designated as novel if its lead SNP was located > 1 megabase (Mb) away from any previously reported lead SNP; loci within this distance were considered known. We established criteria whereby the lead SNVs would possess a P-value less than 5 × 10− 8, and a maximal P-value cutoff set below 0.05. Independent SNVs were characterized by an r2 value below 0.2, and lead SNVs by an r2 value below 0.1 within a 1 Mb range. Genomic risk loci were determined by amalgamating genomic regions whenever the distance between lead SNVs was under 500 kb [16]. These SNVs underwent annotation using ANNOVAR and were evaluated for their combined annotation-dependent depletion (CADD) and RegulomeDB scores [42–44].

For further validation of the reproducibility of independent risk loci, we applied a statistical fine mapping approach to all loci exhibiting significant associations within the MAMT [45]. This technique, an augmentation of Bayesian fine mapping, leverages heterogeneity estimates across ancestries, assigning a lower prior probability to variants with disparate effect estimates among populations as potential causal variants [46]. A 99% confidence set for each variant was ascertained by ranking SNPs within an r2 > 0.1 of the lead variants according to their posterior probabilities, continuing until the aggregate posterior probabilities met or exceeded 0.99 [25].

To assess the impact of SNV-associated loci (± 250 kb range) on MDD and metabolic traits, we conducted Bayesian colocalization analysis. Here, hypothesis H3 denotes association with different causal variants between traits, whereas H4 suggests a shared causal variant for the associated traits. Utilizing the “coloc” R package, we determined the posterior probability for each hypothesis based on summary statistics [47]. In doing so, we calculated the conditional posterior probability (P(H4)/(P(H3) + P(H4))) for colocalization, marking variants with a value above 0.7 as colocalized [48]. To reduce reliance on the single-causal-variant assumption, SuSiE fine-mapping was performed for all loci with PP.H4.abf values greater than 0.7 [49].

Annotation of genes

The summary statistics for MAMT underwent initial annotation using the FUMA platform. We derived both positional and expression quantitative trait loci (eQTL) mapping outcomes from FUMA. Furthermore, eQTL annotations for genes in brain tissues were facilitated by datasets accessible within FUMA [25, 41].

We conducted gene-based association analyses employing tools such as Multi-marker Analysis of Genomic Annotation (MAGMA), Polygenic Priority Score (POPS), and mBAT-combo [50–52]. POPS leverages gene-level association statistics from MAGMA, incorporating data from an extensive array of public bulk and single-cell expression datasets, annotated biological pathways, and anticipated protein-protein interactions, including 79 gene features, to rank genes [51]. The mBAT-combo method, a multivariate technique, excels in identifying gene-trait associations amidst masking effects using GWAS summary data to enhance overall effectiveness, surpassing the conventional sum-χ2 method [52]. Significance for MAGMA and mBAT-combo was determined by a FDR below 0.05 after Benjamini-Hochberg (BH) correction, while for POPS, a score greater than 1 was set as the threshold for inclusion.

Transcriptome-wide association studies

For the transcriptome-wide association studies (TWAS), we utilized the FUSION software, sourcing SNP weights from its website, which aggregates data from numerous external research efforts. This includes SNP weights for various tissues such as all available brain tissues, adrenal and pituitary glands, thyroid, and whole blood as cataloged in GTEx v8, alongside those from the CommonMind Consortium focusing on the dorsolateral prefrontal cortex [53, 54]. The analysis was underpinned by a Multi-Ancestry LD reference panel. We established a transcriptome-wide significance threshold for TWAS associations at P < 1.37 × 10− 6, a value determined through an alignment-based procedure that calculates significance thresholds relative to the number of features examined [53].

Gene-phenotype correlation analysis

To investigate the relationships between pleiotropic genes and phenotypes, we utilized summary data-based Mendelian randomization (SMR) analyses. This involved employing MAMT alongside eQTL data, which encompassed brain tissues from BrainMeta, blood samples from the eQTLGen consortium, and data on nine specific brain cell types derived from both sources [55–58]. The SMR methodology integrates information from GWAS and QTL research to assess pleiotropic connections between gene expression and complex traits of interest [59]. A multi-ancestry LD reference panel facilitated LD estimations. In our analyses, gene-trait associations were deemed significant when the BH adjusted FDR was less than 0.05, and the HEIDI test indicated non-significance (P > 0.01).

Identification of credible genes

To ascertain a robust set of pleiotropic genes, we synthesized insights from an array of gene mapping approaches. This entailed merging data from FUMA annotations, MAGMA, POPS, mBAT-combo, TWAS, and SMR [60]. Genes that consistently emerged across all four analytical techniques were earmarked as credible candidates.

In discerning potential pleiotropic genes, we conducted a phenotypic enrichment analysis utilizing the Mammalian Phenotype (MP) ontology from Mouse Genome Informatics (MGI) [61]. This analysis aimed to delineate the phenotypic distinctiveness of pleiotropic versus non-pleiotropic genes. By comparing the frequency of genes associated with specific phenotypes within the pleiotropic gene set against that in the non-pleiotropic set, we could gauge phenotypic enrichment. The Fisher exact test facilitated the identification of significant differences in phenotype associations.

Computational drug repurposing

We utilized the TWAS Z-values of identified credible genes as proxies for their phenotypic impact. Subsequently, we employed the Connectivity Map (CMap) algorithm to pinpoint drugs capable of negating the disease’s genetic signature. This process involved matching the TWAS association signals from our target genes against the reference maps within the CMap database, which documents the gene expression alterations resulting from chemical perturbations, thus facilitating drug-gene pair characterization. Our analysis was restricted to the CMap touchstone dataset, featuring reference signatures from nine cell lines treated with approximately 3,000 well-documented small molecule drugs. We gauged a drug’s connectivity to the genome based on its consistency in inversely affecting the expression patterns of the entire gene set. CMap’s τ score quantifies the alignment between the query and reference expression profiles, with a negative τ score suggesting that the implicated molecule can normalize the trait-associated gene expression. Thus, such molecules are potential candidates for repurposing in treating the disease, where a more negative τ score strengthens the case for drug repurposing [34, 62].

Protein-based association analysis

In an effort to navigate the intricate proteomic landscape associated with the phenotype, we conducted a Biomarker Expression Level Imputation using Summary-level Statistics (BLISS) analysis. This approach stands in contrast to traditional Proteome-wide Association Studies (PWAS), which primarily depend on individual-level reference proteomes, thereby constraining their capacity to harness publicly available summary-level pQTL data [63]. We undertook an analysis of pQTLs for European ancestry by integrating MT EUR with pQTL data from deCODE, UK Biobank, and ARIC, and for East Asian ancestry by combining MT EAS with pQTL data from the UK Biobank, employing both standard PWAS and BLISS methodologies [64]. For proteins yielding significant associations across different ancestries (P < 0.05), we aggregated findings from all sources for these proteins and conducted a meta-analysis using the metafor package [65]. Heterogeneity was assessed via the Paule-Mandel estimator, with the application of random effects for I2 ≥ 40, and otherwise, a fixed-effects model was employed.

Mendelian randomization analysis

Acknowledging that individuals with MDD are predisposed to unhealthy dietary patterns, which in turn may precipitate MetS [66], we embarked on Mendelian randomization (MR) analysis to decipher the causal dynamics at play. This analysis incorporated GWAS summary data for 8 dietary habits (Supplementary Table 2) [67–70], conducting separate meta-analyses for each diet across East Asian and European ancestries using the IVW fixed-effects model, thus generating insights into multiethnic dietary influences15. Given the potential for sample overlap between the MAMT and multi-ancestry diet (MA Diet) datasets, such as those derived from the UK Biobank, we employed bidirectional MR analysis through MRlap. This method utilizes cross-trait LDSC to estimate overlap, thereby addressing and adjusting for potential biases [71]. We applied a clustering criterion of r2 = 0.2 within a 1000 kb window, with genome-wide significance set at P = 5 × 10− 8 as the threshold for selecting instrumental variables (IV) for exposure.

Results

Meta-analysis

Our study’s design and the analytical workflow are illustrated in Fig. 1. Initially, we assessed λmeta for each of the three European MDD GWAS cohorts, confirming the absence of significant sample overlap (λmeta > 1) as depicted in Supplementary Fig. 1. Based on the meta-analysis, we compiled MA EUR GWAS summary data, encompassing 20,016,977 SNPs and a cohort of 971,060 individuals, which included 245,959 patients with MDD and 725,101 controls.

Fig. 1.

Fig. 1

Open in a new tab

Flowchart. Input GWAS: PGC MDD EUR 170 756/329 443, FinnGen MDD EUR 47 696/359 290, iPSYCH MDD EUR 27 507/36 368, MDD EAS meta 21 980/360 956; 10 MetS traits EUR 44 696–476 326, EAS 36 422–173 646; 8 dietary habits EUR 1.13–1.45 M, EAS 153–165 k. Key thresholds: GWAS P ≤ 5 × 10⁻⁸, gene-based FDR < 0.05, TWAS P < 1.37 × 10⁻⁶, colocalisation/SuSIE PP.H4 > 0.7, MR FDR < 0.05, CMap τ < − 90. Abbreviations: MDD, Major Depressive Disorder; MAMT, Multi-ancestry Multi-trait meta-analysis; MGAMA, Multi-marker Analysis of GenoMic Annotation; POPS, Polygenic Priority Score; TWAS, Transcriptome-Wide Association Study; SMR, Summary-data-based Mendelian Randomization; CMap, Connectivity Map

Through LDSC analysis, genetic correlations were discerned between MA EUR and six metabolic indicators, whereas MA EAS exhibited correlations with five indicators. Remarkably, Body Mass Index and High-Density Lipoprotein Cholesterol levels were genetically correlated with MDD across both ancestries (Fig. 2 and Supplementary Table 4).

Fig. 2.

Fig. 2

Open in a new tab

Genetic correlation analysis of mdd with mets-related metrics across ancestries

Subsequent MTAG analyses of MA EUR with six metabolic traits, and MA EAS with five, facilitated the derivation of MT EUR and MT EAS, respectively. The IVW fixed-effects model enabled the amalgamation of findings from both ancestral backgrounds, culminating in the MAMT. Post-quality control, the analysis revealed 8,967,483 SNP results, with no indication of underlying population stratification (λ = 1.04) and a robust concordance between observed P-values and those inferred from Z-scores (Fig. 3, Supplementary Tables 3, and Supplementary Fig. 2).

Fig. 3.

Fig. 3

Open in a new tab

Manhattan Plot for MAMT. This figure accentuates the newly discovered SNVs associated with MDD, each situated more than a million base pairs apart from previously reported variants, adhering to a significance threshold of 5 × 10 − 8 (for aesthetic purposes, only the most significant SNV per chromosome is retained)

Moreover, we pinpointed 313 independent SNPs corresponding to 63 loci, each distanced by at least 500 kb, among which 106 SNVs marked novel MDD associations, distinct from previously identified GWAS loci by at least 1 million base pairs (Fig. 3). A multiple ancestry Bayesian fine mapping approach ascertained that nearly all SNVs had posterior probabilities of 0.99 or higher, with the exception of rs17231506 (cumSum = 0.80), suggesting no evidence of multiple independent signals (Supplementary Table 6). At a Bonferroni-adjusted threshold of 0.05/313, 305 SNVs demonstrated no significant heterogeneity as per METAL’s heterogeneity test (Supplementary Table 6).

Through Pearson correlation coefficient calculations, we found that the effect sizes of the 313 SNVs within MAMT were generally consistent with those observed in MA (a combined meta-analysis based on MA EUR and MA EAS), MT-EAS, and MT-EUR, with the correlations of the absolute values of effect sizes being r = 0.56, 0.99, and 0.65, respectively. However, the consistency with MA-EAS and MA-EUR was lesser, yet still presented moderate correlations, with the absolute values of effect sizes being r = 0.48 and 0.34, respectively (Supplementary Fig. 3).

Colocalization analysis revealed 66 SNVs (21.09% of total) with conditional posterior probabilities exceeding 0.7 (Supplementary Table 7), including 13 SNVs spanning 8 unique loci also surpassing a PP.H4 value of 0.7. Notably, the newly discovered SNV rs464161 colocalized across five East Asian lineage trait pairs. Moreover, rs11066015 demonstrated colocalization within trait pairs of both East Asian (e.g., Diastolic Blood Pressure-MA EAS) and European ancestries (e.g., Low-Density Lipoprotein Cholesterol Levels-MA EUR). To reduce reliance on the single-causal-variant assumption, we further performed SuSIE fine-mapping for these 13 loci (Wallace, 2021). Among them, 7 SNVs exhibited SuSIE posterior probability (SUSIE.PP.H4) > 0.7, all of which also showed PP.H4.abf > 0.7, providing robust evidence for colocalization with multiple causal variants (Supplementary Table 19).

ANNOVAR annotation of these SNVs identified 17 with CADD scores above the designated threshold (> 12.37), including three novel SNVs (rs2740480, rs429358, and rs72642490). Significantly, rs429358, a variant within the mRNA exonic region, is also linked to brain structure across multiple ancestries [72, 73], with the associated gene APOE highlighted in subsequent investigations (Supplementary Table 5).

Genetic analysis

To elucidate the biological underpinnings revealed by GWAS, we deployed a suite of computational strategies for the functional annotation and prioritization of genes potentially implicated in disease pathology. Initially, FUMA annotation pinpointed 1,612 genes of interest (Supplementary Table 8). Subsequent MAGMA analysis narrowed our focus to 409 genes (Supplementary Table 9), while POPS analysis highlighted 59 genes with PoPS scores exceeding 1 (Supplementary Table 10). Additionally, the mBAT-combo approach identified 563 genes related to the conditions under study (Supplementary Table 11).

TWAS further revealed 93 genes significantly associated with the phenotype (P < 1.37 × 10− 6), among which BUD13 (in blood, Z = 15.22, P = 2.52 × 10− 52) and CETP (in thyroid, Z=-32.42, P = 1.46 × 10− 230) were the most noteworthy (Fig. 4 and Supplementary Table 12). Notably, CCDC116 was significantly correlated across 14 tissue types, spanning various brain regions, and including the adrenal gland, pituitary, and thyroid, aligning with previous research [53].

Fig. 4.

Fig. 4

Open in a new tab

Manhattan Plot of Z-Scores for Genes Associated with MAMT in TWAS Significant gene associations are marked with red dots. For genes at the top, increased expression is associated with elevated MDD or MetS risk, whereas genes at the bottom show a negative correlation with expression

An SMR analysis identified 47 pleiotropic genes (Supplementary Table 13), among them repeatedly emphasized MDD-related genes like DCC. Remarkably, AEBP1, HCG22, and PM20D1 demonstrated significant correlations within astrocytes, while TM6SF2 was notably inversely correlated with phenotypes in excitatory neurons.

Integrating findings from the aforementioned six methodologies, we identified 65 genes concurrently featured in four or more techniques, designating them as high-confidence candicate genes. Of these, nine appeared across five methods (Fig. 5A). Our phenotypic analysis of these high-confidence genes revealed seven distinct phenotypes significantly associated with the genetic findings (Fig. 5B and Supplementary Table 14), encompassing muscle, hepatobiliary system, adipose tissue, homeostasis/metabolism, growth/body shape/body region, cellular, and cardiovascular system phenotypes. These results underscore MDD’s complexity and its linkage with MetS, highlighting aspects like muscle fatigue and diminished activity common among MDD sufferers, the implications of hepatobiliary system abnormalities, the association of adipose tissue phenotype with obesity and MDD, the perturbation in metabolic processes influencing mood and energy, and the increased cardiovascular disease risk in MDD patients, potentially attributable to lifestyle impacts and shared genetic factors [74–80].

Fig. 5.

Fig. 5

Open in a new tab

Pleiotropic Genes Analysis. (a) Nine pleiotropic genes identified using six distinct methodologies, each gene recognized by five approaches. The left side represents the risk genes, while the right-side details the various methods employed. (b) Parallel phenotype analysis of pleiotropic genes

Computational drug repurposing analysis

Through the extraction of TWAS Z-values for 49 credible pleiotropic genes, we identified 12 drug classes with clinical relevance using computational drug repurposing (CDR) strategies facilitated by the Connectivity Map (CMap) algorithm (Supplementary Table 15). Notably, some of the discerned drug classes, such as HDAC inhibitors (e.g., Pyroxamide) and Cytochrome P450 inhibitors, are already established in the clinical management of Major Depressive Disorder (MDD), lending credence to our findings [81–83]. Additionally, BCL inhibitors have demonstrated efficacy in mitigating depressive symptoms through their regulatory impact on Bcl-2 and Bcl-xl expression, alongside adjustments in insulin component expressions [84–86]. Furthermore, Piperlongumine, categorized as a Glutathione transferase inhibitor, has shown to counteract neuroinflammation by influencing the NF-KB pathway in lipopolysaccharide-induced BV2 microglial cells, exhibiting both anxiolytic and antidepressant properties [87, 88].

Protein association analysis

Utilizing the BLISS approach, we integrated plasma protein data across diverse ancestries with MT EUR and MT EAS findings, identifying 239 proteins with significant associations (P < 0.05) (Supplementary Table 16). Excluding seven proteins with data only from a single source, a meta-analysis on the remainder pinpointed 66 pertinent proteins. Heterogeneity assessments revealed that two proteins exhibited I2 > 40, necessitating the use of a random-effects model, while the rest were analyzed using fixed-effects models (Supplementary Table 17). Among these 73 pleiotropic proteins, GPHA2 emerged as the most significantly negatively correlated with the phenotype (Z = -5.83, P = 5.46 × 10^-9), and ARHGEF2 exhibited the most substantial positive correlation (Z = 5.70, P = 1.22 × 10^-8).

Mendelian randomization

The results from the multiethnic meta-analysis of eight dietary habits demonstrated well-calibrated λ values and LDSC intercepts, ensuring robustness in our findings (Supplementary Table 3). Further, our MR analysis delved into the causal relationships between MAMT and diverse dietary habits across ancestries (Supplementary Table 18). The analysis indicated that alcohol consumption may be linked to an increased risk of MDD and MetS, pointing to a critical area for lifestyle intervention. Conversely, data on coffee consumption relayed a positive signal, suggesting that its consumers might face a reduced risk for both conditions. Moreover, the association analysis regarding milk consumption suggested a potential risk increase for these diseases, though further research is required for confirmation. These insights are pivotal for understanding how diet, through genetic mechanisms, influences mental and metabolic health.

Discussion

This research leveraged a multitude of computational methodologies to unravel the intertwined biological processes underlying MDD and MetS. By amalgamating GWAS datasets across various ancestries, we delineated the intricate genetic interplay between MDD and MetS, underscored by nuanced genetic association analyses.

To clarify the novelty of our findings, we performed a systematic comparison with two landmark large-scale MDD GWAS: the European cohort study by Howard et al. [21] and the multi-ancestry analysis by Meng et al. [25]. Among our identified variants, 106 SNVs represent novel discoveries, distanced > 1 Mb from previously reported MDD loci with r² < 0.1 (indicating no linkage disequilibrium). These loci were underpowered in single-trait analyses but were successfully identified through our multi-ancestry multi-trait framework (MTAG λmeta = 1.04), underscoring the unique advantage of integrating MDD and MetS-related traits for pleiotropic locus discovery (Supplementary Table 20).

Notably, Bayesian fine mapping and co-localization analyses identified several associated SNVs linked to traits in both diseases. For instance, the newly discovered SNV rs464161 exhibited co-localization across traits within East Asian lineages, while rs11066015 demonstrated co-localization across traits in both East Asian and European ancestries. This locus, also implicated in GWAS for coffee consumption, corroborates findings from subsequent MR analyses [89], aligning with further MR outcomes. Moreover, rs429358, a variant located in an mRNA exonic region associated with brain structure across ancestries [72, 73], and its related gene APOE, were identified among our likely pleiotropic genes. Existing research has revealed that the rs429358 mutation augments MDD prevalence by 2.17 times, correlating with elevated hypercholesterolemia and coronary heart disease risks [90–92], potentially due to this locus’s impact on APOE protein structure and function, thus affecting its receptor binding affinity, lipid metabolism, and cellular repair mechanisms [93–95]. The association between this variant and both MDD and MetS might reflect its dual regulatory role in brain and systemic metabolism, proposing APOE as a critical nexus for mental and metabolic health [96, 97].

Our genetic investigation underscored 65 probable pleiotropic genes crucial to disease pathogenesis and progression. Utilizing diverse bioinformatics instruments, such as TWAS and SMR, has deepened our comprehension of the gene-phenotype nexus, offering fresh insights into the mechanisms underpinning these genes’ roles in brain functionality and structure.

The drug repurposing analysis outcomes are promising, generating hypotheses for novel therapeutic possibilities for MDD and MetS. Specifically, HDAC inhibitors’ identification underscores their potential broader application in MDD management. Additionally, the discovery of Cytochrome P450 inhibitors supports antidepressant therapy diversification, potentially aiding depressive symptom alleviation by modifying neurotransmitter metabolism within the brain. Moreover, due to their influence on drug metabolism, such inhibitors could enhance other antidepressants’ efficacy and minimize adverse effects [98–100]. Nonetheless, employing Cytochrome P450 inhibitors in MDD and MetS treatment necessitates caution, given their broad impact on drug metabolism, necessitating vigilant monitoring for drug interactions and potential side effects during clinical usage [101–104].

Proteomic analyses, facilitated by BLISS and meta-analysis, have furnished insights into proteins associated with MDD and MetS across ancestries, shedding light on diseases’ molecular underpinnings. Discoveries like GPHA2 and ARHGEF2 not only highlight these proteins’ biological significance but may also inform future treatment strategies. This proteomic approach promises a more profound understanding of how genetic findings translate into complex diseases’ molecular characteristics.

Moreover, MR analysis delved into the causal links between dietary habits and MDD and MetS. We found that alcohol consumption elevates both conditions’ risk, whereas coffee consumption seems to mitigate risk, aligning with extant cohort study outcomes [105–108]. These insights underscore the importance of healthy living in preventing and managing these conditions.

In essence, our study elucidates the genetic complexity interlinking MDD and MetS, paving the way for further functional explorations and potential clinical applications. Future endeavors should integrate more data dimensions and functional studies on genetic variation’s impact on disease phenotypes and treatment responses. Moreover, probing the intricate gene-phenotype interactions will foster new avenues for uncovering disease mechanisms and therapeutic targets.

However, this study has several important limitations that should be considered when interpreting our findings. Firstly, a primary constraint is that our key findings—including the prioritized pleiotropic genes, proteins, and drug candidates—are derived entirely from computational analyses of summary-level data. These results represent statistical associations and predictions that require validation through functional studies (e.g., in vitro and in vivo model systems) to confirm their biological relevance and therapeutic potential. Specifically, our drug repurposing analysis relied solely on CMap-derived transcriptomic signatures without experimental validation; while these in silico predictions provide valuable hypotheses, their clinical applicability must be substantiated by extensive laboratory and clinical investigation. Secondly, although we utilized multi-ancestry data, the analytical power was likely uneven across populations due to disparities in sample sizes and SNP coverage between European and East Asian cohorts; consequently, the generalizability of specific loci and mechanisms across all ancestral groups necessitates further dedicated investigation. Thirdly, our Mendelian randomization analysis provides evidence consistent with causal relationships but is subject to key assumptions (e.g., absence of pleiotropy and horizontal pleiotropy) that cannot be fully verified using summary-level data alone; therefore, these results should be interpreted as supportive rather than definitive proof of causation. Additionally, as noted, the absence of individual-level data access constrained our ability to explore complex genetic interactions, and our reliance on public databases means the completeness and accuracy of these datasets directly influenced our outcomes. Future research must address these limitations by integrating functional validation experiments, more comprehensive and balanced multi-ancestry datasets, and rigorous experimental follow-up to translate these computational findings into clinically meaningful insights.

Recent genetic advances offer new tools for early detection and personalized treatment of MDD and MetS. The GWAShug platform [60] provides a valuable resource to decode shared genetic bases of complex traits using summary statistics, supporting future research into the genetic complexity of MDD and MetS. Additionally, polygenic scores (PGS) calculated using tools like DBSLMM [109] or obtained from platforms such as PGS-Depot and PGS-server can enhance risk prediction for complex diseases [110, 111]. Cai et al. demonstrated improved risk prediction accuracy for ischemic stroke by integrating transcriptomic data with PGS [112], suggesting similar potential for MDD and MetS.

However, beyond genetics, socioeconomic status (SES) and lifestyle factors significantly impact MDD and MetS [113]. Low SES is associated with higher disease prevalence due to poor living conditions and limited health resources. Future research should consider SES in understanding disease mechanisms. Improving health education and living conditions, particularly for low-SES populations, could substantially reduce MDD and MetS prevalence.

Supplementary Information

Below is the link to the electronic supplementary material.

12888_2026_7816_MOESM1_ESM.pdf (429.8KB, pdf)

Supplementary Material 1: Supplementary Figs. 1–3: These figures provide additional visual data and analyses that support the findings presented in the main text.

Supplementary Material 2 (913.2KB, docx)
Supplementary Material 3 (556.7KB, xlsx)
12888_2026_7816_MOESM4_ESM.pdf (5.9MB, pdf)

Supplementary Material 4: Supplementary Tables 1–18: These tables offer detailed statistical results, extended datasets, and comprehensive methodological descriptions that complement the main manuscript.

Acknowledgements

We acknowledge that an earlier version of this manuscript has been presented as a preprint. The preprint version can be accessed at the following link: [https://www.researchsquare.com/article/rs-4193051/v1]. In addition, please ensure to properly cite any original sources in the main text of your manuscript. You can check the mentioned source at [https://www.researchsquare.com/article/rs-4193051/v1].

Author contributions

Y.F. and Y.L. contributed equally to the study by conceptualizing the research, performing the data analysis, and drafting the manuscript. H.Z. and H.X. were responsible for the collection and preprocessing of the data. Y.L., H.Z., and P.H. provided expertise in genetic epidemiology and guided the analytical approach. P.H. supervised the study, secured funding, and contributed to the interpretation of results. All authors critically reviewed and approved the final version of the manuscript.

Funding

This study did not receive financial support from any external agencies, and there are no relevant financial conflicts of interest.

Data availability

The data that support the findings of this study are openly available in MD and METs at https://fwhyu-capybaralab.shinyapps.io/md_and_mets/.

Declarations

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Yiheng Feng and Yihao Li contributed equally to this work.

References

  • 1.Smith K. Mental health: a world of depression. Nature. 2014;515(7526):181. [DOI] [PubMed] [Google Scholar]
  • 2.Uchida S, Yamagata H, Seki T, Watanabe Y. Epigenetic mechanisms of major depression: targeting neuronal plasticity. Psychiatry Clin Neurosci. 2018;72(4):212–27. [DOI] [PubMed] [Google Scholar]
  • 3.Uher R, Payne JL, Pavlova B, Perlis RH. Major depressive disorder in DSM-5: implications for clinical practice and research of changes from DSM-IV. Depress Anxiety. 2014;31(6):459–71. [DOI] [PubMed] [Google Scholar]
  • 4.Schramm E, Klein DN, Elsaesser M, Furukawa TA, Domschke K. Review of dysthymia and persistent depressive disorder: history, correlates, and clinical implications. Lancet Psychiatry. 2020;7(9):801–12. [DOI] [PubMed] [Google Scholar]
  • 5.Vancampfort D, Stubbs B, Mitchell AJ, De Hert M, Wampers M, Ward PB, Rosenbaum S, Correll CU. Risk of metabolic syndrome and its components in people with schizophrenia and related psychotic disorders, bipolar disorder and major depressive disorder: a systematic review and meta-analysis. World Psychiatry. 2015;14(3):339–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Al-Khatib Y, Akhtar MA, Kanawati MA, Mucheke R, Mahfouz M, Al-Nufoury M. Depression and metabolic syndrome: A narrative review. Cureus. 2022;14(2):e22153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Marazziti D, Arone A, Palermo S, Annuzzi E, Cappellato G, Chiarantini I, Prete LD, Dell’Osso L. The wicked relationship between depression and metabolic syndrome. Clin Neuropsychiatry. 2023;20(2):100–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Lamers F, Vogelzangs N, Merikangas KR, de Jonge P, Beekman AT, Penninx BW. Evidence for a differential role of HPA-axis function, inflammation and metabolic syndrome in melancholic versus atypical depression. Mol Psychiatry. 2013;18(6):692–9. [DOI] [PubMed] [Google Scholar]
  • 9.Slavich GM, Irwin MR. From stress to inflammation and major depressive disorder: a social signal transduction theory of depression. Psychol Bull. 2014;140(3):774–815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chan KL, Poller WC, Swirski FK, Russo SJ. Central regulation of stress-evoked peripheral immune responses. Nat Rev Neurosci. 2023;24(10):591–604. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Sekula P, Del Greco MF, Pattaro C, Köttgen A. Mendelian randomization as an approach to assess causality using observational data. J Am Soc Nephrol. 2016;27(11):3253–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Postolache TT, Del Bosque-Plata L, Jabbour S, Vergare M, Wu R, Gragnoli C. Co-shared genetics and possible risk gene pathway partially explain the comorbidity of schizophrenia, major depressive disorder, type 2 diabetes, and metabolic syndrome. Am J Med Genet B Neuropsychiatr Genet. 2019;180(3):186–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Hagenaars SP, Coleman JRI, Choi SW, Gaspar H, Adams MJ, Howard DM, Hodgson K, Traylor M, Air TM, Andlauer TFM, et al. Genetic comorbidity between major depression and cardio-metabolic traits, stratified by age at onset of major depression. Am J Med Genet B Neuropsychiatr Genet. 2020;183(6):309–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Gao X, Qin Y, Jiao S, Hao J, Zhao J, Wang J, Wen Y, Wang T. Genetic evidence for the causal relations between metabolic syndrome and psychiatric disorders: a Mendelian randomization study. Transl Psychiatry. 2024;14(1):46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Zhang Y, Lu Q, Ye Y, Huang K, Liu W, Wu Y, Zhong X, Li B, Yu Z, Travers BG, et al. SUPERGNOVA: local genetic correlation analysis reveals heterogeneous etiologic sharing of complex traits. Genome Biol. 2021;22(1):262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Gong W, Guo P, Li Y, Liu L, Yan R, Liu S, Wang S, Xue F, Zhou X, Yuan Z. Role of the Gut-Brain axis in the shared genetic etiology between Gastrointestinal tract diseases and psychiatric disorders: A Genome-Wide pleiotropic analysis. JAMA Psychiatry. 2023;80(4):360–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Solovieff N, Cotsapas C, Lee PH, Purcell SM, Smoller JW. Pleiotropy in complex traits: challenges and strategies. Nat Rev Genet. 2013;14(7):483–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal Pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Chung D, Yang C, Li C, Gelernter J, Zhao H. GPA: a statistical approach to prioritizing GWAS results by integrating Pleiotropy and annotation. PLoS Genet. 2014;10(11):e1004787. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Lee CH, Shi H, Pasaniuc B, Eskin E, Han B. PLEIO: a method to map and interpret pleiotropic loci with GWAS summary statistics. Am J Hum Genet. 2021;108(1):36–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Howard DM, Adams MJ, Clarke TK, Hafferty JD, Gibson J, Shirali M, Coleman JRI, Hagenaars SP, Ward J, Wigmore EM, et al. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci. 2019;22(3):343–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, Reeve MP, Laivuori H, Aavikko M, Kaunisto MA, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613(7944):508–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Buniello A, MacArthur JAL, Cerezo M, Harris LW, Hayhurst J, Malangone C, McMahon A, Morales J, Mountjoy E, Sollis E, et al. The NHGRI-EBI GWAS catalog of published genome-wide association studies, targeted arrays and summary statistics 2019. Nucleic Acids Res. 2019;47(D1):D1005–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Pedersen EM, Agerbo E, Plana-Ripoll O, Steinbach J, Krebs MD, Hougaard DM, Werge T, Nordentoft M, Børglum AD, Musliner KL, et al. ADuLT: an efficient and robust time-to-event GWAS. Nat Commun. 2023;14(1):5553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Meng X, Navoly G, Giannakopoulou O, Levey DF, Koller D, Pathak GA, Koen N, Lin K, Adams MJ, Rentería ME, et al. Multi-ancestry genome-wide association study of major depression aids locus discovery, fine mapping, gene prioritization and causal inference. Nat Genet. 2024;56(2):222–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Loh PR, Kichaev G, Gazal S, Schoech AP, Price AL. Mixed-model association for biobank-scale datasets. Nat Genet. 2018;50(7):906–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Pulit SL, Stoneman C, Morris AP, Wood AR, Glastonbury CA, Tyrrell J, Yengo L, Ferreira T, Marouli E, Ji Y, et al. Meta-analysis of genome-wide association studies for body fat distribution in 694 649 individuals of European ancestry. Hum Mol Genet. 2019;28(1):166–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Wang H, Zhang F, Zeng J, Wu Y, Kemper KE, Xue A, Zhang M, Powell JE, Goddard ME, Wray NR, et al. Genotype-by-environment interactions inferred from genetic effects on phenotypic variability in the UK biobank. Sci Adv. 2019;5(8):eaaw3538. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Barton AR, Sherman MA, Mukamel RE, Loh PR. Whole-exome imputation within UK biobank powers rare coding variant association and fine-mapping analyses. Nat Genet. 2021;53(8):1260–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Graham SE, Clarke SL, Wu KH, Kanoni S, Zajac GJM, Ramdas S, Surakka I, Ntalla I, Vedantam S, Winkler TW, et al. The power of genetic diversity in genome-wide association studies of lipids. Nature. 2021;600(7890):675–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Lagou V, Jiang L, Ulrich A, Zudina L, González KSG, Balkhiyarova Z, Faggian A, Maina JG, Chen S, Todorov PV, et al. GWAS of random glucose in 476,326 individuals provide insights into diabetes pathophysiology, complications and treatment stratification. Nat Genet. 2023;55(9):1448–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Chen CY, Chen TT, Feng YA, Yu M, Lin SC, Longchamps RJ, Wang SH, Hsu YH, Yang HI, Kuo PH, et al. Analysis across Taiwan biobank, biobank Japan, and UK biobank identifies hundreds of novel loci for 36 quantitative traits. Cell Genom. 2023;3(12):100436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Chen GB, Lee SH, Robinson MR, Trzaskowski M, Zhu ZX, Winkler TW, Day FR, Croteau-Chonka DC, Wood AR, Locke AE, et al. Across-cohort QC analyses of GWAS summary statistics from complex traits. Eur J Hum Genet. 2016;25(1):137–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Khunsriraksakul C, Li Q, Markus H, Patrick MT, Sauteraud R, McGuire D, Wang X, Wang C, Wang L, Chen S, et al. Multi-ancestry and multi-trait genome-wide association meta-analyses inform clinical risk prediction for systemic lupus erythematosus. Nat Commun. 2023;14(1):668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics. 2010;26(17):2190–1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Han B, Eskin E. Random-effects model aimed at discovering associations in meta-analysis of genome-wide association studies. Am J Hum Genet. 2011;88(5):586–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Bulik-Sullivan BK, Loh PR, Finucane HK, Ripke S, Yang J, Patterson N, Daly MJ, Price AL, Neale BM. LD score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet. 2015;47(3):291–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA, Abecasis GR. A global reference for human genetic variation. Nature. 2015;526(7571):68–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Turley P, Walters RK, Maghzian O, Okbay A, Lee JJ, Fontana MA, Nguyen-Viet TA, Wedow R, Zacher M, Furlotte NA, et al. Multi-trait analysis of genome-wide association summary statistics using MTAG. Nat Genet. 2018;50(2):229–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Guo P, Gong W, Li Y, Liu L, Yan R, Wang Y, Zhang Y, Yuan Z. Pinpointing novel risk loci for lewy body dementia and the shared genetic etiology with alzheimer’s disease and parkinson’s disease: a large-scale multi-trait association analysis. BMC Med. 2022;20(1):214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Watanabe K, Taskesen E, van Bochoven A, Posthuma D. Functional mapping and annotation of genetic associations with FUMA. Nat Commun. 2017;8(1):1826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010;38(16):e164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res. 2019;47(D1):D886–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Dong S, Zhao N, Spragins E, Kagda MS, Li M, Assis P, Jolanki O, Luo Y, Cherry JM, Boyle AP, et al. Annotating and prioritizing human non-coding variants with RegulomeDB v.2. Nat Genet. 2023;55(5):724–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Lam M, Chen CY, Li Z, Martin AR, Bryois J, Ma X, Gaspar H, Ikeda M, Benyamin B, Brown BC, et al. Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet. 2019;51(12):1670–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Gormley P, Anttila V, Winsvold BS, Palta P, Esko T, Pers TH, Farh KH, Cuenca-Leon E, Muona M, Furlotte NA, et al. Meta-analysis of 375,000 individuals identifies 38 susceptibility loci for migraine. Nat Genet. 2016;48(8):856–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Giambartolomei C, Vukcevic D, Schadt EE, Franke L, Hingorani AD, Wallace C, Plagnol V. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 2014;10(5):e1004383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Huang X, Yao M, Tian P, Wong JYY, Li Z, Liu Z, Zhao JV. Genome-wide cross-trait analysis and Mendelian randomization reveal a shared genetic etiology and causality between COVID-19 and venous thromboembolism. Commun Biol. 2023;6(1):441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Wallace C. A more accurate method for colocalisation analysis allowing for multiple causal variants. PLoS Genet. 2021;17(9):e1009440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.de Leeuw CA, Mooij JM, Heskes T, Posthuma D. MAGMA: generalized gene-set analysis of GWAS data. PLoS Comput Biol. 2015;11(4):e1004219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Weeks EM, Ulirsch JC, Cheng NY, Trippe BL, Fine RS, Miao J, Patwardhan TA, Kanai M, Nasser J, Fulco CP, et al. Leveraging polygenic enrichments of gene features to predict genes underlying complex traits and diseases. Nat Genet. 2023;55(8):1267–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Li A, Liu S, Bakshi A, Jiang L, Chen W, Zheng Z, Sullivan PF, Visscher PM, Wray NR, Yang J, et al. mBAT-combo: A more powerful test to detect gene-trait associations from GWAS data. Am J Hum Genet. 2023;110(1):30–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Dall’Aglio L, Lewis CM, Pain O. Delineating the genetic component of gene expression in major depression. Biol Psychiatry. 2021;89(6):627–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Gusev A, Ko A, Shi H, Bhatia G, Chung W, Penninx BW, Jansen R, de Geus EJ, Boomsma DI, Wright FA, et al. Integrative approaches for large-scale transcriptome-wide association studies. Nat Genet. 2016;48(3):245–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Qi T, Wu Y, Fang H, Zhang F, Liu S, Zeng J, Yang J. Genetic control of RNA splicing and its distinct role in complex trait variation. Nat Genet. 2022;54(9):1355–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Võsa U, Claringbould A, Westra HJ, Bonder MJ, Deelen P, Zeng B, Kirsten H, Saha A, Kreuzhuber R, Yazar S, et al. Large-scale cis- and trans-eQTL analyses identify thousands of genetic loci and polygenic scores that regulate blood gene expression. Nat Genet. 2021;53(9):1300–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Bryois J, Calini D, Macnair W, Foo L, Urich E, Ortmann W, Iglesias VA, Selvaraj S, Nutma E, Marzin M, et al. Cell-type-specific cis-eQTLs in eight human brain cell types identify novel risk genes for psychiatric and neurological disorders. Nat Neurosci. 2022;25(8):1104–12. [DOI] [PubMed] [Google Scholar]
  • 58.Jerber J, Seaton DD, Cuomo ASE, Kumasaka N, Haldane J, Steer J, Patel M, Pearce D, Andersson M, Bonder MJ, et al. Population-scale single-cell RNA-seq profiling across dopaminergic neuron differentiation. Nat Genet. 2021;53(3):304–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Zhu Z, Zhang F, Hu H, Bakshi A, Robinson MR, Powell JE, Montgomery GW, Goddard ME, Wray NR, Visscher PM, et al. Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat Genet. 2016;48(5):481–7. [DOI] [PubMed] [Google Scholar]
  • 60.Cao C, Tian M, Li Z, Zhu W, Huang P, Yang S. GWAShug: a comprehensive platform for decoding the shared genetic basis between complex traits based on summary statistics. Nucleic Acids Res. 2025;53(D1):D1006–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Blake JA, Baldarelli R, Kadin JA, Richardson JE, Smith CL, Bult CJ. Mouse genome database (MGD): knowledgebase for mouse-human comparative biology. Nucleic Acids Res. 2021;49(D1):D981–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Wang Z, Lachmann A, Keenan AB, Ma’ayan A. L1000FWD: fireworks visualization of drug-induced transcriptomic signatures. Bioinformatics. 2018;34(12):2150–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Wu C, Zhang Z, Yang X, Zhao B. Large-scale imputation models for multi-ancestry proteome-wide association analysis. bioRxiv. 2023:2023.2010.2005.561120.
  • 64.Shao K, Luo Z, Huang P, Yang S. ProteoNexus: an integrative database to characterize genetic architecture, estimate mediation effects, and construct and evaluate prediction models of the plasma proteome. Nucleic Acids Res. 2025. [DOI] [PMC free article] [PubMed]
  • 65.Viechtbauer W. Conducting Meta-Analyses in R with the metafor package. J Stat Softw. 2010;36(3):1–48. [Google Scholar]
  • 66.Kim IS, Hwang JY. Does better diet quality offset the association between depression and metabolic syndrome? Nutrients. 2023;15(4). [DOI] [PMC free article] [PubMed]
  • 67.Karlsson Linnér R, Biroli P, Kong E, Meddens SFW, Wedow R, Fontana MA, Lebreton M, Tino SP, Abdellaoui A, Hammerschlag AR, et al. Genome-wide association analyses of risk tolerance and risky behaviors in over 1 million individuals identify hundreds of loci and shared genetic influences. Nat Genet. 2019;51(2):245–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Matoba N, Akiyama M, Ishigaki K, Kanai M, Takahashi A, Momozawa Y, Ikegawa S, Ikeda M, Iwata N, Hirata M, et al. GWAS of 165,084 Japanese individuals identified nine loci associated with dietary habits. Nat Hum Behav. 2020;4(3):308–16. [DOI] [PubMed] [Google Scholar]
  • 69.Cole JB, Florez JC, Hirschhorn JN. Comprehensive genomic analysis of dietary habits in UK biobank identifies hundreds of genetic associations. Nat Commun. 2020;11(1):1467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Pirastu N, McDonnell C, Grzeszkowiak EJ, Mounier N, Imamura F, Merino J, Day FR, Zheng J, Taba N, Concas MP, et al. Using genetic variation to disentangle the complex relationship between food intake and health outcomes. PLoS Genet. 2022;18(6):e1010162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Mounier N, Kutalik Z. Bias correction for inverse variance weighting Mendelian randomization. Genet Epidemiol. 2023;47(4):314–31. [DOI] [PubMed] [Google Scholar]
  • 72.Park JY, Lee D, Lee JJ, Gim J, Gunasekaran TI, Choi KY, Kang S, Do AR, Jo J, Park J, et al. A missense variant in SHARPIN mediates alzheimer’s disease-specific brain damages. Transl Psychiatry. 2021;11(1):590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Brouwer RM, Klein M, Grasby KL, Schnack HG, Jahanshad N, Teeuw J, Thomopoulos SI, Sprooten E, Franz CE, Gogtay N, et al. Genetic variants associated with longitudinal changes in brain structure across the lifespan. Nat Neurosci. 2022;25(4):421–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 74.Trivedi MH. The link between depression and physical symptoms. Prim Care Companion J Clin Psychiatry. 2004;6(Suppl 1):12–6. [PMC free article] [PubMed] [Google Scholar]
  • 75.van Os E, van den Broek WW, Mulder PG, ter Borg PC, Bruijn JA, van Buuren HR. Depression in patients with primary biliary cirrhosis and primary sclerosing cholangitis. J Hepatol. 2007;46(6):1099–103. [DOI] [PubMed] [Google Scholar]
  • 76.Fu X, Wang Y, Zhao F, Cui R, Xie W, Liu Q, Yang W. Shared biological mechanisms of depression and obesity: focus on adipokines and lipokines. Aging. 2023;15(12):5917–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Gu X, Ke S, Wang Q, Zhuang T, Xia C, Xu Y, Yang L, Zhou M. Energy metabolism in major depressive disorder: recent advances from omics technologies and imaging. Biomed Pharmacother. 2021;141:111869. [DOI] [PubMed] [Google Scholar]
  • 78.Paul ER, Östman L, Heilig M, Mayberg HS, Hamilton JP. Towards a multilevel model of major depression: genes, immuno-metabolic function, and cortico-striatal signaling. Transl Psychiatry. 2023;13(1):171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.Nemeroff CB, Goldschmidt-Clermont PJ. Heartache and heartbreak–the link between depression and cardiovascular disease. Nat Rev Cardiol. 2012;9(9):526–39. [DOI] [PubMed] [Google Scholar]
  • 80.Dhar AK, Barton DA. Depression and the link with cardiovascular disease. Front Psychiatry. 2016;7:33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.Covington HE 3rd, Maze I, LaPlant QC, Vialou VF, Ohnishi YN, Berton O, Fass DM, Renthal W, Rush AJ 3rd, Wu EY, et al. Antidepressant actions of histone deacetylase inhibitors. J Neurosci. 2009;29(37):11451–60. [DOI] [PMC free article] [PubMed]
  • 82.Sahafnejad Z, Ramazi S, Allahverdi A. An update of epigenetic drugs for the treatment of cancers and brain diseases: a comprehensive review. Genes (Basel). 2023;14(4). [DOI] [PMC free article] [PubMed]
  • 83.Perlis RH. Cytochrome P450 genotyping and antidepressants. BMJ. 2007;334(7597):759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 84.Kosten TA, Galloway MP, Duman RS, Russell DS, D’Sa C. Repeated unpredictable stress and antidepressants differentially regulate expression of the bcl-2 family of apoptotic genes in rat cortical, hippocampal, and limbic brain structures. Neuropsychopharmacology. 2008;33(7):1545–58. [DOI] [PubMed] [Google Scholar]
  • 85.Tinahones FJ, Coín Aragüez L, Murri M, Oliva Olivera W, Mayas Torres MD, Barbarroja N, Gomez Huelgas R, Malagón MM, El Bekay R. Caspase induction and BCL2 Inhibition in human adipose tissue: a potential relationship with insulin signaling alteration. Diabetes Care. 2013;36(3):513–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 86.Zhang C, Wu Z, Hong W, Wang Z, Peng D, Chen J, Yuan C, Yu S, Xu L, Fang Y. Influence of BCL2 gene in major depression susceptibility and antidepressant treatment outcome. J Affect Disord. 2014;155:288–94. [DOI] [PubMed] [Google Scholar]
  • 87.Kumar S, Kamboj J, Suman, Sharma S. Overview for various aspects of the health benefits of Piper longum linn. Fruit. J Acupunct Meridian Stud. 2011;4(2):134–40. [DOI] [PubMed] [Google Scholar]
  • 88.Kim N, Do J, Bae JS, Jin HK, Kim JH, Inn KS, Oh MS, Lee JK. Piperlongumine inhibits neuroinflammation via regulating NF-κB signaling pathways in lipopolysaccharide-stimulated BV2 microglia cells. J Pharmacol Sci. 2018;137(2):195–201. [DOI] [PubMed] [Google Scholar]
  • 89.Jin T, Youn J, Kim AN, Kang M, Kim K, Sung J, Lee JE. Interactions of Habitual Coffee Consumption by Genetic Polymorphisms with the Risk of Prediabetes and Type 2 Diabetes Combined. Nutrients. 2020;12(8). [DOI] [PMC free article] [PubMed]
  • 90.Lumsden AL, Mulugeta A, Zhou A, Hyppönen E. Apolipoprotein E (APOE) genotype-associated disease risks: a phenome-wide, registry-based, case-control study utilising the UK Biobank. EBioMedicine. 2020;59:102954. [DOI] [PMC free article] [PubMed]
  • 91.Huang YP, Xue JJ, Li C, Chen X, Fu HJ, Fei T, Bi PX. Depression and APOEε4 status in individuals with subjective cognitive decline: A Meta-Analysis. Psychiatry Investig. 2020;17(9):858–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 92.Gafarov V, Gromova E, Shakhtshneider E, Gagulin I, Gafarova A. Association of the APOE gene polymorphism with depression in white adults in the WHO MONICA-Psychosocial program. J Pers Med. 2023;13(9). [DOI] [PMC free article] [PubMed]
  • 93.Zhen J, Huang X, Van Halm-Lutterodt N, Dong S, Ma W, Xiao R, Yuan L. ApoE rs429358 and rs7412 polymorphism and gender differences of serum lipid profile and cognition in aging Chinese population. Front Aging Neurosci. 2017;9:248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 94.Wang C, Ji X, Tang Z, Zhang Z, Gu X, Fang X. Combined homocysteine and ApoE rs429358 and rs7412 polymorphism in association with serum lipid levels and cognition in Chinese community-dwelling older adults. BMC Psychiatry. 2022;22(1):223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 95.Yang LG, March ZM, Stephenson RA, Narayan PS. Apolipoprotein E in lipid metabolism and neurodegenerative disease. Trends Endocrinol Metab. 2023;34(8):430–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 96.Li XB, Wang J, Xu AD, Huang JM, Meng LQ, Huang RY, Wang JL. Apolipoprotein E polymorphisms increase the risk of post-stroke depression. Neural Regen Res. 2016;11(11):1790–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 97.Rao H, Wu H, Yu Z, Huang Q. APOe genetic polymorphism rs7412 T/T genotype may be a risk factor for essential hypertension among Hakka People in Southern China. Int J Hypertens. 2022;2022:8145896. [DOI] [PMC free article] [PubMed]
  • 98.Porcelli S, Fabbri C, Spina E, Serretti A, De Ronchi D. Genetic polymorphisms of cytochrome P450 enzymes and antidepressant metabolism. Expert Opin Drug Metab Toxicol. 2011;7(9):1101–15. [DOI] [PubMed] [Google Scholar]
  • 99.Höfer P, Schosser A, Calati R, Serretti A, Massat I, Kocabas NA, Konstantinidis A, Linotte S, Mendlewicz J, Souery D, et al. The impact of cytochrome P450 CYP1A2, CYP2C9, CYP2C19 and CYP2D6 genes on suicide attempt and suicide risk-a European multicentre study on treatment-resistant major depressive disorder. Eur Arch Psychiatry Clin Neurosci. 2013;263(5):385–91. [DOI] [PubMed] [Google Scholar]
  • 100.Calabrò M, Fabbri C, Kasper S, Zohar J, Souery D, Montgomery S, Albani D, Forloni G, Ferentinos P, Rujescu D, et al. Metabolizing status of CYP2C19 in response and side effects to medications for depression: results from a naturalistic study. Eur Neuropsychopharmacol. 2022;56:100–11. [DOI] [PubMed] [Google Scholar]
  • 101.Davidson MD, Ballinger KR, Khetani SR. Long-term exposure to abnormal glucose levels alters drug metabolism pathways and insulin sensitivity in primary human hepatocytes. Sci Rep. 2016;6:28178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 102.Gravel S, Chiasson JL, Dallaire S, Turgeon J, Michaud V. Evaluating the impact of type 2 diabetes mellitus on CYP450 metabolic activities: protocol for a case-control Pharmacokinetic study. BMJ Open. 2018;8(2):e020922. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 103.Dos Santos LRB, Fleming I. Role of cytochrome P450-derived, polyunsaturated fatty acid mediators in diabetes and the metabolic syndrome. Prostaglandins Other Lipid Mediat. 2020;148:106407. [DOI] [PubMed] [Google Scholar]
  • 104.Zhao M, Ma J, Li M, Zhang Y, Jiang B, Zhao X, Huai C, Shen L, Zhang N, He L, et al. Cytochrome P450 enzymes and drug metabolism in humans. Int J Mol Sci. 2021;22(23). [DOI] [PMC free article] [PubMed]
  • 105.Wang L, Shen X, Wu Y, Zhang D. Coffee and caffeine consumption and depression: A meta-analysis of observational studies. Aust N Z J Psychiatry. 2016;50(3):228–42. [DOI] [PubMed] [Google Scholar]
  • 106.Grosso G, Micek A, Castellano S, Pajak A, Galvano F. Coffee, tea, caffeine and risk of depression: A systematic review and dose-response meta-analysis of observational studies. Mol Nutr Food Res. 2016;60(1):223–34. [DOI] [PubMed] [Google Scholar]
  • 107.Corbi-Cobo-Losey MJ, Martinez-Gonzalez M, Gribble AK, Fernandez-Montero A, Navarro AM, Domínguez LJ, Bes-Rastrollo M, Toledo E. Coffee Consumption and the Risk of Metabolic Syndrome in the ‘Seguimiento Universidad de Navarra’ Project. Antioxid (Basel). 2023;12(3). [DOI] [PMC free article] [PubMed]
  • 108.Lee J, Go TH, Min S, Koh SB, Choi JR. Association between lifestyle factors and metabolic syndrome in general populations with depressive symptoms in cross-setional based cohort study of Ansung-Ansan. PLoS ONE. 2022;17(3):e0262526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 109.Yang S, Zhou X. Accurate and scalable construction of polygenic scores in large biobank data sets. Am J Hum Genet. 2020;106(5):679–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 110.Cao C, Zhang S, Wang J, Tian M, Ji X, Huang D, Yang S, Gu N. PGS-Depot: a comprehensive resource for polygenic scores constructed by summary statistics based methods. Nucleic Acids Res. 2024;52(D1):D963–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 111.Yang S, Zhou X. PGS-server: accuracy, robustness and transferability of polygenic score methods for biobank scale studies. Brief Bioinform. 2022;23(2). [DOI] [PubMed]
  • 112.Cai X, Li H, Cao X, Ma X, Zhu W, Xu L, Yang S, Yu R, Huang P. Integrating transcriptomic and polygenic risk scores to enhance predictive accuracy for ischemic stroke subtypes. Hum Genet. 2024;144(1):43–54. [DOI] [PubMed] [Google Scholar]
  • 113.Ye X, Wang Y, Zou Y, Tu J, Tang W, Yu R, Yang S, Huang P. Associations of socioeconomic status with infectious diseases mediated by lifestyle, environmental pollution and chronic comorbidities: a comprehensive evaluation based on UK biobank. Infect Dis Poverty. 2023; 12(1). [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

12888_2026_7816_MOESM1_ESM.pdf (429.8KB, pdf)

Supplementary Material 1: Supplementary Figs. 1–3: These figures provide additional visual data and analyses that support the findings presented in the main text.

Supplementary Material 2 (913.2KB, docx)
Supplementary Material 3 (556.7KB, xlsx)
12888_2026_7816_MOESM4_ESM.pdf (5.9MB, pdf)

Supplementary Material 4: Supplementary Tables 1–18: These tables offer detailed statistical results, extended datasets, and comprehensive methodological descriptions that complement the main manuscript.

Data Availability Statement

The data that support the findings of this study are openly available in MD and METs at https://fwhyu-capybaralab.shinyapps.io/md_and_mets/.

Articles from BMC Psychiatry are provided here courtesy of BMC

RESOURCES