Context-specific eQTL dynamics uncover genetic pleiotropy and chromatin-mediated target gene switching in schizophrenia

Context-specific eQTL dynamics uncover genetic pleiotropy and chromatin-mediated target gene switching in schizophrenia | 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 Context-specific eQTL dynamics uncover genetic pleiotropy and chromatin-mediated target gene switching in schizophrenia Linyan Ye, Zongrui Shen, Qi Yang, Xiaohui Wu, JunPing Ye, Zhongwei Li, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6582362/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 Schizophrenia is a highly heritable psychiatric disorder, but functional mechanisms linking genetic risk to pathogenesis remain poorly understood. Most schizophrenia-associated variants reside in non-coding regions, highlighting the need to investigate their regulatory roles. This study employs context-specific expression quantitative trait loci (eQTL) analysis using the BrainSeq Phase I dataset to dissect schizophrenia-associated regulatory dynamics. Comparative eQTL analysis revealed widespread loss and gain of regulatory associations in schizophrenia group versus controls, alongside consistent eQTLs. A notable target gene switching phenomenon emerged, where specific SNPs regulated distinct genes across disease states, indicative of genetic pleiotropy mediated by competition for shared regulatory elements. Pleiotropic SNPs exhibited stronger schizophrenia associations, localized farther from target genes, and were enriched in repressive chromatin domains marked by H3K27me3. Transcription factor binding site analysis implicated EZH2, a polycomb repressive complex component, in mediating these regulatory shifts. Integration of schizophrenia-specific eQTLs with GWAS data via mendelian randomization prioritized risk genes like ANKRD45, which showed disease-context regulation and links to behavioral deficits in knockout models. Overall, we found the universality of eQTL specificity, and revealed a new mechanism that multiple genes competing for the shared regulatory sites, leading to phenotype-dependent gene expression shifts. This study establish context-specific eQTL dynamics as a critical layer of schizophrenia's genetic architecture. These insights advance functional interpretation of non-coding risk variants and provide new insights into regulatory mechanisms contributing to disease susceptibility. Molecular Genetics Expression quantitative trait loci Mendelian randomization Genome-wide association study Transcriptional Regulation Dynamics Polycomb repressive complex H3K27me3 Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Schizophrenia is a complex polygenic disorder with an estimated heritability of up to 80%. 1 Genome-wide association studies (GWAS) have identified numerous susceptibility loci associated with schizophrenia, 2 – 4 with recent research pinpointing 287 risk loci. 5 Notably, most of schizophrenia-associated genetic variants reside in non-coding regions and are significantly enriched in regulatory elements like enhancers. This underscores the importance of integrating expression quantitative trait loci (eQTL) analysis to elucidate the molecular mechanisms by which these variants contribute to schizophrenia pathogenesis. The regulatory effects of genetic variants on gene expression are highly heterogeneous. For instance, comparative eQTL studies across different ethnic populations have revealed substantial genetic heterogeneity, largely attributable to differences in allele frequencies. 6 Additionally, gene-environment interactions play a crucial role, account for approximately 35% of phenotypic variation. 7 eQTLs effects are also context-dependent, exhibiting tissue and cell specificity, 8 , 9 can be influenced by non-genetic factors such as age, smoking, alcohol consumption, infections, cannabis use, and psychosocial stress. 10 – 13 Furthermore, the complexity of genetic variation in regulatory functions is further demonstrated through their pleiotropy, where a single genetic variation can regulate the expression of multiple genes. 14 , 15 Moreover, genetic pleiotropy complicates the identification of functional genes mediating disease susceptibility. Despite these insights, context-specific studies on brain tissue eQTLs and their pleiotropic effects in schizophrenia remain limited. In this study, we aim to investigate context-specific eQTLs and their impact on schizophrenia susceptibility. Using BrainSeq project Phase I dataset, 16 which includes genotype, gene expression, and sample metadata, we stratified individuals based on phenotype, age, and sex to conduct comprehensive eQTL analyses. We then examine differences in eQTLs and their pleiotropic effects across these contexts, focusing on schizophrenia-specific eQTLs and their transcriptional factor driving these regulatory changes. Additionally, we employed summary-data-based mendelian randomization (SMR), 17 , 18 to integrate context-specific eQTLs with PGC3-SZ GWAS data to identify risk genes mediating schizophrenia susceptibility. Through this investigation, we aim to address two key questions: (1) What are the underlying drivers of context-specific of eQTLs? and (2) How do these findings enhance our understanding of schizophrenia risk and pathogenesis? Materials and methods Data and Sample Description This study utilized data from the BrainSeq Project Phase I dataset. 16 Genotype data (SNP arrays) were obtained from dbGaP (phs000448.v1.p1), and RNA sequencing data was downloaded from Synapse (syn12299750). After initial filtering, we identified 456 samples with both genotyping and transcriptomic data. To reduce age-related confounding in eQTL analysis, embryonic samples were excluded, resulting in a final dataset of 414 postnatal samples. Context-Specific cis-eQTL Mapping SNP array data were initially processed using GenomeStudio 2.0. To expand the number of available variants for eQTL mapping, we performed genotyping imputation using IMPUTE2 19 referencing the 1000 Genomes Project dataset. 20 Imputed results were filtered using an INFO score threshold of 0.8 to ensure quality, and converted to PLINK 21 format via gtool. This yielded genotype data for approximately 8.12 million SNPs. For transcriptomic data, RNA-seq reads were aligned to the hg19 reference genome 22 , 23 using STAR. 24 Gene-level quantification was performed using featureCounts, 25 and expression values were normalized as FPKM. Samples were stratified based on phenotype (schizophrenia vs. control), sex, chronological age (cutoff: 45 years), and epigenetic age (determined in our previous study, 26 Fig. 1 A and Supplementary Fig. 1 ). We conducted cis-eQTL analysis using MatrixEQTL, 27 considering SNPs within 1 Mb of each gene. Covariates included sex, age, ethnicity, and phenotype. The parameter pvOutputThreshold was set to 1 to retain all cis-eQTL results for downstream analyses. Enrichment Analysis Since the BrainSeq Phase I dataset derives from the Dorsolateral Prefrontal Cortex (DLPFC), we obtained chromatin state annotations for the adult DLPFC (Roadmap Epigenomics sample E073) via ChromHMM. 28 , 29 Due to limited ChIP-seq dataset in the DLPFC, we utilized transcription factor binding data from ENCODE 30 SK-N-SH neuroblastoma cell line, which includes 65 ChIP-seq datasets for corresponding to 62 transcription factors. To avoid enrichment bias caused by linkage disequilibrium, we used GREGOR 31 to asses of eQTLs in transcription factor binding sites and chromatin statte. GREGOR generated linkage maps based on 1000 Genomes Project data, controlling for LD structure among SNPs. Summary-data-based Mendelian Randomization (SMR) To assess causal relationships between gene expression and schizophrenia risk, we applied the SMR method. 17 , 18 This integrative approach uses SNPs as instrumental variables, combining phenotype-specific cis-eQTLs from schizophrenia and control groups with PGC3-SZ GWAS summary statistics. 5 Only genes with at least one significant eQTL ( P < 1E-5) were considered, corresponding to an estimated false discovery rate (FDR) < 0.05. To account for multiple independent SNPs per gene, we used multi-SMR, which aggregates signals across all qualifying SNPs in a gene region to infer its putative effect on schizophrenia risk. Data Availability All data generated in this study are available. The context-specific eQTL summary data is available on Zenodo, with the identifier 10.5281/zenodo.14333896 . Results Identifying Context-Specific eQTLs To examine eQTL patterns under different contextual conditions, we analyzed 414 samples from the BrainSeq Project Phase I dataset, 16 integrating genomic (genetic variation) and transcriptomic (gene expression) data (Fig. 1 A and Supplementary Fig. 1) . Samples were categorized based on sex, disease status, age, and epigenetic age acceleration, 26 forming nine distinct groups (including an all-sample group). A genome-wide cis-eQTL analysis (1Mb window) was conducted using MatrixEQTL. 27 For each group, we applied a false discovery rate (FDR) adjustment to the eQTL P -values, setting a significance threshold of FDR < 0.05 for significant eQTL pairs. The corresponding P -value threshold at this FDR level was recorded, revealing a strong correlation with sample size (Fig. 1 B). Notably, schizophrenia patients (SCZ) samples exhibited a higher-than-expected P -value threshold, deviating from the regression curve. This suggests a distribution shift towards significance, implying schizophrenia-specific regulatory loci. We quantified the number of significant eQTL pairs, associated genes (eGene), and SNPs (eSNP), as well as the average number of eSNP per eGene. While the number of eQTLs and eSNPs in the SCZ group significantly deviated from the expected trend, the number of eGenes did not (Fig. 1 C, Supplementary Fig. 2 ). This suggests that the schizophrenia involves an unexpectedly high number of regulatory variants influencing gene expression. Focused on the SCZ group, we compared eQTL distribution with the control (Con) group, identifying numerous genome-wide eQTL pairs without clear chromosomal bias (Fig. 1 D). However, eQTL distributions varied across chromosomes between the two groups (Fig. 1 E-F and Supplementary Fig. 3 ). For example, on chromosome 1 eQTLs in SCZ samples exhibited greater significance than in the Con group (Fig. 1 E), whereas chromosome 22 exhibited more significant eQTLs in the Con group (Fig. 1 F). The top SCZ-specific eQTL on chromosome 1 was rs138513841, associated with the NKAIN1 , while the leading eQTL on chromosome 22 was rs75493558, linked to PARVG . Both showed reduced significance in the control group (Fig. 1 G-H). NKAIN1 has been implicated in autism and alcohol dependence, while PARVG is linked to Alzheimer's disease and alcohol dependence 32 , 33 . These findings suggest that SCZ-specific eQTLs may contribute to broader neuropsychiatric disorder mechanisms. Widespread Loss, Gain, and Consistent eQTLs in Schizophrenia To systematically assess similarities and differences in eQTL patterns between the Con and SCZ groups, we employed the upsetR package 34 to visualize intersection counts of eQTL pairs across all samples (ALL), Con samples, and SCZ samples. The Con group contained 229,392 eQTL pairs ( FDR 0.05; Fig. 2 A). Analysis of the P -value distribution for these eQTLs in the SCZ group (Fig. 2 B) showed that 51.34% had P > 0.05, indicating a widespread loss of eQTLs in schizophrenia. Conversely, the SCZ group contained 168,806 significant eQTLs ( FDR 0.05). Analysis of the P -value distribution for these SCZ-Gain eQTLs in the Con group (Fig. 2 C) revealed that 53.4% had P > 0.05, suggesting a substantial gain of eQTLs in schizophrenia. After filtering out eSNPs targeting any gene in the SCZ group with P > 0.05, we identified 54,146 eSNPs corresponding to 7,605 eGenes and 61,326 SCZ-Loss eQTL pairs ( Supplementary Table 1 ). Similarly, after excluding eSNPs targeting any gene in the Con group with P < 0.05, we identified 34,331 eSNPs corresponding to 5,142 eGenes and 37,907 SCZ-Gain eQTL pairs ( Supplementary Table 2 ). Additionally, 83,656 eQTL pairs were significant in both the Con and SCZ groups ( Supplementary Table 3 ). Comparing effect sizes of these shared eQTL pairs revealed that their directional effect remained consistent. When one group was fixed at FDR < 0.05 and the other was relaxed to P < 0.05, only a small subset of eSNPs displayed opposing effect directions. Notably, as P -values became more significant, the number of opposing cases declined, indicating that opposing effect sizes are relatively rare ( Supplementary Fig. 4 ). Genes Competing for Regulatory Elements Lead to Genetic Pleiotropy in Genetic Regulation To investigate the impact of genetic pleiotropy on gene expression regulation in schizophrenia, we analyzed the overlap of eSNP between Con-Specific eQTLs (Con FDR 0.05) and SCZ-Specific eQTLs (Con FDR > 0.05 & SCZ FDR < 0.05). We identified 6,821 overlapping eSNPs, which target different genes in the Con and SCZ groups (Fig. 2 D). For clarity, we denote genes significantly targeted by eSNPs in the Con group ( FDR < 0.05) as Gene A and those significantly targeted in the SCZ group ( FDR 0.05 (Fig. 2 E, eSNP-Gene A-Gene B Pair N = 2,583, OR = 2.94, P = 2.213E-145). After excluding eSNPs with at least one eQTL P 0.05 in either group, we identified 1,144 eSNPs corresponding to 265 eGenes, forming 351 eGene-eGene pairs and 1,358 eQTL-eQTL pairs ( Supplementary Table 4 ). This subset, indicative of eGene switch, exhibited a strong tendency for consistent effect size, with 1,218 eQTL-eQTL pairs (89.68%) maintaining the same directional effect (Fig. 2 F). In contrast, only 55.39% of other eQTL-eQTL pairs showed consistent effects, highlighting a stronger consistency in the eGene switch subset ( OR = 7.588, P = 4.725E-161). Expression correlation analyses within the eGene switch subset revealed 163 negatively correlated gene pairs (46.44%), significantly higher than the 176 negatively correlated pairs (25.69%) observed in other gene sets ( OR = 2.50, P = 3.543E-11, Fig. 2 G). This suggests that genes undergoing eGene switch are more likely to exhibit inverse expression relationships. In summary, within the eGene switch subset, SNP effects targeting different genes align in the same direction, while their gene expression correlations are inversely related. This suggests competition for regulatory loci, leading to genetic pleiotropy in gene expression regulation across different contexts (control and schizophrenia). Notably, the strongest negatively correlated gene pair identified was regulated by SNP rs77548189, which influences ZNF268 expression in the Con group (Fig. 2 H) and GALNT9 expression in the SCZ group (Fig. 2 I). The competition for rs77548189 between ZNF268 and GALNT9 leads to a negative correlation in their expression levels (Fig. 2 J). GALNT9 has been associated with autism, 35 while ZNF268 has not been linked to neuropsychiatric disorders, suggesting that genetic pleiotropy in gene expression regulation plays a multifaceted role in neuropsychiatric disorder development. Repressive Histone Mark Induced by EZH2 is involved in Genetic Pleiotropy Associated with Schizophrenia We categorized eSNPs corresponding to eQTLs into Loss, Gain, Consistent, and Switch groups, further classifying Switch eQTLs into Negative and Positive correlations based on gene expression relationships ( Supplementary Table 1–4 ). To explore the genetic characteristics for these categories, we analyzed their associations with PGC3-SZ GWAS 5 beta values, P -values, minor allele frequencies (MAF), and distances between eSNPs and their respective eGenes. Notably, eSNPs in the Negative Switch category exhibited the most significant P -values, followed by those in the Positive Switch, Consistent, Gain, and Loss categories. (Fig. 3 A). Similarly, Negative Switch eSNPs had the highest GWAS beta values, followed by those in the Positive Switch category (Fig. 3 B). MAF analysis revealed that eSNPs in the Loss, Gain, and Switch Category had MAF around 0.1, whereas Consistent eSNPs had higher frequencies (0.3) (Fig. 3 C). Additionally, eSNPs in the Negative Switch Category were positioned farther from their target genes, with a median distance approaching 500 kb (Fig. 3 D). These findings suggest that Switch eQTLs, particularly those in the Negative Switch category, involve variants with low MAF and long-distance regulatory effects and strong associations with schizophrenia. To investigate potential epigenetic mechanism, we performed chromatin state enrichment analysis using ChromHMM 28 annotations from the Roadmap Epigenomics project. 29 This revealed a significant enrichment of Negative Switch eSNPs in repressive chromatin regions marked by the ReprPC state (Fig. 3 E), implicating repressive histone modification in genetic pleiotropy. Further analysis of 65 transcriptional factor (TF) binding sites in SK-N-SH ChIP-Seq data 30 showed that Negative Switch eSNPs category were highly enriched in these regulatory regions. Notably, the most enriched TFs were EZH2 (phospho-T487) and EP300 (Fig. 3 F), suggesting that repressive histone modifications, such as H3K27me3 induced by EZH2, may contribute to the schizophrenia-associated genetic pleiotropy. Phenotype-Specific eQTL Contributed to Identifying Schizophrenia Risk Genes Using SMR Analysis To further assess the role of schizophrenia-specific eQTL in disease risk, we conducted summary data-based Mendelian randomization (SMR) analysis. 17 , 18 This approach integrates eQTL data with PGC3-SZ GWAS data to identify gene whose expression levels may be causally linked to schizophrenia. We analyzed 2,854 genes using eQTL data from control individuals ( Supplementary Table 6 ) and 2,187 genes using eQTL data from schizophrenia patients ( Supplementary Table 7 ), with 969 overlapped genes between two groups. To evaluate the impact of genetic pleiotropy on SMR analysis, we extracted the top SNPs (i.e., SNPs with the most significant eQTL P -value) corresponding to these 969 overlapping genes. We identified 144 shared top SNPs (Fig. 4 A), which regulated 133 overlapping genes, 15 control-specific genes, and 28 schizophrenia-specific genes (Fig. 4 B). This resulted in 33 unique gene-gene pairs further supporting the notion that eSNPs regulate different genes in the control and schizophrenia groups (Fig. 4 C, Supplementary Table 8 ). Applying an FDR threshold < 0.1, we identified a key SNP, rs16846667, located on chromosome 1 (Fig. 4 D). This variant regulates SERPINC1 in the control group (Fig. 4 E) but ANKRD45 in the schizophrenia group (Fig. 4 F-G). ANKRD45 belongs to the ankyrin repeat family, and knockout mice exhibit abnormal behaviors in open field test, including significantly reduced resting time 36 , 37 (Fig. 4 H). In contrast, there is no known association between SERPINC1 and nervous system function. These findings suggest that rs16846667 may contribute to schizophrenia susceptibility by modulating ANKRD45 expression in affected individuals. Discussion A key finding of this study is the universality of eQTL specificity, particularly among individuals with schizophrenia, where the significance of eQTLs exceeded expectations. This suggests that genetic variation may have distinct regulatory functions within schizophrenia patients. While eQTLs are known to exhibit tissue and cell specificity due to differences in TF expression and cellular environments, 8 , 38 – 40 disease specificity remains underexplored. Recent studies on non-alcoholic fatty liver disease (NAFLD) 41 and idiopathic pulmonary fibrosis (IPF) 42 have identified disease-specific eQTLs that influence gene expression and disease progression, supporting the relevance of our findings. Notably, we identified a large number of specific eQTLs in schizophrenia, with over 50% of eQTLs in both control and schizophrenia groups not reaching significance in the other group, highlighting the need for focused investigations into schizophrenia-specific eQTLs. Another major contribution of this study is the exploration of genetic pleiotropy in gene regulation and its role in schizophrenia susceptibility. Traditionally, pleiotropy refers to a single genetic variant influencing multiple genes through transcriptional regulatory complexes involving enhancers and promoters. 43 , 44 However, our findings suggest a different mechanism—multiple genes competing for the same regulatory sites, leading to phenotype-dependent gene expression shifts. This pattern parallels hemoglobin switching, where locus control regions in the β-globin cluster regulate gene expression during development. 45 – 47 Furthermore, our analysis revealed that schizophrenia-associated pleiotropic genetic variants are significantly enriched in EZH2 binding sites, a key component of the PRC2 complex. PRC2, through H3K27me3-mediated chromatin modifications, plays a crucial role in embryonic development, cell differentiation, cell fate determination, and gene silencing. 48 – 50 Dysregulation of PRC2 has been implicated in neurodevelopmental abnormalities and neurotransmitter metabolism. 48 , 51 , 52 Our previous work also demonstrated that PRC2’s role in the allele-specific methylation changes in EIPR1 in monozygotic twins discordant for schizophrenia. 53 Collectively, our findings suggest that PRC2-mediated regulatory shifts contribute to schizophrenia susceptibility by altering gene expression in context-dependent manner. To functionally interpret these regulatory changes, we integrate phenotype-specific eQTL data with PGC3-SZ GWAS data using the SMR tool. This analysis identified schizophrenia-associated variant rs16846667 on chromosome 1, which regulates SERPINC1 in control individuals but ANKRD45 in schizophrenia patients. Notably, Ankrd45 knockout mice exhibited behavioral abnormalities, whereas SERPINC1 has no known neurological associations. This highlights the utility of phenotype-specific eQTLs in pinpointing genes mediating the functional effects of genetic variation on disease risk. As a resource for future research, we provide eQTL data stratified by age, sex, phenotype, and epigenetic age, creating eight distinct groups. This data can be further analyzed using methods such as SMR, coloc, 54 , 55 and TWAS 56 , 57 to explore the influence of genetic regulation across different biological contexts. Despite these insights, our study has limitations. The BrainSeq Phase I dataset includes both European and African American individuals, which may introduce population-specific biases. Given the need for subgroup analyses based on age, sex, and phenotype, restricting our sample to a single population would have significantly reduced statistical power. However, previous research has shown high consistency in eQTL results across these populations, 6 and we account for population as a covariate in our analyses to mitigate potential biases. In conclusion, our study establishes the context-specificity of eQTLs in schizophrenia and demonstrates that genetic pleiotropy arises from multiple genes competing for shared regulatory sites. The PRC2 complex, particularly EZH2, plays a crucial role in mediating these shifts in gene regulation between control and schizophrenia groups. These findings provide new insights into the molecular mechanisms underlying schizophrenia susceptibility and emphasize the importance of investigating phenotype-specific eQTLs. By making our context-specific eQTL dataset publicly available, we offer a valuable resource for future studies delving deeper into the molecular mechanisms of neuropsychiatric disorders. Declarations Acknowledgements Author contributions: Conceived and designed the analysis: ZW, LY, and ZS; Provided the rawdata of BrainSeq project: CZ; Analyzed the data: ZW, LY, XW, JY, and ZL; Wrote the paper: ZW, LY, CZ, QY, SC, and FX. Funding This work was supported by a grant of the National Natural Science Foundation of China [grant numbers 82101577]; China Postdoctoral Science Foundation [grant numbers 2023M730724, 2020M682806]. Competing interests The authors report no competing interests. Supplementary material Supplementary material is available online. Author contributions: Conceived and designed the analysis: ZW, LY, and ZS; Provided the rawdata of BrainSeq project: CZ; Analyzed the data: ZW, LY, XW, JY, and ZL; Wrote the paper: ZW, LY, CZ, QY, SC, and FX. References Hilker R, Helenius D, Fagerlund B et al (2018) Heritability of Schizophrenia and Schizophrenia Spectrum Based on the Nationwide Danish Twin Register. Biological psychiatry . 15(6):492–498. 10.1016/j.biopsych.2017.08.017 Biological (2014) insights from 108 schizophrenia-associated genetic loci. Nature . Jul 24 ;511(7510):421-7. 10.1038/nature13595 Lam M, Chen CY, Li Z et al (2019) Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet Dec 51(12):1670–1678. 10.1038/s41588-019-0512-x Pardiñas AF, Holmans P, Pocklington AJ et al (2018) Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection. Nat Genet Mar 50(3):381–389. 10.1038/s41588-018-0059-2 Trubetskoy V, Pardiñas AF, Qi T et al (2022) Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nat Apr 604(7906):502–508. 10.1038/s41586-022-04434-5 Chen Y, Liu S, Ren Z et al (2024) Cross-ancestry analysis of brain QTLs enhances interpretation of schizophrenia genome-wide association studies. Am J Hum Genet Nov 7(11):2444–2457. 10.1016/j.ajhg.2024.09.001 Zhang H, Khan A, Kushner SA, Rzhetsky A (2022) Dissecting schizophrenia phenotypic variation: the contribution of genetic variation, environmental exposures, and gene-environment interactions. Schizophrenia (Heidelberg Germany) May 10(1):51. 10.1038/s41537-022-00257-5 Zhernakova DV, Deelen P, Vermaat M et al (2017) Identification of context-dependent expression quantitative trait loci in whole blood. Nat Genet Jan 49(1):139–145. 10.1038/ng.3737 Findley AS, Monziani A, Richards AL et al (2021) Functional dynamic genetic effects on gene regulation are specific to particular cell types and environmental conditions. eLife May 14(10). 10.7554/eLife.67077 Di Forti M, Marconi A, Carra E et al (2015) Proportion of patients in south London with first-episode psychosis attributable to use of high potency cannabis: a case-control study. lancet Psychiatry Mar 2(3):233–238. 10.1016/s2215-0366(14)00117-5 Hjorthøj C, Compton W, Starzer M et al (2023) Association between cannabis use disorder and schizophrenia stronger in young males than in females. Psychol Med Nov 53(15):7322–7328. 10.1017/s0033291723000880 Lupien SJ, McEwen BS, Gunnar MR, Heim C (2009) Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat reviews Neurosci Jun 10(6):434–445. 10.1038/nrn2639 Benros ME, Trabjerg BB, Meier S et al (2016) Influence of Polygenic Risk Scores on the Association Between Infections and Schizophrenia. Biol psychiatry Oct 15(8):609–616. 10.1016/j.biopsych.2016.04.008 Delaneau O, Zazhytska M, Borel C et al (2019) Chromatin three-dimensional interactions mediate genetic effects on gene expression. Science (New York, NY) . May 3. ;364(6439) 10.1126/science.aat8266 Mumbach MR, Satpathy AT, Boyle EA et al (2017) Enhancer connectome in primary human cells identifies target genes of disease-associated DNA elements. Nat Genet Nov 49(11):1602–1612. 10.1038/ng.3963 Jaffe AE, Straub RE, Shin JH et al (2018) Developmental and genetic regulation of the human cortex transcriptome illuminate schizophrenia pathogenesis. Nat Neurosci Aug 21(8):1117–1125. 10.1038/s41593-018-0197-y Zhu Z, Zhang F, Hu H et al (2016) Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat Genet May 48(5):481–487. 10.1038/ng.3538 Wu Y, Zeng J, Zhang F et al (2018) Integrative analysis of omics summary data reveals putative mechanisms underlying complex traits. Nat Commun Mar 2(1):918. 10.1038/s41467-018-03371-0 Howie BN, Donnelly P, Marchini J (2009) A flexible and accurate genotype imputation method for the next generation of genome-wide association studies. PLoS Genet Jun 5(6):e1000529. 10.1371/journal.pgen.1000529 Abecasis GR, Altshuler D, Auton A et al (2010) A map of human genome variation from population-scale sequencing. Nat Oct 28(7319):1061–1073. 10.1038/nature09534 Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. GigaScience 4:7. 10.1186/s13742-015-0047-8 Mudge JM, Carbonell-Sala S, Diekhans M et al (2025) GENCODE 2025: reference gene annotation for human and mouse. Nucleic acids Res Jan 6(D1):D966–d975. 10.1093/nar/gkae1078 Frankish A, Diekhans M, Ferreira AM et al (2019) GENCODE reference annotation for the human and mouse genomes. Nucleic acids Res Jan 8(D1):D766–d773. 10.1093/nar/gky955 Dobin A, Davis CA, Schlesinger F et al (2013) STAR: ultrafast universal RNA-seq aligner. Bioinformatics (Oxford, England) . Jan 1. ;29(1):15–21. 10.1093/bioinformatics/bts635 Liao Y, Smyth GK, Shi W (2013) The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic acids Res May 1(10):e108. 10.1093/nar/gkt214 Wu X, Ye J, Wang Z, Zhao C (2021) Epigenetic Age Acceleration Was Delayed in Schizophrenia. Schizophr Bull Apr 29(3):803–811. 10.1093/schbul/sbaa164 Shabalin AA (2012) Matrix eQTL: ultra fast eQTL analysis via large matrix operations. Bioinf (Oxford England) May 15(10):1353–1358. 10.1093/bioinformatics/bts163 Ernst J, Kellis M (2012) ChromHMM: automating chromatin-state discovery and characterization. Nat methods Feb 28(3):215–216. 10.1038/nmeth.1906 Roadmap Epigenomics C, Kundaje A, Meuleman W et al (2015) Integrative analysis of 111 reference human epigenomes. Nat Feb 19(7539):317–330. 10.1038/nature14248 An integrated encyclopedia (2012) of DNA elements in the human genome. Nat Sep 6 489(7414):57–74. 10.1038/nature11247 Schmidt EM, Zhang J, Zhou W et al (2015) GREGOR: evaluating global enrichment of trait-associated variants in epigenomic features using a systematic, data-driven approach. Bioinformatics (Oxford, England) . Aug 15. ;31(16):2601-6. 10.1093/bioinformatics/btv201 Rappaport N, Nativ N, Stelzer G et al (2013) MalaCards: an integrated compendium for diseases and their annotation. Database: J Biol databases curation 2013:bat018. 10.1093/database/bat018 Rappaport N, Twik M, Plaschkes I et al (2017) MalaCards: an amalgamated human disease compendium with diverse clinical and genetic annotation and structured search. Nucleic acids Res Jan 4(D1):D877–d887. 10.1093/nar/gkw1012 Lex A, Gehlenborg N, Strobelt H, Vuillemot R, Pfister H, UpSet (2014) Visualization of Intersecting Sets. IEEE Trans visualization Comput graphics Dec 20(12):1983–1992. 10.1109/tvcg.2014.2346248 Stelzer G, Rosen N, Plaschkes I et al (2016) The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr protocols Bioinf Jun 20(1301–). 10.1002/cpbi.5 Groza T, Gomez FL, Mashhadi HH et al (2023) The International Mouse Phenotyping Consortium: comprehensive knockout phenotyping underpinning the study of human disease. Nucleic acids Res Jan 6(D1):D1038–d1045. 10.1093/nar/gkac972 Dickinson ME, Flenniken AM, Ji X et al (2016) High-throughput discovery of novel developmental phenotypes. Nat Sep 22(7621):508–514. 10.1038/nature19356 Consortium GT, Laboratory DA, Coordinating Center -Analysis, Working G et al (2017) Genetic effects on gene expression across human tissues. Nat Oct 11(7675):204–213. 10.1038/nature24277 Kim-Hellmuth S, Aguet F, Oliva M et al (2020) Cell type-specific genetic regulation of gene expression across human tissues. Science (New York, NY) . Sep 11. ;369(6509) 10.1126/science.aaz8528 Lappalainen T, Sammeth M, Friedländer MR et al (2013) Transcriptome and genome sequencing uncovers functional variation in humans. Nat Sep 26(7468):506–511. 10.1038/nature12531 Yoo T, Joo SK, Kim HJ et al (2021) Disease-specific eQTL screening reveals an anti-fibrotic effect of AGXT2 in non-alcoholic fatty liver disease. J Hepatol Sep 75(3):514–523. 10.1016/j.jhep.2021.04.011 Dalapati T, Wang L, Jones AG et al (2025) Context-specific eQTLs provide deeper insight into causal genes underlying shared genetic architecture of COVID-19 and idiopathic pulmonary fibrosis. HGG Adv Jan 27(2):100410. 10.1016/j.xhgg.2025.100410 Kim TK, Shiekhattar R (2015) Architectural and Functional Commonalities between Enhancers and Promoters. Cell Aug 27(5):948–959. 10.1016/j.cell.2015.08.008 Levo M, Raimundo J, Bing XY et al (2022) Transcriptional coupling of distant regulatory genes in living embryos. Nat May 605(7911):754–760. 10.1038/s41586-022-04680-7 Karlsson S, Nienhuis AW (1985) Developmental regulation of human globin genes. Annu Rev Biochem 54:1071–1108. 10.1146/annurev.bi.54.070185.005231 Stamatoyannopoulos G (1991) Human hemoglobin switching. Science (New York, NY) . Apr 19. ;252(5004):383. 10.1126/science.2017679 Kim S, Kim YW, Shim SH, Kim CG, Kim A (2012) Chromatin structure of the LCR in the human β-globin locus transcribing the adult δ- and β-globin genes. Int J Biochem cell biology Mar 44(3):505–513. 10.1016/j.biocel.2011.12.001 Margueron R, Reinberg D (2011) The Polycomb complex PRC2 and its mark in life. Nat Jan 20(7330):343–349. 10.1038/nature09784 Greer EL, Shi Y (2012) Histone methylation: a dynamic mark in health, disease and inheritance. Nature reviews Genetics . Apr 3. ;13(5):343 – 57. 10.1038/nrg3173 Majewski IJ, Blewitt ME, de Graaf CA et al (2008) Polycomb repressive complex 2 (PRC2) restricts hematopoietic stem cell activity. PLoS biology Apr 15(4):e93. 10.1371/journal.pbio.0060093 Wang YY, Deng YS, Dai SK et al (2022) Loss of microglial EED impairs synapse density, learning, and memory. Mol psychiatry Jul 27(7):2999–3009. 10.1038/s41380-022-01576-w Schröder S, Fuchs U, Gisa V et al (2024) PRDM16-DT is a novel lncRNA that regulates astrocyte function in Alzheimer's disease. Acta Neuropathol Aug 29(1):32. 10.1007/s00401-024-02787-x Li Q, Wang Z, Zong L et al (2021) Allele-specific DNA methylation maps in monozygotic twins discordant for psychiatric disorders reveal that disease-associated switching at the EIPR1 regulatory loci modulates neural function. Mol Psychiatry Nov 26 26(11):6630–6642. 10.1038/s41380-021-01126-w Giambartolomei C, Vukcevic D, Schadt EE et al (2014) Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet May 10(5):e1004383. 10.1371/journal.pgen.1004383 Wallace C (2021) A more accurate method for colocalisation analysis allowing for multiple causal variants. PLoS Genet Sep 17(9):e1009440. 10.1371/journal.pgen.1009440 Gamazon ER, Wheeler HE, Shah KP et al (2015) A gene-based association method for mapping traits using reference transcriptome data. Nat Genet Sep 47(9):1091–1098. 10.1038/ng.3367 Mai J, Lu M, Gao Q, Zeng J, Xiao J (2023) Transcriptome-wide association studies: recent advances in methods, applications and available databases. Commun biology Sep 1(1):899. 10.1038/s42003-023-05279-y Additional Declarations The authors declare no competing interests. Supplementary Files Supplementarytables.xlsx Supplementary Tables Supplementaryfigures.docx Supplementary Figures 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-6582362","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":451589427,"identity":"ecc86cf7-3e4d-4e0f-ade2-53fc352144db","order_by":0,"name":"Linyan Ye","email":"","orcid":"","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":false,"prefix":"","firstName":"Linyan","middleName":"","lastName":"Ye","suffix":""},{"id":451591627,"identity":"bc9166a5-a5ee-4860-96ed-17f1ad8b8be0","order_by":1,"name":"Zongrui Shen","email":"","orcid":"","institution":"Gaozhou People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zongrui","middleName":"","lastName":"Shen","suffix":""},{"id":451591628,"identity":"a85fbf23-3d8a-477e-b5eb-7114d637aceb","order_by":2,"name":"Qi Yang","email":"","orcid":"","institution":"Department of Genetic and Metabolic Central Laboratory, Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Yang","suffix":""},{"id":451591629,"identity":"fb64dac8-ff31-4d44-86a8-0d4d9fdd8c44","order_by":3,"name":"Xiaohui Wu","email":"","orcid":"","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":false,"prefix":"","firstName":"Xiaohui","middleName":"","lastName":"Wu","suffix":""},{"id":451591630,"identity":"b52a59c8-5701-4dc0-ae60-9501348a7111","order_by":4,"name":"JunPing Ye","email":"","orcid":"","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":false,"prefix":"","firstName":"JunPing","middleName":"","lastName":"Ye","suffix":""},{"id":451591631,"identity":"efaa3f2b-dd5d-4fe5-8464-6502c30debf9","order_by":5,"name":"Zhongwei Li","email":"","orcid":"","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":false,"prefix":"","firstName":"Zhongwei","middleName":"","lastName":"Li","suffix":""},{"id":451591632,"identity":"75634a02-f80f-4fa4-9e61-9eaf27c6fd23","order_by":6,"name":"Fu Xiong","email":"","orcid":"https://orcid.org/0000-0003-2428-6165","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":false,"prefix":"","firstName":"Fu","middleName":"","lastName":"Xiong","suffix":""},{"id":451591633,"identity":"10344a8e-3f68-4aed-bd30-65329109eaa9","order_by":7,"name":"Siyao Che","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7ElEQVRIiWNgGAWjYDACCSBOAGJ+ZsbmBx8MbOyI1yLZ3txmOKMgLZk4LSBgcOZ4gzTPh0OMDYR08M9uPvzhYZudHMONxAZjG4MDzAzsh49uwGvJnWNpEoltycaMMxIbHucY3OFj4ElLu4FPi4FEjhlDYhtzYrME0JYcg2fMDBI8ZgS05H/+kNhWn9gG1CJtYXCYsYGwlhwGoMMOJ/bwHGyQZiBGi8SNNDOJhHPHjSXYG9sMewzSktkI+YV/RvLjjz/KquXsD7M/fvDjj40dP/vhY3i1gAEjGxKHDacyFPCHOGWjYBSMglEwQgEAkGBMiHZW7HwAAAAASUVORK5CYII=","orcid":"","institution":"Gaozhou People's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Siyao","middleName":"","lastName":"Che","suffix":""},{"id":451591634,"identity":"c95033b1-0260-4874-8c9a-9bd3c3378a0b","order_by":8,"name":"Cunyou Zhao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwElEQVRIiWNgGAWjYHACNiC2gTB5SNCSRrqWwyRo4Z+R/uzBj4rz9rozEhgfvG1jkDcnpEXiRkK6Yc+Z24nbbiQwG85tYzDc2UBAi4FEwjEJ3rbbCWY3EtikedsYEgwOENSS2Cb5t+2cPVAL+28itSSDDD/ACHQYGzNRWiTOPGOTljmTnLjtzMNmyTnnJAw3ENLC357+TPJNhZ292fHkgx/elNnIE7SFQSABxmJsANlKSD3IGoKGjoJRMApGwYgHAMMCPf9KFD5LAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-6116-4584","institution":"Department of Medical Genetics, School of Basic Medical Sciences, Southern Medical University","correspondingAuthor":true,"prefix":"","firstName":"Cunyou","middleName":"","lastName":"Zhao","suffix":""},{"id":451591635,"identity":"956d7582-e662-454c-b20a-7e7a10103545","order_by":9,"name":"Zhongju Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBElEQVRIiWNgGAWjYDACCRDBA2YyP0j8Y8PDz99AvBY2g4cNaTKSMw4QowUKJB82HLYxaEjAr0N+dvOxh19kDif2z26/YJC44zyPAcMBxg8fc3BrYZxzLN1Yhudw4ow7ZwoeJJ65zWPO3MAsOXMbbi3MEjlm0hI8h3MbbuQkGCSw3eaxbDjAxsyLRwsbTMt8oBaJBLZzPAYHEvBr4QFqkfwA1LLhRvoBicS2A4S1SEikpUkz8KTXb7yRw2aQcCaZR3LGwWa8fpGfkXxM8mePtbHcjfTHD39U2Nnz8zcf/PARjxZwEPD2gN1oAOUzNuBXD1Ly4weIYn9AUOUoGAWjYBSMTAAAzZVWkG2H4AgAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-0098-6100","institution":"Gaozhou People's Hospital","correspondingAuthor":true,"prefix":"","firstName":"Zhongju","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2025-05-03 06:40:49","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-6582362/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6582362/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82047582,"identity":"5f46da8d-d0fd-46aa-9ab1-07b5ac75ccf8","added_by":"auto","created_at":"2025-05-06 09:42:40","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":620724,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eContext-Specific eQTL analysis.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e Schematic overview of context-specific eQTL mapping workflow, stratified by disease status, gender, and age. \u003cstrong\u003e(B, C)\u003c/strong\u003e Regression analyses showing the relationship between sample size and the \u003cem\u003eP\u003c/em\u003e-value threshold corresponding to \u003cem\u003eFDR\u003c/em\u003e=0.05 \u003cstrong\u003e(B)\u003c/strong\u003e, and the average number of eSNPs per gene \u003cstrong\u003e(C)\u003c/strong\u003e. \u003cstrong\u003e(D)\u003c/strong\u003e Circos plot of eQTL \u003cem\u003eP\u003c/em\u003e-values across the genome for all individuals (inner ring), controls (middle ring) and individuals with schizophrenia (outer ring). \u003cstrong\u003e(E, F)\u003c/strong\u003e Q-Q plots of eQTL \u003cem\u003eP\u003c/em\u003e-values on chromosomes chr1 \u003cstrong\u003e(E)\u003c/strong\u003e and 22 \u003cstrong\u003e(F)\u003c/strong\u003e, illustrating deviation from the expected null distribution. \u003cstrong\u003e(G, H)\u003c/strong\u003eRepresentative phenotype-specific eQTLs: rs138513841 associated with \u003cem\u003eNKAN1\u003c/em\u003eexpression \u003cstrong\u003e(G)\u003c/strong\u003e and rs75493558 associated with \u003cem\u003ePARVG\u003c/em\u003e expression \u003cstrong\u003e(H)\u003c/strong\u003e.\u003c/p\u003e","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/23163f9acbbe41bc87ed45f5.jpg"},{"id":82045365,"identity":"e94ccfd3-868a-460b-a41f-607dc5ca9428","added_by":"auto","created_at":"2025-05-06 09:34:41","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":692367,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparative analysis of eQTL between control and schizophrenia groups.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e UpsetR plot illustrating the overlap of significant eQTLs (\u003cem\u003eFDR\u003c/em\u003e\u0026lt;0.05) among the three groups: all individuals (ALL), controls (Con), and schizophrenia patients (SCZ). \u003cstrong\u003e(B, C)\u003c/strong\u003e Pie charts showing the significance of Con-specific eQTLs in the schizophrenia group \u003cstrong\u003e(B)\u003c/strong\u003eand vice versa \u003cstrong\u003e(C)\u003c/strong\u003e, highlighting differential regulatory effects. \u003cstrong\u003e(D)\u003c/strong\u003eOverlap of eSNPs regulating different target genes between Control (labeled as Gene A) and Schizophrenia (labeled as Gene B), representing a gene-switching phenomenon. \u003cstrong\u003e(E)\u003c/strong\u003e Distribution of \u003cem\u003eP\u003c/em\u003e-values for non-significant eQTLs from panel D, illustrating the statistical significance of Gene A eQTLs in the Schizophrenia group and Gene B eQTLs in the Control group. \u003cstrong\u003e(F)\u003c/strong\u003eProportion of overlapping eQTLs with consistent direction of effect sizes between the Control and Schizophrenia groups. \u003cstrong\u003e(G)\u003c/strong\u003e Distribution of Pearson correlation coefficients representing co-expression relationships Gene A and Gene B. \u003cstrong\u003e(H, I)\u003c/strong\u003e Representative examples of a phenotype-specific eQTL (rs77548189) showing regulation of \u003cem\u003eZNF268\u003c/em\u003e expression in controls \u003cstrong\u003e(H)\u003c/strong\u003eand \u003cem\u003eGALNT9\u003c/em\u003e expression in schizophrenia patients \u003cstrong\u003e(I)\u003c/strong\u003e. \u003cstrong\u003e(J) \u003c/strong\u003ePearson correlation analysis between \u003cem\u003eZNF268\u003c/em\u003e and \u003cem\u003eGALNT9\u003c/em\u003e expression levels.\u003c/p\u003e","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/70952d8efdfa9955b9d68ec4.jpg"},{"id":82045366,"identity":"77c9fdaa-8662-4799-a49c-85d27e7ad718","added_by":"auto","created_at":"2025-05-06 09:34:41","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":421559,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCharacterization of eSNPs across context-specific regulatory categories.\u003c/strong\u003e \u003cstrong\u003e(A-D)\u003c/strong\u003e Comparison of eSNP properties across five regulatory categories: Loss, Gain, Consistent, Negative Switch, and Positive Switch. These properties includes: -Log\u003csub\u003e10\u003c/sub\u003e(GWAS \u003cem\u003eP\u003c/em\u003e-values), indicating schizophrenia association strength \u003cstrong\u003e(A)\u003c/strong\u003e, Absolute values of GWAS effect sizes \u003cstrong\u003e(\u003c/strong\u003e|β|; \u003cstrong\u003eB)\u003c/strong\u003e, Minor allele frequency \u003cstrong\u003e(\u003c/strong\u003eMAF; \u003cstrong\u003eC)\u003c/strong\u003e, Distance between eSNPs and their respective eGenes \u003cstrong\u003e(D)\u003c/strong\u003e. \u003cstrong\u003e(E, F)\u003c/strong\u003e Enrichment of eSNPs in chromatin states based on ChromHMM annotations from the Roadmap Epigenomics Project (DLPFC, E073) \u003cstrong\u003e(E)\u003c/strong\u003e, and in transcription factor binding sites identified in the SK-N-SH neuroblastoma cell line (ENCODE dataset) \u003cstrong\u003e(F)\u003c/strong\u003e, shown as fold enrichment for each regulatory category.\u003c/p\u003e","description":"","filename":"Fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/476362f9763140117e0b8ec7.jpg"},{"id":82045367,"identity":"be996c27-9a06-492d-ad51-f4830e9c2260","added_by":"auto","created_at":"2025-05-06 09:34:41","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":536730,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhenotype-specific eQTLs aid in identifying risk genes for schizophrenia.\u003c/strong\u003e \u003cstrong\u003e(A-C)\u003c/strong\u003e Diagram illustrating the selection of genetic pleiotropy affecting the results of SMR analysis. \u003cstrong\u003e(A)\u003c/strong\u003e Intersection of top SNPs identified in the Control and Schizophrenia SMR results, \u003cstrong\u003e(B)\u003c/strong\u003eIntersection of genes corresponding to the 144 top SNPs, \u003cstrong\u003e(C)\u003c/strong\u003e Among these, 15 and 28 specific genes regulate Gene A in the Control group and Gene B in the Schizophrenia group, respectively. \u003cstrong\u003e(D-F)\u003c/strong\u003e \u003cem\u003eP\u003c/em\u003e-value distributions from the PGC3-SZ GWAS for the locus of top SNP rs16846667 \u003cstrong\u003e(D)\u003c/strong\u003e, the eQTL \u003cem\u003eP\u003c/em\u003e-value distribution for SERPINC1 in the Control group \u003cstrong\u003e(E)\u003c/strong\u003e, and the eQTL \u003cem\u003eP\u003c/em\u003e-value distribution for ANKRD45 in the Schizophrenia group \u003cstrong\u003e(F)\u003c/strong\u003e. \u003cstrong\u003e(G)\u003c/strong\u003e Boxplot demonstrating the phenotype specificity of eQTL showing association between rs16846667 and \u003cem\u003eANKRD45\u003c/em\u003e expression. \u003cstrong\u003e(H)\u003c/strong\u003eAnkrd45 knockout mice exhibits a reduced frequency of whole arena resting time in the open field test, as recorded in the IMPC database.\u003c/p\u003e","description":"","filename":"Fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/68202098afd98b5708ee31c8.jpg"},{"id":82049196,"identity":"f1a2d16c-2bab-4155-9047-f6d0386656d6","added_by":"auto","created_at":"2025-05-06 09:50:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3317762,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/a8ea177c-522d-434d-a2b8-8ede3c93d7c4.pdf"},{"id":82045372,"identity":"511044d7-f645-42ec-8286-b1de86d6dd08","added_by":"auto","created_at":"2025-05-06 09:34:42","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":29735651,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Tables\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Supplementarytables.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/1f4df870eb94a101edfccabb.xlsx"},{"id":82047583,"identity":"4c37fd2f-d2cc-4d70-934b-0ae90c666ed0","added_by":"auto","created_at":"2025-05-06 09:42:41","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3329574,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplementary Figures\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Supplementaryfigures.docx","url":"https://assets-eu.researchsquare.com/files/rs-6582362/v1/1e1f7c3b2218af2ac52be182.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eContext-specific eQTL dynamics uncover genetic pleiotropy and chromatin-mediated target gene switching in schizophrenia\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSchizophrenia is a complex polygenic disorder with an estimated heritability of up to 80%.\u003csup\u003e1\u003c/sup\u003e Genome-wide association studies (GWAS) have identified numerous susceptibility loci associated with schizophrenia,\u003csup\u003e\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e with recent research pinpointing 287 risk loci.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Notably, most of schizophrenia-associated genetic variants reside in non-coding regions and are significantly enriched in regulatory elements like enhancers. This underscores the importance of integrating expression quantitative trait loci (eQTL) analysis to elucidate the molecular mechanisms by which these variants contribute to schizophrenia pathogenesis.\u003c/p\u003e \u003cp\u003eThe regulatory effects of genetic variants on gene expression are highly heterogeneous. For instance, comparative eQTL studies across different ethnic populations have revealed substantial genetic heterogeneity, largely attributable to differences in allele frequencies.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e Additionally, gene-environment interactions play a crucial role, account for approximately 35% of phenotypic variation.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e eQTLs effects are also context-dependent, exhibiting tissue and cell specificity,\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e can be influenced by non-genetic factors such as age, smoking, alcohol consumption, infections, cannabis use, and psychosocial stress.\u003csup\u003e\u003cspan additionalcitationids=\"CR11 CR12\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Furthermore, the complexity of genetic variation in regulatory functions is further demonstrated through their pleiotropy, where a single genetic variation can regulate the expression of multiple genes.\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e Moreover, genetic pleiotropy complicates the identification of functional genes mediating disease susceptibility. Despite these insights, context-specific studies on brain tissue eQTLs and their pleiotropic effects in schizophrenia remain limited.\u003c/p\u003e \u003cp\u003eIn this study, we aim to investigate context-specific eQTLs and their impact on schizophrenia susceptibility. Using BrainSeq project Phase I dataset,\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e which includes genotype, gene expression, and sample metadata, we stratified individuals based on phenotype, age, and sex to conduct comprehensive eQTL analyses. We then examine differences in eQTLs and their pleiotropic effects across these contexts, focusing on schizophrenia-specific eQTLs and their transcriptional factor driving these regulatory changes. Additionally, we employed summary-data-based mendelian randomization (SMR),\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e to integrate context-specific eQTLs with PGC3-SZ GWAS data to identify risk genes mediating schizophrenia susceptibility. Through this investigation, we aim to address two key questions: (1) What are the underlying drivers of context-specific of eQTLs? and (2) How do these findings enhance our understanding of schizophrenia risk and pathogenesis?\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData and Sample Description\u003c/h2\u003e \u003cp\u003eThis study utilized data from the BrainSeq Project Phase I dataset.\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e Genotype data (SNP arrays) were obtained from dbGaP (phs000448.v1.p1), and RNA sequencing data was downloaded from Synapse (syn12299750). After initial filtering, we identified 456 samples with both genotyping and transcriptomic data. To reduce age-related confounding in eQTL analysis, embryonic samples were excluded, resulting in a final dataset of 414 postnatal samples.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eContext-Specific cis-eQTL Mapping\u003c/h3\u003e\n\u003cp\u003eSNP array data were initially processed using GenomeStudio 2.0. To expand the number of available variants for eQTL mapping, we performed genotyping imputation using IMPUTE2\u003csup\u003e19\u003c/sup\u003e referencing the 1000 Genomes Project dataset.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e Imputed results were filtered using an INFO score threshold of 0.8 to ensure quality, and converted to PLINK\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e format via gtool. This yielded genotype data for approximately 8.12\u0026nbsp;million SNPs. For transcriptomic data, RNA-seq reads were aligned to the hg19 reference genome\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e using STAR.\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e Gene-level quantification was performed using featureCounts,\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e and expression values were normalized as FPKM. Samples were stratified based on phenotype (schizophrenia vs. control), sex, chronological age (cutoff: 45 years), and epigenetic age (determined in our previous study,\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e). We conducted cis-eQTL analysis using MatrixEQTL,\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e considering SNPs within 1 Mb of each gene. Covariates included sex, age, ethnicity, and phenotype. The parameter pvOutputThreshold was set to 1 to retain all cis-eQTL results for downstream analyses.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eEnrichment Analysis\u003c/h3\u003e\n\u003cp\u003eSince the BrainSeq Phase I dataset derives from the Dorsolateral Prefrontal Cortex (DLPFC), we obtained chromatin state annotations for the adult DLPFC (Roadmap Epigenomics sample E073) via ChromHMM.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e Due to limited ChIP-seq dataset in the DLPFC, we utilized transcription factor binding data from ENCODE\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e SK-N-SH neuroblastoma cell line, which includes 65 ChIP-seq datasets for corresponding to 62 transcription factors. To avoid enrichment bias caused by linkage disequilibrium, we used GREGOR\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e to asses of eQTLs in transcription factor binding sites and chromatin statte. GREGOR generated linkage maps based on 1000 Genomes Project data, controlling for LD structure among SNPs.\u003c/p\u003e\n\u003ch3\u003eSummary-data-based Mendelian Randomization (SMR)\u003c/h3\u003e\n\u003cp\u003eTo assess causal relationships between gene expression and schizophrenia risk, we applied the SMR method.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e This integrative approach uses SNPs as instrumental variables, combining phenotype-specific cis-eQTLs from schizophrenia and control groups with PGC3-SZ GWAS summary statistics.\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Only genes with at least one significant eQTL (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;1E-5) were considered, corresponding to an estimated false discovery rate (FDR)\u0026thinsp;\u0026lt;\u0026thinsp;0.05. To account for multiple independent SNPs per gene, we used multi-SMR, which aggregates signals across all qualifying SNPs in a gene region to infer its putative effect on schizophrenia risk.\u003c/p\u003e\n\u003ch3\u003eData Availability\u003c/h3\u003e\n\u003cp\u003eAll data generated in this study are available. The context-specific eQTL summary data is available on Zenodo, with the identifier \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.5281/zenodo.14333896\u003c/span\u003e\u003cspan address=\"10.5281/zenodo.14333896\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eIdentifying Context-Specific eQTLs\u003c/h2\u003e \u003cp\u003eTo examine eQTL patterns under different contextual conditions, we analyzed 414 samples from the BrainSeq Project Phase I dataset,\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e integrating genomic (genetic variation) and transcriptomic (gene expression) data (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and \u003cb\u003eSupplementary Fig.\u0026nbsp;1)\u003c/b\u003e. Samples were categorized based on sex, disease status, age, and epigenetic age acceleration,\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e forming nine distinct groups (including an all-sample group). A genome-wide cis-eQTL analysis (1Mb window) was conducted using MatrixEQTL.\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eFor each group, we applied a false discovery rate (FDR) adjustment to the eQTL \u003cem\u003eP\u003c/em\u003e-values, setting a significance threshold of \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 for significant eQTL pairs. The corresponding \u003cem\u003eP\u003c/em\u003e-value threshold at this FDR level was recorded, revealing a strong correlation with sample size (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Notably, schizophrenia patients (SCZ) samples exhibited a higher-than-expected \u003cem\u003eP\u003c/em\u003e-value threshold, deviating from the regression curve. This suggests a distribution shift towards significance, implying schizophrenia-specific regulatory loci. We quantified the number of significant eQTL pairs, associated genes (eGene), and SNPs (eSNP), as well as the average number of eSNP per eGene. While the number of eQTLs and eSNPs in the SCZ group significantly deviated from the expected trend, the number of eGenes did not (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC, \u003cb\u003eSupplementary Fig.\u0026nbsp;2\u003c/b\u003e). This suggests that the schizophrenia involves an unexpectedly high number of regulatory variants influencing gene expression.\u003c/p\u003e \u003cp\u003eFocused on the SCZ group, we compared eQTL distribution with the control (Con) group, identifying numerous genome-wide eQTL pairs without clear chromosomal bias (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). However, eQTL distributions varied across chromosomes between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE-F and \u003cb\u003eSupplementary Fig.\u0026nbsp;3\u003c/b\u003e). For example, on chromosome 1 eQTLs in SCZ samples exhibited greater significance than in the Con group (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE), whereas chromosome 22 exhibited more significant eQTLs in the Con group (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF). The top SCZ-specific eQTL on chromosome 1 was rs138513841, associated with the \u003cem\u003eNKAIN1\u003c/em\u003e, while the leading eQTL on chromosome 22 was rs75493558, linked to \u003cem\u003ePARVG\u003c/em\u003e. Both showed reduced significance in the control group (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG-H). \u003cem\u003eNKAIN1\u003c/em\u003e has been implicated in autism and alcohol dependence, while \u003cem\u003ePARVG\u003c/em\u003e is linked to Alzheimer's disease and alcohol dependence \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. These findings suggest that SCZ-specific eQTLs may contribute to broader neuropsychiatric disorder mechanisms.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eWidespread Loss, Gain, and Consistent eQTLs in Schizophrenia\u003c/h3\u003e\n\u003cp\u003eTo systematically assess similarities and differences in eQTL patterns between the Con and SCZ groups, we employed the upsetR package\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e to visualize intersection counts of eQTL pairs across all samples (ALL), Con samples, and SCZ samples. The Con group contained 229,392 eQTL pairs (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), of which 145,736 eQTLs (63.53%) were not significant in the SCZ group (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Analysis of the \u003cem\u003eP\u003c/em\u003e-value distribution for these eQTLs in the SCZ group (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB) showed that 51.34% had \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05, indicating a widespread loss of eQTLs in schizophrenia. Conversely, the SCZ group contained 168,806 significant eQTLs (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05), with 85,150 (50.44%) not significant in the Con group (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05). Analysis of the \u003cem\u003eP\u003c/em\u003e-value distribution for these SCZ-Gain eQTLs in the Con group (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC) revealed that 53.4% had \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05, suggesting a substantial gain of eQTLs in schizophrenia. After filtering out eSNPs targeting any gene in the SCZ group with \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05, we identified 54,146 eSNPs corresponding to 7,605 eGenes and 61,326 SCZ-Loss eQTL pairs (\u003cb\u003eSupplementary Table\u0026nbsp;1\u003c/b\u003e). Similarly, after excluding eSNPs targeting any gene in the Con group with \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, we identified 34,331 eSNPs corresponding to 5,142 eGenes and 37,907 SCZ-Gain eQTL pairs (\u003cb\u003eSupplementary Table\u0026nbsp;2\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAdditionally, 83,656 eQTL pairs were significant in both the Con and SCZ groups (\u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e). Comparing effect sizes of these shared eQTL pairs revealed that their directional effect remained consistent. When one group was fixed at \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and the other was relaxed to \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, only a small subset of eSNPs displayed opposing effect directions. Notably, as \u003cem\u003eP\u003c/em\u003e-values became more significant, the number of opposing cases declined, indicating that opposing effect sizes are relatively rare (\u003cb\u003eSupplementary Fig.\u0026nbsp;4\u003c/b\u003e).\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eGenes Competing for Regulatory Elements Lead to Genetic Pleiotropy in Genetic Regulation\u003c/h2\u003e \u003cp\u003eTo investigate the impact of genetic pleiotropy on gene expression regulation in schizophrenia, we analyzed the overlap of eSNP between Con-Specific eQTLs (Con \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 \u0026amp; SCZ \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05) and SCZ-Specific eQTLs (Con \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05 \u0026amp; SCZ \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). We identified 6,821 overlapping eSNPs, which target different genes in the Con and SCZ groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). For clarity, we denote genes significantly targeted by eSNPs in the Con group (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) as Gene A and those significantly targeted in the SCZ group (\u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05) as Gene B. Analyzing eQTL \u003cem\u003eP\u003c/em\u003e-value distributions for Gene B in the Con group and Gene A in the SCZ group, we observed a significant enrichment in both eQTL categories with \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE, eSNP-Gene A-Gene B Pair \u003cem\u003eN\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2,583, \u003cem\u003eOR\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.94, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.213E-145).\u003c/p\u003e \u003cp\u003eAfter excluding eSNPs with at least one eQTL \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 \u0026amp; \u003cem\u003eFDR\u003c/em\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.05 in either group, we identified 1,144 eSNPs corresponding to 265 eGenes, forming 351 eGene-eGene pairs and 1,358 eQTL-eQTL pairs (\u003cb\u003eSupplementary Table\u0026nbsp;4\u003c/b\u003e). This subset, indicative of eGene switch, exhibited a strong tendency for consistent effect size, with 1,218 eQTL-eQTL pairs (89.68%) maintaining the same directional effect (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). In contrast, only 55.39% of other eQTL-eQTL pairs showed consistent effects, highlighting a stronger consistency in the eGene switch subset (\u003cem\u003eOR\u003c/em\u003e\u0026thinsp;=\u0026thinsp;7.588, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;4.725E-161). Expression correlation analyses within the eGene switch subset revealed 163 negatively correlated gene pairs (46.44%), significantly higher than the 176 negatively correlated pairs (25.69%) observed in other gene sets (\u003cem\u003eOR\u003c/em\u003e\u0026thinsp;=\u0026thinsp;2.50, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;3.543E-11, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG). This suggests that genes undergoing eGene switch are more likely to exhibit inverse expression relationships.\u003c/p\u003e \u003cp\u003eIn summary, within the eGene switch subset, SNP effects targeting different genes align in the same direction, while their gene expression correlations are inversely related. This suggests competition for regulatory loci, leading to genetic pleiotropy in gene expression regulation across different contexts (control and schizophrenia). Notably, the strongest negatively correlated gene pair identified was regulated by SNP rs77548189, which influences \u003cem\u003eZNF268\u003c/em\u003e expression in the Con group (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH) and \u003cem\u003eGALNT9\u003c/em\u003e expression in the SCZ group (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eI). The competition for rs77548189 between \u003cem\u003eZNF268\u003c/em\u003e and \u003cem\u003eGALNT9\u003c/em\u003e leads to a negative correlation in their expression levels (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eJ). \u003cem\u003eGALNT9\u003c/em\u003e has been associated with autism,\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e while \u003cem\u003eZNF268\u003c/em\u003e has not been linked to neuropsychiatric disorders, suggesting that genetic pleiotropy in gene expression regulation plays a multifaceted role in neuropsychiatric disorder development.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eRepressive Histone Mark Induced by EZH2 is involved in Genetic Pleiotropy Associated with Schizophrenia\u003c/h2\u003e \u003cp\u003eWe categorized eSNPs corresponding to eQTLs into Loss, Gain, Consistent, and Switch groups, further classifying Switch eQTLs into Negative and Positive correlations based on gene expression relationships (\u003cb\u003eSupplementary Table\u0026nbsp;1\u0026ndash;4\u003c/b\u003e). To explore the genetic characteristics for these categories, we analyzed their associations with PGC3-SZ GWAS\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e beta values, \u003cem\u003eP\u003c/em\u003e-values, minor allele frequencies (MAF), and distances between eSNPs and their respective eGenes. Notably, eSNPs in the Negative Switch category exhibited the most significant \u003cem\u003eP\u003c/em\u003e-values, followed by those in the Positive Switch, Consistent, Gain, and Loss categories. (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Similarly, Negative Switch eSNPs had the highest GWAS beta values, followed by those in the Positive Switch category (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). MAF analysis revealed that eSNPs in the Loss, Gain, and Switch Category had MAF around 0.1, whereas Consistent eSNPs had higher frequencies (0.3) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Additionally, eSNPs in the Negative Switch Category were positioned farther from their target genes, with a median distance approaching 500 kb (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). These findings suggest that Switch eQTLs, particularly those in the Negative Switch category, involve variants with low MAF and long-distance regulatory effects and strong associations with schizophrenia.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo investigate potential epigenetic mechanism, we performed chromatin state enrichment analysis using ChromHMM\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e annotations from the Roadmap Epigenomics project.\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e This revealed a significant enrichment of Negative Switch eSNPs in repressive chromatin regions marked by the ReprPC state (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), implicating repressive histone modification in genetic pleiotropy. Further analysis of 65 transcriptional factor (TF) binding sites in SK-N-SH ChIP-Seq data\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e showed that Negative Switch eSNPs category were highly enriched in these regulatory regions. Notably, the most enriched TFs were EZH2 (phospho-T487) and EP300 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF), suggesting that repressive histone modifications, such as H3K27me3 induced by EZH2, may contribute to the schizophrenia-associated genetic pleiotropy.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePhenotype-Specific eQTL Contributed to Identifying Schizophrenia Risk Genes Using SMR Analysis\u003c/h2\u003e \u003cp\u003eTo further assess the role of schizophrenia-specific eQTL in disease risk, we conducted summary data-based Mendelian randomization (SMR) analysis.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e This approach integrates eQTL data with PGC3-SZ GWAS data to identify gene whose expression levels may be causally linked to schizophrenia. We analyzed 2,854 genes using eQTL data from control individuals (\u003cb\u003eSupplementary Table\u0026nbsp;6\u003c/b\u003e) and 2,187 genes using eQTL data from schizophrenia patients (\u003cb\u003eSupplementary Table\u0026nbsp;7\u003c/b\u003e), with 969 overlapped genes between two groups.\u003c/p\u003e \u003cp\u003eTo evaluate the impact of genetic pleiotropy on SMR analysis, we extracted the top SNPs (i.e., SNPs with the most significant eQTL \u003cem\u003eP\u003c/em\u003e-value) corresponding to these 969 overlapping genes. We identified 144 shared top SNPs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), which regulated 133 overlapping genes, 15 control-specific genes, and 28 schizophrenia-specific genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). This resulted in 33 unique gene-gene pairs further supporting the notion that eSNPs regulate different genes in the control and schizophrenia groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC, \u003cb\u003eSupplementary Table\u0026nbsp;8\u003c/b\u003e). Applying an \u003cem\u003eFDR\u003c/em\u003e threshold\u0026thinsp;\u0026lt;\u0026thinsp;0.1, we identified a key SNP, rs16846667, located on chromosome 1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). This variant regulates \u003cem\u003eSERPINC1\u003c/em\u003e in the control group (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) but \u003cem\u003eANKRD45\u003c/em\u003e in the schizophrenia group (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF-G). ANKRD45 belongs to the ankyrin repeat family, and knockout mice exhibit abnormal behaviors in open field test, including significantly reduced resting time\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e(Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH). In contrast, there is no known association between \u003cem\u003eSERPINC1\u003c/em\u003e and nervous system function. These findings suggest that rs16846667 may contribute to schizophrenia susceptibility by modulating \u003cem\u003eANKRD45\u003c/em\u003e expression in affected individuals.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eA key finding of this study is the universality of eQTL specificity, particularly among individuals with schizophrenia, where the significance of eQTLs exceeded expectations. This suggests that genetic variation may have distinct regulatory functions within schizophrenia patients. While eQTLs are known to exhibit tissue and cell specificity due to differences in TF expression and cellular environments,\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e disease specificity remains underexplored. Recent studies on non-alcoholic fatty liver disease (NAFLD)\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e and idiopathic pulmonary fibrosis (IPF)\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e have identified disease-specific eQTLs that influence gene expression and disease progression, supporting the relevance of our findings. Notably, we identified a large number of specific eQTLs in schizophrenia, with over 50% of eQTLs in both control and schizophrenia groups not reaching significance in the other group, highlighting the need for focused investigations into schizophrenia-specific eQTLs.\u003c/p\u003e \u003cp\u003eAnother major contribution of this study is the exploration of genetic pleiotropy in gene regulation and its role in schizophrenia susceptibility. Traditionally, pleiotropy refers to a single genetic variant influencing multiple genes through transcriptional regulatory complexes involving enhancers and promoters.\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e However, our findings suggest a different mechanism\u0026mdash;multiple genes competing for the same regulatory sites, leading to phenotype-dependent gene expression shifts. This pattern parallels hemoglobin switching, where locus control regions in the β-globin cluster regulate gene expression during development.\u003csup\u003e\u003cspan additionalcitationids=\"CR46\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eFurthermore, our analysis revealed that schizophrenia-associated pleiotropic genetic variants are significantly enriched in EZH2 binding sites, a key component of the PRC2 complex. PRC2, through H3K27me3-mediated chromatin modifications, plays a crucial role in embryonic development, cell differentiation, cell fate determination, and gene silencing.\u003csup\u003e\u003cspan additionalcitationids=\"CR49\" citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e Dysregulation of PRC2 has been implicated in neurodevelopmental abnormalities and neurotransmitter metabolism.\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e Our previous work also demonstrated that PRC2\u0026rsquo;s role in the allele-specific methylation changes in \u003cem\u003eEIPR1\u003c/em\u003e in monozygotic twins discordant for schizophrenia.\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e Collectively, our findings suggest that PRC2-mediated regulatory shifts contribute to schizophrenia susceptibility by altering gene expression in context-dependent manner.\u003c/p\u003e \u003cp\u003eTo functionally interpret these regulatory changes, we integrate phenotype-specific eQTL data with PGC3-SZ GWAS data using the SMR tool. This analysis identified schizophrenia-associated variant rs16846667 on chromosome 1, which regulates \u003cem\u003eSERPINC1\u003c/em\u003e in control individuals but \u003cem\u003eANKRD45\u003c/em\u003e in schizophrenia patients. Notably, Ankrd45 knockout mice exhibited behavioral abnormalities, whereas \u003cem\u003eSERPINC1\u003c/em\u003e has no known neurological associations. This highlights the utility of phenotype-specific eQTLs in pinpointing genes mediating the functional effects of genetic variation on disease risk.\u003c/p\u003e \u003cp\u003eAs a resource for future research, we provide eQTL data stratified by age, sex, phenotype, and epigenetic age, creating eight distinct groups. This data can be further analyzed using methods such as SMR, coloc,\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e and TWAS\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e to explore the influence of genetic regulation across different biological contexts.\u003c/p\u003e \u003cp\u003eDespite these insights, our study has limitations. The BrainSeq Phase I dataset includes both European and African American individuals, which may introduce population-specific biases. Given the need for subgroup analyses based on age, sex, and phenotype, restricting our sample to a single population would have significantly reduced statistical power. However, previous research has shown high consistency in eQTL results across these populations,\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e and we account for population as a covariate in our analyses to mitigate potential biases.\u003c/p\u003e \u003cp\u003eIn conclusion, our study establishes the context-specificity of eQTLs in schizophrenia and demonstrates that genetic pleiotropy arises from multiple genes competing for shared regulatory sites. The PRC2 complex, particularly EZH2, plays a crucial role in mediating these shifts in gene regulation between control and schizophrenia groups. These findings provide new insights into the molecular mechanisms underlying schizophrenia susceptibility and emphasize the importance of investigating phenotype-specific eQTLs. By making our context-specific eQTL dataset publicly available, we offer a valuable resource for future studies delving deeper into the molecular mechanisms of neuropsychiatric disorders.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAuthor contributions:\u003c/em\u003e Conceived and designed the analysis: ZW, LY, and ZS; Provided the rawdata of BrainSeq project: CZ; Analyzed the data: ZW, LY, XW, JY, and ZL; Wrote the paper: ZW, LY, CZ, QY, SC, and FX.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by a grant of the National Natural Science Foundation of China [grant numbers 82101577]; China Postdoctoral Science Foundation [grant numbers 2023M730724, 2020M682806].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors report no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSupplementary material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSupplementary material is available online.\u003c/p\u003e\u003ch2\u003eAuthor contributions:\u003c/h2\u003e \u003cp\u003eConceived and designed the analysis: ZW, LY, and ZS; Provided the rawdata of BrainSeq project: CZ; Analyzed the data: ZW, LY, XW, JY, and ZL; Wrote the paper: ZW, LY, CZ, QY, SC, and FX.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eHilker R, Helenius D, Fagerlund B et al (2018) Heritability of Schizophrenia and Schizophrenia Spectrum Based on the Nationwide Danish Twin Register. \u003cem\u003eBiological psychiatry\u003c/em\u003e. 15(6):492\u0026ndash;498. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.biopsych.2017.08.017\u003c/span\u003e\u003cspan address=\"10.1016/j.biopsych.2017.08.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBiological (2014) insights from 108 schizophrenia-associated genetic loci. \u003cem\u003eNature\u003c/em\u003e. Jul 24 ;511(7510):421-7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature13595\u003c/span\u003e\u003cspan address=\"10.1038/nature13595\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLam M, Chen CY, Li Z et al (2019) Comparative genetic architectures of schizophrenia in East Asian and European populations. Nat Genet Dec 51(12):1670\u0026ndash;1678. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41588-019-0512-x\u003c/span\u003e\u003cspan address=\"10.1038/s41588-019-0512-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePardi\u0026ntilde;as AF, Holmans P, Pocklington AJ et al (2018) Common schizophrenia alleles are enriched in mutation-intolerant genes and in regions under strong background selection. Nat Genet Mar 50(3):381\u0026ndash;389. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41588-018-0059-2\u003c/span\u003e\u003cspan address=\"10.1038/s41588-018-0059-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrubetskoy V, Pardi\u0026ntilde;as AF, Qi T et al (2022) Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nat Apr 604(7906):502\u0026ndash;508. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-022-04434-5\u003c/span\u003e\u003cspan address=\"10.1038/s41586-022-04434-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Y, Liu S, Ren Z et al (2024) Cross-ancestry analysis of brain QTLs enhances interpretation of schizophrenia genome-wide association studies. Am J Hum Genet Nov 7(11):2444\u0026ndash;2457. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.ajhg.2024.09.001\u003c/span\u003e\u003cspan address=\"10.1016/j.ajhg.2024.09.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Khan A, Kushner SA, Rzhetsky A (2022) Dissecting schizophrenia phenotypic variation: the contribution of genetic variation, environmental exposures, and gene-environment interactions. Schizophrenia (Heidelberg Germany) May 10(1):51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41537-022-00257-5\u003c/span\u003e\u003cspan address=\"10.1038/s41537-022-00257-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhernakova DV, Deelen P, Vermaat M et al (2017) Identification of context-dependent expression quantitative trait loci in whole blood. Nat Genet Jan 49(1):139\u0026ndash;145. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3737\u003c/span\u003e\u003cspan address=\"10.1038/ng.3737\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFindley AS, Monziani A, Richards AL et al (2021) Functional dynamic genetic effects on gene regulation are specific to particular cell types and environmental conditions. eLife May 14(10). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.7554/eLife.67077\u003c/span\u003e\u003cspan address=\"10.7554/eLife.67077\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDi Forti M, Marconi A, Carra E et al (2015) Proportion of patients in south London with first-episode psychosis attributable to use of high potency cannabis: a case-control study. lancet Psychiatry Mar 2(3):233\u0026ndash;238. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/s2215-0366(14)00117-5\u003c/span\u003e\u003cspan address=\"10.1016/s2215-0366(14)00117-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHjorth\u0026oslash;j C, Compton W, Starzer M et al (2023) Association between cannabis use disorder and schizophrenia stronger in young males than in females. Psychol Med Nov 53(15):7322\u0026ndash;7328. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/s0033291723000880\u003c/span\u003e\u003cspan address=\"10.1017/s0033291723000880\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLupien SJ, McEwen BS, Gunnar MR, Heim C (2009) Effects of stress throughout the lifespan on the brain, behaviour and cognition. Nat reviews Neurosci Jun 10(6):434\u0026ndash;445. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nrn2639\u003c/span\u003e\u003cspan address=\"10.1038/nrn2639\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenros ME, Trabjerg BB, Meier S et al (2016) Influence of Polygenic Risk Scores on the Association Between Infections and Schizophrenia. Biol psychiatry Oct 15(8):609\u0026ndash;616. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.biopsych.2016.04.008\u003c/span\u003e\u003cspan address=\"10.1016/j.biopsych.2016.04.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelaneau O, Zazhytska M, Borel C et al (2019) Chromatin three-dimensional interactions mediate genetic effects on gene expression. \u003cem\u003eScience (New York, NY)\u003c/em\u003e. May 3. ;364(6439)\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.aat8266\u003c/span\u003e\u003cspan address=\"10.1126/science.aat8266\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMumbach MR, Satpathy AT, Boyle EA et al (2017) Enhancer connectome in primary human cells identifies target genes of disease-associated DNA elements. Nat Genet Nov 49(11):1602\u0026ndash;1612. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3963\u003c/span\u003e\u003cspan address=\"10.1038/ng.3963\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJaffe AE, Straub RE, Shin JH et al (2018) Developmental and genetic regulation of the human cortex transcriptome illuminate schizophrenia pathogenesis. Nat Neurosci Aug 21(8):1117\u0026ndash;1125. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41593-018-0197-y\u003c/span\u003e\u003cspan address=\"10.1038/s41593-018-0197-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhu Z, Zhang F, Hu H et al (2016) Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat Genet May 48(5):481\u0026ndash;487. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3538\u003c/span\u003e\u003cspan address=\"10.1038/ng.3538\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu Y, Zeng J, Zhang F et al (2018) Integrative analysis of omics summary data reveals putative mechanisms underlying complex traits. Nat Commun Mar 2(1):918. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41467-018-03371-0\u003c/span\u003e\u003cspan address=\"10.1038/s41467-018-03371-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHowie BN, Donnelly P, Marchini J (2009) A flexible and accurate genotype imputation method for the next generation of genome-wide association studies. PLoS Genet Jun 5(6):e1000529. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pgen.1000529\u003c/span\u003e\u003cspan address=\"10.1371/journal.pgen.1000529\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbecasis GR, Altshuler D, Auton A et al (2010) A map of human genome variation from population-scale sequencing. Nat Oct 28(7319):1061\u0026ndash;1073. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature09534\u003c/span\u003e\u003cspan address=\"10.1038/nature09534\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ (2015) Second-generation PLINK: rising to the challenge of larger and richer datasets. GigaScience 4:7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13742-015-0047-8\u003c/span\u003e\u003cspan address=\"10.1186/s13742-015-0047-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMudge JM, Carbonell-Sala S, Diekhans M et al (2025) GENCODE 2025: reference gene annotation for human and mouse. Nucleic acids Res Jan 6(D1):D966\u0026ndash;d975. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkae1078\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkae1078\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrankish A, Diekhans M, Ferreira AM et al (2019) GENCODE reference annotation for the human and mouse genomes. Nucleic acids Res Jan 8(D1):D766\u0026ndash;d773. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gky955\u003c/span\u003e\u003cspan address=\"10.1093/nar/gky955\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobin A, Davis CA, Schlesinger F et al (2013) STAR: ultrafast universal RNA-seq aligner. \u003cem\u003eBioinformatics (Oxford, England)\u003c/em\u003e. Jan 1. ;29(1):15\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/bts635\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/bts635\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiao Y, Smyth GK, Shi W (2013) The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic acids Res May 1(10):e108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkt214\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkt214\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWu X, Ye J, Wang Z, Zhao C (2021) Epigenetic Age Acceleration Was Delayed in Schizophrenia. Schizophr Bull Apr 29(3):803\u0026ndash;811. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/schbul/sbaa164\u003c/span\u003e\u003cspan address=\"10.1093/schbul/sbaa164\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShabalin AA (2012) Matrix eQTL: ultra fast eQTL analysis via large matrix operations. Bioinf (Oxford England) May 15(10):1353\u0026ndash;1358. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/bts163\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/bts163\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eErnst J, Kellis M (2012) ChromHMM: automating chromatin-state discovery and characterization. Nat methods Feb 28(3):215\u0026ndash;216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nmeth.1906\u003c/span\u003e\u003cspan address=\"10.1038/nmeth.1906\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoadmap Epigenomics C, Kundaje A, Meuleman W et al (2015) Integrative analysis of 111 reference human epigenomes. Nat Feb 19(7539):317\u0026ndash;330. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature14248\u003c/span\u003e\u003cspan address=\"10.1038/nature14248\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAn integrated encyclopedia (2012) of DNA elements in the human genome. Nat Sep 6 489(7414):57\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature11247\u003c/span\u003e\u003cspan address=\"10.1038/nature11247\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmidt EM, Zhang J, Zhou W et al (2015) GREGOR: evaluating global enrichment of trait-associated variants in epigenomic features using a systematic, data-driven approach. \u003cem\u003eBioinformatics (Oxford, England)\u003c/em\u003e. Aug 15. ;31(16):2601-6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/bioinformatics/btv201\u003c/span\u003e\u003cspan address=\"10.1093/bioinformatics/btv201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRappaport N, Nativ N, Stelzer G et al (2013) MalaCards: an integrated compendium for diseases and their annotation. Database: J Biol databases curation 2013:bat018. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/database/bat018\u003c/span\u003e\u003cspan address=\"10.1093/database/bat018\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRappaport N, Twik M, Plaschkes I et al (2017) MalaCards: an amalgamated human disease compendium with diverse clinical and genetic annotation and structured search. Nucleic acids Res Jan 4(D1):D877\u0026ndash;d887. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkw1012\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkw1012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLex A, Gehlenborg N, Strobelt H, Vuillemot R, Pfister H, UpSet (2014) Visualization of Intersecting Sets. IEEE Trans visualization Comput graphics Dec 20(12):1983\u0026ndash;1992. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1109/tvcg.2014.2346248\u003c/span\u003e\u003cspan address=\"10.1109/tvcg.2014.2346248\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStelzer G, Rosen N, Plaschkes I et al (2016) The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr protocols Bioinf Jun 20(1301\u0026ndash;). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1002/cpbi.5\u003c/span\u003e\u003cspan address=\"10.1002/cpbi.5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGroza T, Gomez FL, Mashhadi HH et al (2023) The International Mouse Phenotyping Consortium: comprehensive knockout phenotyping underpinning the study of human disease. Nucleic acids Res Jan 6(D1):D1038\u0026ndash;d1045. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1093/nar/gkac972\u003c/span\u003e\u003cspan address=\"10.1093/nar/gkac972\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDickinson ME, Flenniken AM, Ji X et al (2016) High-throughput discovery of novel developmental phenotypes. Nat Sep 22(7621):508\u0026ndash;514. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature19356\u003c/span\u003e\u003cspan address=\"10.1038/nature19356\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConsortium GT, Laboratory DA, Coordinating Center -Analysis, Working G et al (2017) Genetic effects on gene expression across human tissues. Nat Oct 11(7675):204\u0026ndash;213. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature24277\u003c/span\u003e\u003cspan address=\"10.1038/nature24277\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim-Hellmuth S, Aguet F, Oliva M et al (2020) Cell type-specific genetic regulation of gene expression across human tissues. \u003cem\u003eScience (New York, NY)\u003c/em\u003e. Sep 11. ;369(6509)\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.aaz8528\u003c/span\u003e\u003cspan address=\"10.1126/science.aaz8528\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLappalainen T, Sammeth M, Friedl\u0026auml;nder MR et al (2013) Transcriptome and genome sequencing uncovers functional variation in humans. Nat Sep 26(7468):506\u0026ndash;511. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature12531\u003c/span\u003e\u003cspan address=\"10.1038/nature12531\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYoo T, Joo SK, Kim HJ et al (2021) Disease-specific eQTL screening reveals an anti-fibrotic effect of AGXT2 in non-alcoholic fatty liver disease. J Hepatol Sep 75(3):514\u0026ndash;523. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jhep.2021.04.011\u003c/span\u003e\u003cspan address=\"10.1016/j.jhep.2021.04.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDalapati T, Wang L, Jones AG et al (2025) Context-specific eQTLs provide deeper insight into causal genes underlying shared genetic architecture of COVID-19 and idiopathic pulmonary fibrosis. HGG Adv Jan 27(2):100410. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.xhgg.2025.100410\u003c/span\u003e\u003cspan address=\"10.1016/j.xhgg.2025.100410\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim TK, Shiekhattar R (2015) Architectural and Functional Commonalities between Enhancers and Promoters. Cell Aug 27(5):948\u0026ndash;959. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.cell.2015.08.008\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2015.08.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLevo M, Raimundo J, Bing XY et al (2022) Transcriptional coupling of distant regulatory genes in living embryos. Nat May 605(7911):754\u0026ndash;760. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41586-022-04680-7\u003c/span\u003e\u003cspan address=\"10.1038/s41586-022-04680-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarlsson S, Nienhuis AW (1985) Developmental regulation of human globin genes. Annu Rev Biochem 54:1071\u0026ndash;1108. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1146/annurev.bi.54.070185.005231\u003c/span\u003e\u003cspan address=\"10.1146/annurev.bi.54.070185.005231\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStamatoyannopoulos G (1991) Human hemoglobin switching. \u003cem\u003eScience (New York, NY)\u003c/em\u003e. Apr 19. ;252(5004):383. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.2017679\u003c/span\u003e\u003cspan address=\"10.1126/science.2017679\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim S, Kim YW, Shim SH, Kim CG, Kim A (2012) Chromatin structure of the LCR in the human β-globin locus transcribing the adult δ- and β-globin genes. Int J Biochem cell biology Mar 44(3):505\u0026ndash;513. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.biocel.2011.12.001\u003c/span\u003e\u003cspan address=\"10.1016/j.biocel.2011.12.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMargueron R, Reinberg D (2011) The Polycomb complex PRC2 and its mark in life. Nat Jan 20(7330):343\u0026ndash;349. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nature09784\u003c/span\u003e\u003cspan address=\"10.1038/nature09784\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreer EL, Shi Y (2012) Histone methylation: a dynamic mark in health, disease and inheritance. \u003cem\u003eNature reviews Genetics\u003c/em\u003e. Apr 3. ;13(5):343\u0026thinsp;\u0026ndash;\u0026thinsp;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nrg3173\u003c/span\u003e\u003cspan address=\"10.1038/nrg3173\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMajewski IJ, Blewitt ME, de Graaf CA et al (2008) Polycomb repressive complex 2 (PRC2) restricts hematopoietic stem cell activity. PLoS biology Apr 15(4):e93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pbio.0060093\u003c/span\u003e\u003cspan address=\"10.1371/journal.pbio.0060093\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang YY, Deng YS, Dai SK et al (2022) Loss of microglial EED impairs synapse density, learning, and memory. Mol psychiatry Jul 27(7):2999\u0026ndash;3009. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41380-022-01576-w\u003c/span\u003e\u003cspan address=\"10.1038/s41380-022-01576-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchr\u0026ouml;der S, Fuchs U, Gisa V et al (2024) PRDM16-DT is a novel lncRNA that regulates astrocyte function in Alzheimer's disease. Acta Neuropathol Aug 29(1):32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00401-024-02787-x\u003c/span\u003e\u003cspan address=\"10.1007/s00401-024-02787-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Q, Wang Z, Zong L et al (2021) Allele-specific DNA methylation maps in monozygotic twins discordant for psychiatric disorders reveal that disease-associated switching at the EIPR1 regulatory loci modulates neural function. Mol Psychiatry Nov 26 26(11):6630\u0026ndash;6642. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41380-021-01126-w\u003c/span\u003e\u003cspan address=\"10.1038/s41380-021-01126-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiambartolomei C, Vukcevic D, Schadt EE et al (2014) Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet May 10(5):e1004383. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pgen.1004383\u003c/span\u003e\u003cspan address=\"10.1371/journal.pgen.1004383\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWallace C (2021) A more accurate method for colocalisation analysis allowing for multiple causal variants. PLoS Genet Sep 17(9):e1009440. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pgen.1009440\u003c/span\u003e\u003cspan address=\"10.1371/journal.pgen.1009440\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGamazon ER, Wheeler HE, Shah KP et al (2015) A gene-based association method for mapping traits using reference transcriptome data. Nat Genet Sep 47(9):1091\u0026ndash;1098. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/ng.3367\u003c/span\u003e\u003cspan address=\"10.1038/ng.3367\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMai J, Lu M, Gao Q, Zeng J, Xiao J (2023) Transcriptome-wide association studies: recent advances in methods, applications and available databases. Commun biology Sep 1(1):899. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s42003-023-05279-y\u003c/span\u003e\u003cspan address=\"10.1038/s42003-023-05279-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[{"identity":"fa416202-c701-457d-8e26-7c52f8e9b19c","identifier":"10.13039/501100001809","name":"National Natural Science Foundation of China","awardNumber":"82101577","order_by":0},{"identity":"98054625-589a-4ba6-920c-982d933b1c1d","identifier":"10.13039/501100002858","name":"China Postdoctoral Science Foundation","awardNumber":"2023M730724","order_by":1},{"identity":"85ee4a6f-9aeb-4b90-a920-e965efde66b2","identifier":"10.13039/501100002858","name":"China Postdoctoral Science Foundation","awardNumber":"2020M682806","order_by":2}],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Gaozhou People's Hospital","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":"Expression quantitative trait loci, Mendelian randomization, Genome-wide association study, Transcriptional Regulation Dynamics, Polycomb repressive complex, H3K27me3","lastPublishedDoi":"10.21203/rs.3.rs-6582362/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6582362/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSchizophrenia is a highly heritable psychiatric disorder, but functional mechanisms linking genetic risk to pathogenesis remain poorly understood. Most schizophrenia-associated variants reside in non-coding regions, highlighting the need to investigate their regulatory roles.\u003c/p\u003e \u003cp\u003eThis study employs context-specific expression quantitative trait loci (eQTL) analysis using the BrainSeq Phase I dataset to dissect schizophrenia-associated regulatory dynamics.\u003c/p\u003e \u003cp\u003eComparative eQTL analysis revealed widespread loss and gain of regulatory associations in schizophrenia group versus controls, alongside consistent eQTLs. A notable target gene switching phenomenon emerged, where specific SNPs regulated distinct genes across disease states, indicative of genetic pleiotropy mediated by competition for shared regulatory elements. Pleiotropic SNPs exhibited stronger schizophrenia associations, localized farther from target genes, and were enriched in repressive chromatin domains marked by H3K27me3. Transcription factor binding site analysis implicated EZH2, a polycomb repressive complex component, in mediating these regulatory shifts. Integration of schizophrenia-specific eQTLs with GWAS data via mendelian randomization prioritized risk genes like ANKRD45, which showed disease-context regulation and links to behavioral deficits in knockout models.\u003c/p\u003e \u003cp\u003eOverall, we found the universality of eQTL specificity, and revealed a new mechanism that multiple genes competing for the shared regulatory sites, leading to phenotype-dependent gene expression shifts. This study establish context-specific eQTL dynamics as a critical layer of schizophrenia's genetic architecture. These insights advance functional interpretation of non-coding risk variants and provide new insights into regulatory mechanisms contributing to disease susceptibility.\u003c/p\u003e","manuscriptTitle":"Context-specific eQTL dynamics uncover genetic pleiotropy and chromatin-mediated target gene switching in schizophrenia","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-06 09:34:36","doi":"10.21203/rs.3.rs-6582362/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":"4b8c013d-be9e-4d2d-84ad-5effae327d17","owner":[],"postedDate":"May 6th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48036728,"name":"Molecular Genetics"}],"tags":[],"updatedAt":"2025-05-06T09:34:36+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-06 09:34:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6582362","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6582362","identity":"rs-6582362","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}