Gene-Environment Interaction through single-cell transcriptomics in Type 2 diabetes

Gene-Environment Interaction through single-cell transcriptomics in Type 2 diabetes | 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 Systematic Review Gene-Environment Interaction through single-cell transcriptomics in Type 2 diabetes Rupanjali Singh This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9283816/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Type 2 diabetes mellitus (T2DM) arises from the interplay between inherited genetic factors and environmental influences such as lifestyle, metabolic challenges, and inflammation. Objective: To provide a systematic review and meta-analysis of human studies that use single-cell transcriptomic methods to examine gene–environment interactions in T2DM, focusing on the integration of genome-wide association studies (GWAS), polygenic risk scores (PRS), environmental exposures, and comprehensive multi-omics analysis frameworks. Methods: We conducted a systematic literature search of PubMed/MEDLINE, EMBASE, Web of Science, Scopus, and Google Scholar for eligible human studies published from January 2021 to December 2025. Studies were included if they applied single-cell transcriptomics to tissues relevant to T2DM and combined genetic risk assessment with environmental or clinical data. Study quality was evaluated using modified STROBE and QUADAS-2 tools. When suitable, random-effects meta-analyses were conducted, and heterogeneity was estimated using the I² statistic. Results: Seven studies of high methodological quality were included. Single-cell transcriptomic analyses consistently revealed significant cellular diversity in pancreatic, immune, and peripheral tissues, with disease-related gene expression changes restricted to specific cell populations. The integration of PRS showed that genetic risk was distributed unevenly across various cell types, identifying subsets of cells with heightened genetic susceptibility. Environmental exposures were found to influence the relationship between genetic variations and gene expression, indicating genuine gene–environment interactions at the single-cell level. The meta-analysis revealed a significant overall association between molecular dysregulation and T2DM-related traits, with moderate heterogeneity (I² ≈ 48%). Conclusion: The available evidence demonstrates that gene–environment interactions in T2DM are highly specific to individual cell types and are best elucidated through single-cell transcriptomic studies. Combining genetic risk, environmental exposures, and multi-omics analysis yields valuable mechanistic insights and informs the advancement of precision medicine for T2DM. Systems Biology Computational Biology Type 2 diabetes mellitus Single-cell transcriptomics Gene–environment interaction Polygenic risk score Genome-wide association studies Cellular heterogeneity Multiomics integration Machine learning Precision medicine Environmental exposure Figures Figure 1 Figure 2 Figure 3 Introduction Type 2 diabetes mellitus (T2DM) is a chronic metabolic disorder characterized by persistent hyperglycaemia resulting from a combination of insulin resistance and progressive pancreatic β-cell dysfunction [1] . The development and progression of T2DM are driven by a complex interplay between inherited genetic susceptibility and a wide range of environmental influences, including diet, physical activity, obesity, metabolic stress, and inflammatory exposures [2] . Despite substantial advances in understanding disease epidemiology and clinical management, the biological mechanisms linking genetic risk to cellular dysfunction in specific tissues remain incompletely defined. Large-scale genome-wide association studies (GWAS) have successfully identified numerous genetic loci associated with T2DM risk. However, the majority of these variants reside in non-coding regions of the genome, limiting direct interpretation of their functional consequences [3] . As a result, traditional genetic association approaches provide limited insight into how risk variants influence gene regulation within specific cell types or how these effects are modified by environmental exposures [4] . This gap has hindered the translation of genetic discoveries into mechanistic understanding and precision therapeutic strategies. Recent technological advances in single-cell transcriptomics have transformed the study of complex metabolic diseases by enabling gene expression profiling at the resolution of individual cells [5] . Unlike bulk transcriptomic analyses, which average signals across heterogeneous cell populations, single-cell RNA sequencing (scRNA-seq) captures cell-type– specific transcriptional states, uncovering rare cell populations, dynamic regulatory programs, and disease-associated cellular subtypes [6] . Over the last several years, these approaches have been increasingly applied to human tissues relevant to T2DM, including pancreatic islets, immune cells, adipose tissue, liver, and skeletal muscle. Such studies have revealed substantial cellular heterogeneity and context-dependent transcriptional alterations that are not detectable using conventional methods. Importantly, single-cell transcriptomics offers a unique framework to investigate gene– environment interactions at cellular resolution. Genetic variants may exert differential regulatory effects depending on cell identity, metabolic state, or exposure to environmental stressors such as hyperglycaemia, lipotoxicity, or chronic inflammation [7] . By linking genotype information to cell-specific transcriptional profiles, it becomes possible to delineate how environmental factors modulate the functional impact of genetic risk within defined cellular contexts. Parallel to advances in transcriptomics, the development of polygenic risk scores (PRS) has enabled quantitative estimation of cumulative genetic liability for T2DM [8] . PRS integrate information from thousands of genetic variants identified through GWAS, providing a continuous measure of inherited risk. Studies have demonstrated that the predictive performance of PRS varies across populations and environmental conditions, reinforcing the concept that genetic susceptibility does not operate independently of external exposures [9] . Widely adopted analytical tools, such as PLINK, have facilitated standardized workflows for GWAS analysis, genotype quality control, population stratification adjustment, and PRS calculation, thereby enhancing reproducibility across human studies. More recently, integrative analytical strategies combining single-cell transcriptomics, genetic risk profiling, environmental data, and machine-learning–based models have emerged. These multiomics approaches enable high-dimensional data integration, identification of regulatory networks, and improved detection of non-linear gene–environment interactions. Machine learning methods, in particular, have shown promise in uncovering complex patterns that link genetic variation, transcriptional regulation, and disease phenotypes, thereby advancing precision medicine frameworks in T2DM research [10] . In this systematic literature review and meta-analysis, we synthesize contemporary human studies published between 2021 and 2025 that employ single-cell transcriptomic technologies to investigate gene–environment interactions in T2DM. Specifically, this review aims to: (1) characterize cell-type–specific transcriptional alterations associated with T2DM; (2) evaluate how GWAS and PRS are integrated with single-cell data to localize genetic risk within cellular populations; (3) assess the influence of environmental and clinical exposures on genotype–expression relationships; and (4) examine the role of machine learning and multiomics integration in advancing mechanistic understanding and precision medicine applications. By consolidating evidence across these domains, this review seeks to clarify how genetic and environmental factors converge at the cellular level to drive T2DM pathogenesis. Materials and Methods 1. Objective and Scope The primary objective of this systematic review and meta-analysis was to identify, evaluate, and synthesize human research published from January 1, 2021 to December 31, 2025 that investigates gene–environment interactions (G×E) in Type 2 diabetes mellitus (T2DM) using single-cell transcriptomics . Secondary goals included assessment of studies integrating GWAS and Polygenic Risk Scores (PRS) with single-cell data, leveraging PLINK software for genetic analyses, and highlighting multi-dimensional analytic frameworks such as machine learning , multi-omics integration , and gene regulatory network reconstruction . 2. Eligibility Criteria 2.1. Study Design • Original research articles, cohort studies, case–control studies, cross-sectional analyses, and clinical observational studies. • Meta-analyses and systematic reviews were surveyed to identify additional primary research but not included in quantitative synthesis. • Preprints were considered only if subsequently published or peer-reviewed by December 31, 2025. 2.2. Population • Human subjects diagnosed with T2DM. • Studies exclusively using rodent or other non-human models were excluded. 2.3. Technologies and Techniques Single-cell transcriptomic approaches (e.g., scRNA-seq, single-cell multi-omics) applied to human tissues relevant to T2DM (e.g., pancreatic islets, adipose, liver, muscle, blood immune cells). • Studies that combined genetic association data (e.g., GWAS, PRS) with single-cell resolution. 2.4. Gene–Environment Interaction • Explicit or implicit investigation of environmental exposures (diet, lifestyle, pollutants, medication response) in interaction with genetic variation at the single-cell level. • Inclusion of analyses linking genotype with transcriptomic variation conditioned on environmental factors. 2.5. Language and Time Frame • Publications in English from 2021 to 2025. 3. Information Sources and Search Strategy 3.1. Databases Systematic searches were conducted in the following electronic databases: • PubMed/MEDLINE • EMBASE • Web of Science (Core Collection) • Scopus • Google Scholar (for supplementary grey literature and citation chasing) 3.2. Search Terms and Boolean Strategy Search strings were developed using combinations of controlled vocabulary (e.g., MeSH) and keywords related to: • Single-Cell Transcriptomics (e.g., “single-cell RNA-seq”, “scRNA-seq”, “single cell transcriptomics”) • Type 2 Diabetes (e.g., “T2D”, “Type 2 Diabetes Mellitus”) • Gene-Environment Interaction (e.g., “gene–environment”, “G×E”, “environmental exposure”) • Genetic Analysis Tools (e.g., “GWAS”, “PLINK”, “polygenic risk score”, “PRS”) • Analytic Frameworks (e.g., “machine learning”, “multi-omics”, “gene regulatory networks”, “cellular heterogeneity”, “precision medicine”) Example combined Boolean search (adapted per database syntax): (“Type 2 diabetes” OR “T2DM”) AND (“single-cell transcriptomics” OR “scRNA-seq” OR “single cell”) AND (“gene-environment interaction” OR “GxE” OR “environment* expos*”) AND (“GWAS” OR “PLINK” OR “polygenic risk score” OR “PRS”) AND (“2021/01/01”[Date – Publication] : “2025/12/31”[Date – Publication]) 3.3. Additional Search Techniques • Citation tracking of key articles. • Manual review of references in review articles and high-impact publications. Search alerts for emerging 2025 publications during manuscript preparation. 4. Study Selection and Screening 4.1. Deduplication All retrieved records were exported into a reference manager (e.g., EndNote, Zotero) and deduplicated. 4.2. Screening Procedure • Title and Abstract Screening: Two independent reviewers screened all titles and abstracts. • Full-Text Assessment: Full texts of potentially eligible studies were retrieved and reviewed against the inclusion/exclusion criteria. 4.3. Discrepancy Resolution Conflicts were resolved by consensus or, when necessary, a third reviewer adjudicated disagreements. 5. Data Extraction and Management 5.1. Extracted Variables For each study meeting eligibility: • Study Characteristics: Authors, year, country, cohort size, clinical characteristics. • Single-Cell Methods: Technology platform (e.g., 10× Genomics), tissue/cell type, quality control metrics. Genetic Data: Source of GWAS data, imputation reference, variant filtering, population stratification adjustments. • PLINK Usage: Specific analyses (e.g., genotype QC, GWAS, PRS computation). • Environmental Factors: Lifestyle, diet, exposures, clinical variables. • Outcomes: Expression signatures, cell clusters implicated, G×E interactions, effect sizes, statistical significance. • Analytic Tools: Machine learning frameworks, network inference tools, multi-omics integration pipelines. • Key Findings and Limitations 5.2. Data Organization Data were entered into standardized spreadsheets and a relational database to facilitate synthesis and meta-analysis. 6. Quality Assessment Each included study was evaluated for: • Methodological Rigor: Appropriateness of sequencing depth, cell filtering, and normalization. • Genetic Analysis Quality: Adequacy of GWAS design, population structure control, PRS validation. • Reporting of Environmental Variables: Clarity and quantification of exposures. • Reproducibility: Transparency of analytic pipelines, code availability. Quality was graded using modified tools suited for multi-omics and single-cell studies (e.g., adapted STROBE and QUADAS-2 frameworks). 7. Meta-Analysis Methods 7.1. Quantitative Synthesis Criteria Studies were pooled if they: • Reported comparable effect measures (e.g., odds ratios, expression fold changes stratified by genotype/environment). • Included compatible cell types or shared analytical outcomes. 7.2. Statistical Aggregation • Random-effects models were used to account for heterogeneity. • Heterogeneity was quantified using I² statistics . 7.3. Sensitivity and Subgroup Analyses • By environmental exposure category (e.g., diet, smoking). • By tissue type (e.g., pancreatic vs peripheral). • By genetic risk strata (high vs low PRS groups). 7.4. Publication Bias Evaluated via funnel plots and Egger’s regression tests where appropriate. 8. Analytical Tools and Software 8.1. Genetic Analysis • PLINK (v1.9 or later) was used for genotype quality control, imputation checks, GWAS, and Polygenic Risk Score calculations. • Standard PLINK workflows ensured: o Removal of low-quality variants (e.g., call rate thresholds). o Population stratification adjustment (e.g., principal component inclusion). o PRS derivation across multiple thresholds and validation sets. 8.2. Single-Cell Transcriptomics • Data processed through commonly accepted pipelines (e.g., Cell Ranger, Seurat, Scanpy). • Normalization, batch correction, and clustering adhered to best-practice guidelines to minimize technical biases. 8.3. Machine Learning and Multi-omics Integration • Algorithms (e.g., random forests, elastic net, deep learning models) applied where indicated to: o Predict phenotypes from integrated features. o Identify interaction signatures. • Tools for gene regulatory network inference (e.g., SCENIC, GRNBoost) supported network elucidation. 8.4. Statistical Software • R and Python environments for data analysis and visualization. • Meta-analysis computations performed with packages such as metafor (R). 9. Keywords for Retrieval and Indexing • Single-Cell Transcriptomics • Polygenic Risk Score • Machine Learning • Gene–Environment Interaction • Cellular Heterogeneity • Multi-omics Integration • Environmental Exposure • Gene Regulatory Networks • Precision Medicine • GWAS • PLINK 10. Reporting Standards This review followed the PRISMA 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, with a detailed flow diagram documenting screening and inclusion processes, and compliance with best practices for transparent reporting of multi-omics research. RESULTS A total of seven original studies published between 2021 and 2025 were included in the final synthesis. These studies collectively explored genetic, transcriptomic, immunological, and polygenic risk mechanisms associated with type 2 diabetes mellitus (T2DM) using advanced methodologies such as single-cell RNA sequencing, Mendelian randomization, and largescale biobank analyses. The results are presented in a structured manner, including study characteristics, risk of bias assessment, quantitative synthesis, heterogeneity analysis, and graphical representation of pooled effects. Risk of bias assessment demonstrated that all included studies were of overall low risk. This reflects robust study designs, well-defined populations, and transparent analytical pipelines. Quantitative Synthesis and Heterogeneity Analysis A meta-analytical framework was applied to synthesize the direction and magnitude of associations reported across studies. The pooled effect estimate demonstrated a significant association between genetic/transcriptomic dysregulation and T2DM-related phenotypes. Heterogeneity analysis revealed moderate between-study variability. The Cochran Q test indicated statistically significant heterogeneity (p < 0.05). The I² statistic was calculated at approximately 48%, suggesting moderate heterogeneity, likely attributable to differences in study populations, tissues analyzed, and analytical platforms. Figure 2. Forest Plot The forest plot illustrates individual and pooled effect estimates from the included studies. All studies demonstrated effect sizes greater than unity, indicating a consistent positive association between molecular dysregulation and T2DM-related outcomes. Figure 3. Funnel Plot The funnel plot demonstrates approximate symmetry around the pooled effect estimate, suggesting a low likelihood of publication bias among the included studies. DISCUSSION This review integrates recent human research conducted between 2021 and 2025 that applies single-cell transcriptomic technologies to explore how genetic susceptibility and environmental influences jointly contribute to Type 2 diabetes mellitus (T2DM). By aggregating evidence from seven independent studies that combine cell-resolved gene expression data with genomic risk measures and environmental context, this analysis advances current understanding of the biological pathways through which inherited risk is translated into disease-relevant cellular dysfunction. Disease Mechanisms Are Embedded in Cellular Subpopulations A defining theme across the reviewed literature is that molecular abnormalities associated with T2DM are concentrated within discrete cellular subgroups rather than distributed uniformly across tissues [ 11 ] . High-resolution transcriptomic profiling consistently demonstrated that disease-associated gene expression changes are confined to specific cell identities and activation states. In pancreatic tissue, insulin-producing cells displayed multiple, distinct transcriptional phenotypes reflecting differential metabolic stress adaptation, inflammatory signaling, and functional decline [ 12 ] . These observations indicate that β-cell impairment in T2DM reflects heterogeneity in cellular responses rather than a single pathological program. Parallel findings in immune compartments further reinforce this concept. Single-cell analyses of circulating and tissue-resident immune cells revealed selective transcriptional reprogramming in defined leukocyte subsets, implicating targeted immune dysregulation in metabolic disease [ 13 ] . Such findings support emerging models in which immune-mediated inflammation in T2DM originates from cell-specific perturbations shaped by both genetic background and metabolic environment. Functional Localization of Genetic Susceptibility The reviewed studies provide compelling evidence that polygenic risk for T2DM is preferentially concentrated within particular cellular populations [ 14 ] . When polygenic risk scores were integrated with single-cell expression data, genetically vulnerable cell types emerged as transcriptionally distinct entities exhibiting dysregulation of pathways central to glucose metabolism, insulin secretion, and immune signaling [ 15 ] . This selective enrichment of genetic risk clarifies the functional relevance of noncoding risk variants identified by GWAS, which often lack clear biological interpretation in the absence of cellular context [ 16 ] . The consistent use of standardized genomic analysis pipelines, including well-established GWAS and PRS methodologies, lends credibility to these findings. Importantly, the observed cell-specific manifestation of genetic risk aligns with population-level evidence showing that genetic predisposition is modulated by behavioral, metabolic, and environmental factors [ 17 ] . Together, these data suggest that genetic susceptibility operates through conditional, celldependent mechanisms rather than fixed deterministic pathways [ 18 ] . Cellular Context Defines Gene–Environment Interactions A key contribution of this review is the demonstration that gene–environment interactions in T2DM become apparent only when examined at single-cell resolution [ 19 – 20 ] . Environmental exposures—such as nutritional excess, obesity-related metabolic stress, and inflammatory stimuli—were shown to influence how genetic variants shape transcriptional outputs within specific cell populations [ 21 ] . In immune cells, environmental cues altered the magnitude and direction of genotype-associated expression patterns, while in pancreatic cells, metabolic conditions modified the expression of genes linked to insulin biosynthesis and cellular resilience [ 22 ] . The moderate variability detected across studies likely reflects authentic biological diversity arising from differences in tissue sources, genetic ancestry, disease stage, and environmental exposure rather than methodological shortcomings [ 23 ] . This heterogeneity underscores the necessity of cell-resolved analytical strategies to capture the complexity of G×E interactions in multifactorial diseases such as T2DM. Analytical Innovation through Integrative Modeling Several studies incorporated computational frameworks capable of integrating genetic, transcriptomic, and environmental data into unified analytical models [ 24 ] . Machine learning approaches proved particularly effective in identifying interaction patterns that are difficult to detect using conventional statistical techniques. These models improved risk prediction accuracy when compared with single-modality analyses and highlighted the value of highdimensional data integration for complex disease research [ 25 ] . In addition, network-based analyses revealed that diverse genetic and environmental influences converge on shared regulatory nodes governing metabolic homeostasis, immune activation, and cellular stress pathways [ 26 ] . Such insights move beyond descriptive associations, offering mechanistic hypotheses that may inform future therapeutic development and biomarker discovery. Implications for Precision Medicine The findings synthesized in this review have direct relevance for the development of personalized approaches to T2DM prevention and management [ 27 ] . The demonstration that genetic risk is expressed through defined cellular contexts suggests that future risk stratification models should incorporate cell-specific molecular information alongside traditional clinical and lifestyle factors. Single-cell–derived transcriptional signatures and cell-informed PRS frameworks may ultimately enable earlier detection of disease susceptibility and more precise targeting of interventions. Moreover, identifying how modifiable environmental factors interact with genetically sensitive cell populations provides a mechanistic foundation for individualized lifestyle and pharmacological strategies. Such approaches hold promise for mitigating disease risk even among individuals with substantial inherited susceptibility [ 28 ] . Study Constraints Despite these advances, several limitations must be acknowledged. The relatively small number of eligible human studies reflects the technical complexity and resource demands of single-cell multiomics research. Differences in sequencing platforms, tissue sampling strategies, analytical workflows, and exposure assessment methods may contribute to residual variability. Furthermore, the predominance of cross-sectional designs limits insight into the temporal evolution of cellular G×E interactions. Population representation across studies was uneven, raising concerns about the transferability of PRS-based conclusions across diverse ancestries. Environmental exposures were often inferred rather than directly quantified, highlighting an important area for methodological improvement. Directions for Future Research Future research should emphasize longitudinal designs that integrate repeated single-cell profiling with comprehensive environmental and lifestyle measurements. Expanding representation across diverse populations and harmonizing analytical standards will be essential for robust inference. Advances in spatial transcriptomics and multi-modal single-cell technologies are expected to further illuminate how microenvironmental cues influence genetically susceptible cells within intact tissues. References 1. Mahajan A, Taliun D, Thurner M, Robertson NR, Torres JM, Rayner NW, et al. Finemapping type 2 diabetes loci to single-variant resolution using high-density imputation and islet-specific epigenome maps. Nat Genet . 2018;50(11):1505–13. 2. Scott RA, Scott LJ, Mägi R, Marullo L, Gaulton KJ, Kaakinen M, et al. An expanded genome-wide association study of type 2 diabetes in Europeans. Diabetes . 2017;66(11):2888–902. 3. Voight BF, Scott LJ, Steinthorsdottir V, Morris AP, Dina C, Welch RP, et al. Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis. Nat Genet . 2010;42(7):579–89. 4. Segerstolpe Å, Palasantza A, Eliasson P, Andersson AC, Andréasson AC, Sun X, et al. Single-cell transcriptome profiling of human pancreatic islets in health and type 2 diabetes. Cell Metab . 2016;24(4):593–607. 5. Baron M, Veres A, Wolock SL, Faust AL, Gaujoux R, Vetere A, et al. A single-cell transcriptomic map of the human and mouse pancreas reveals inter- and intra-cell population structure. Cell Syst . 2016;3(4):346–60. 6. Muraro MJ, Dharmadhikari G, Grün D, Groen N, Dielen T, Jansen E, et al. A singlecell transcriptome atlas of the human pancreas. Cell Syst . 2016;3(4):385–94. 7. Lawlor N, George J, Bolisetty M, Kursawe R, Sun L, Sivakamasundari V, et al. Single-cell transcriptomes identify human islet cell signatures and reveal cell-type– specific expression changes in type 2 diabetes. Genome Res . 2017;27(2):208–22. 8. Fadista J, Vikman P, Laakso EO, Mollet IG, Esguerra JL, Taneera J, et al. Global genomic and transcriptomic analysis of human pancreatic islets reveals novel genes influencing glucose metabolism. Proc Natl Acad Sci U S A . 2014;111(38):13924–9. 9. Khera AV, Chaffin M, Aragam KG, Haas ME, Roselli C, Choi SH, et al. Genomewide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. Nat Genet . 2018;50(9):1219–24. 10. Mars N, Koskela JT, Ripatti P, Kiiskinen T, Havulinna AS, Lindbohm JV, et al. Polygenic and clinical risk scores and their impact on age at onset and prediction of cardiometabolic diseases. Nat Med . 2020;26(4):549–57. 11. Duncan L, Shen H, Gelaye B, Meijsen J, Ressler K, Feldman M, et al. Analysis of polygenic risk score usage and performance in diverse human populations. Nat Commun . 2019;10(1):3328. 12. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. GigaScience . 2015;4:7. 13. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet . 2007;81(3):559–75. 14. Finucane HK, Reshef YA, Anttila V, Slowikowski K, Gusev A, Byrnes A, et al. Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types. Nat Genet . 2018;50(4):621–9. 15. Zhang Y, Hu Y, Wang J, Liu J, Chen L, Zhang S, et al. Cell-type-specific polygenic risk score analysis reveals cellular mechanisms of complex traits. Nat Genet . 2025;57(2):231–42. 16. Argelaguet R, Velten B, Arnol D, Dietrich S, Zenz T, Marioni JC, et al. Multi-omics factor analysis—a framework for unsupervised integration of multi-omics data sets. Mol Syst Biol . 2018;14(6):e8124. 17. Zhou W, Sailani MR, Contrepois K, Zhou Y, Ahadi S, Leopold SR, et al. Longitudinal multi-omics of host–microbe dynamics in prediabetes. Nature . 2019;569(7758):663– 71. 18. Libbrecht MW, Noble WS. Machine learning applications in genetics and genomics. Nat Rev Genet . 2015;16(6):321–32. 19. Moore JH, Williams SM. Epistasis and its implications for personal genetics. Am J Hum Genet . 2009;85(3):309–20. 20. Franks PW, McCarthy MI. Exposing the exposures responsible for type 2 diabetes and obesity. Science . 2016;354(6308):69–73. 21. Martínez-López JA, Lindqvist A, Lopez-Pascual A, et al. Single-cell mRNAregulation analysis reveals cell type-specific mechanisms of type 2 diabetes. Nat Commun. 2025;16:9475. doi:10.1038/s41467-025-65060-z 22. Liu Y, Wang T, Wu R, et al. Single-cell transcriptome atlas and genome-wide Mendelian randomization reveal chemokine involvement in diverse immune cells in type 2 diabetes. Int J Obes (Lond). 2025;49:2213-2228. doi:10.1038/s41366-025-01846 23. Xiao M, Xu Z, Zhu X, Chen J, Wang R, Wang Y, Gao X, Luan S, Pu X. Immunological profiling in type 2 diabetes mellitus and type 2 diabetic kidney disease: insights from single-cell LacNAc sequencing. Front Endocrinol (Lausanne). 2025;16:1550925. doi:10.3389/fendo.2025.1550925 24. Zhao J, Fang Z. Single-cell RNA sequencing reveals the dysfunctional characteristics of PBMCs in patients with type 2 diabetes mellitus. Front Immunol. 2024;15:1501660. doi:10.3389/fimmu.2024.1501660 25. Grenko CM, Taylor HJ, Bonnycastle LL, Xue D, Lee BN, Weiss Z, Yan T, Swift AJ, Mansell EC, Lee A, Robertson CC, Narisu N, Erdos MR, Chen S, Collins FS, Taylor DL. Single-cell transcriptomic profiling of human pancreatic islets reveals genes responsive to glucose exposure over 24 h. Diabetologia. 2024;67(10):2246-2259. doi:10.1007/s00125-024-06214-4 26. Yun JS, Jung SH, Shivakumar M, Xiao B, Khera AV, Won HH, Kim D. Polygenic risk for type 2 diabetes, lifestyle, metabolic health, and cardiovascular disease: a prospective UK Biobank study. Cardiovasc Diabetol. 2022;21(1):131. doi:10.1186/s12933-022-01560-2 27. Zhang S, Shu H, Zhou J, Rubin-Sigler J, Yang X, Liu Y, Cooper-Knock J, Monte E, Zhu C, Tu S, Li H, Tong M, Ecker JR, Ichida JK, Shen Y, Zeng J, Tsao PS, Snyder MP. Single-cell polygenic risk scores dissect cellular and molecular heterogeneity of complex human diseases. Nat Biotechnol. 2025 Jul 25. doi:10.1038/s41587-025-02725-6 28. Zhang S, Shu H, Zhou J, Rubin-Sigler J, Yang X, Liu Y, Cooper-Knock J, Monte E, Zhu C, Tu S, Li H, Tong M, Ecker JR, Ichida JK, Shen Y, Zeng J, Tsao PS, Snyder MP. Deconvolution of polygenic risk score in single cells unravels cellular and molecular heterogeneity of complex human diseases. bioRxiv [Preprint]. 2024 May 14:2024.05.14.594252. doi:10.1101/2024.05.14.594252 Tables Tables 1 and 2 are available in the supplementary files section Additional Declarations The authors declare no competing interests. Supplementary Files Tables.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9283816","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":615493243,"identity":"914f0474-2a19-4aec-a5ed-9f419951b4f8","order_by":0,"name":"Rupanjali Singh","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6ElEQVRIie3QMQrCMBSA4RcKdnnSNVLBEwiK4FTpVSyFuDoqOMSlXQquBS/hEVICdekBdFIoODuJk5jq5JLWTTD/GN5HXgJgMv1iNmcWAEXHXnMBWB0RricoLop43U6ScSEaERpUhHmDwxRANNmrz0NWzhcS4Viu5bULPYdb0VlHxiKfjdJCItkGr8WGqSDxQEuyhLntSKLlvgnZAYmolkjn5rYfElud9/P9epKDuoUzREpeJKgnBai35B5SVIsVSMNU1pHTWf3Yivr+fl9el4k32cTxRUs+I4katprPV92/GzeZTKb/6AnufU3k6iwqxQAAAABJRU5ErkJggg==","orcid":"","institution":"Indian Institute of Technology Jodhpur","correspondingAuthor":true,"prefix":"","firstName":"Rupanjali","middleName":"","lastName":"Singh","suffix":""}],"badges":[],"createdAt":"2026-03-31 19:33:12","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9283816/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9283816/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":106093329,"identity":"e81fe000-bbba-4cf7-9955-63ac8de8b234","added_by":"auto","created_at":"2026-04-03 11:36:49","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":141182,"visible":true,"origin":"","legend":"\u003cp\u003eFigure legend not provided with this version\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9283816/v1/7001c915e4131b8c86393edb.png"},{"id":105982352,"identity":"5d2dedfa-5e19-4cbb-9574-edcd2706c4ab","added_by":"auto","created_at":"2026-04-02 06:58:53","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":123512,"visible":true,"origin":"","legend":"\u003cp\u003eForest Plot\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9283816/v1/edf7cdfebdf788c62cb4c2c4.png"},{"id":105982353,"identity":"c6beeba7-8e18-48af-9a05-e3f0e4df11dc","added_by":"auto","created_at":"2026-04-02 06:58:53","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":71519,"visible":true,"origin":"","legend":"\u003cp\u003eFunnel Plot\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9283816/v1/c86dfe312b80119844a671c3.png"},{"id":106096118,"identity":"086c9b43-38e3-4c7a-ba78-96e04504b46e","added_by":"auto","created_at":"2026-04-03 11:52:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1312678,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9283816/v1/29b5e785-0e80-45f6-a26a-30ac29126051.pdf"},{"id":105982350,"identity":"89a98f74-4b5d-4b8a-9fda-5e19478f59f6","added_by":"auto","created_at":"2026-04-02 06:58:53","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":15822,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-9283816/v1/9d28efd6dad6685fc23e82c3.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eGene-Environment Interaction through single-cell transcriptomics in Type 2 diabetes\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eType 2 diabetes mellitus (T2DM) is a chronic metabolic disorder characterized by persistent hyperglycaemia resulting from a combination of insulin resistance and progressive pancreatic \u0026beta;-cell dysfunction\u003csup\u003e[1]\u003c/sup\u003e. The development and progression of T2DM are driven by a complex interplay between inherited genetic susceptibility and a wide range of environmental influences, including diet, physical activity, obesity, metabolic stress, and inflammatory exposures\u003csup\u003e[2]\u003c/sup\u003e. Despite substantial advances in understanding disease epidemiology and clinical management, the biological mechanisms linking genetic risk to cellular dysfunction in specific tissues remain incompletely defined.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLarge-scale genome-wide association studies (GWAS) have successfully identified numerous genetic loci associated with T2DM risk. However, the majority of these variants reside in non-coding regions of the genome, limiting direct interpretation of their functional consequences\u003csup\u003e[3]\u003c/sup\u003e. As a result, traditional genetic association approaches provide limited insight into how risk variants influence gene regulation within specific cell types or how these effects are modified by environmental exposures\u003csup\u003e[4]\u003c/sup\u003e. This gap has hindered the translation of genetic discoveries into mechanistic understanding and precision therapeutic strategies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRecent technological advances in single-cell transcriptomics have transformed the study of complex metabolic diseases by enabling gene expression profiling at the resolution of individual cells\u003csup\u003e[5]\u003c/sup\u003e. Unlike bulk transcriptomic analyses, which average signals across heterogeneous cell populations, single-cell RNA sequencing (scRNA-seq) captures cell-type\u0026ndash; specific transcriptional states, uncovering rare cell populations, dynamic regulatory programs, and disease-associated cellular subtypes\u003csup\u003e[6]\u003c/sup\u003e. Over the last several years, these approaches have been increasingly applied to human tissues relevant to T2DM, including pancreatic islets, immune cells, adipose tissue, liver, and skeletal muscle. Such studies have revealed substantial cellular heterogeneity and context-dependent transcriptional alterations that are not detectable using conventional methods.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eImportantly, single-cell transcriptomics offers a unique framework to investigate gene\u0026ndash; environment interactions at cellular resolution. Genetic variants may exert differential regulatory effects depending on cell identity, metabolic state, or exposure to environmental stressors such as hyperglycaemia, lipotoxicity, or chronic inflammation\u003csup\u003e[7]\u003c/sup\u003e. By linking genotype information to cell-specific transcriptional profiles, it becomes possible to delineate how environmental factors modulate the functional impact of genetic risk within defined cellular contexts.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eParallel to advances in transcriptomics, the development of polygenic risk scores (PRS) has enabled quantitative estimation of cumulative genetic liability for T2DM\u003csup\u003e[8]\u003c/sup\u003e. PRS integrate information from thousands of genetic variants identified through GWAS, providing a continuous measure of inherited risk. Studies have demonstrated that the predictive performance of PRS varies across populations and environmental conditions, reinforcing the concept that genetic susceptibility does not operate independently of external exposures\u003csup\u003e[9]\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWidely adopted analytical tools, such as PLINK, have facilitated standardized workflows for GWAS analysis, genotype quality control, population stratification adjustment, and PRS calculation, thereby enhancing reproducibility across human studies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMore recently, integrative analytical strategies combining single-cell transcriptomics, genetic risk profiling, environmental data, and machine-learning\u0026ndash;based models have emerged. These multiomics approaches enable high-dimensional data integration, identification of regulatory networks, and improved detection of non-linear gene\u0026ndash;environment interactions. Machine learning methods, in particular, have shown promise in uncovering complex patterns that link genetic variation, transcriptional regulation, and disease phenotypes, thereby advancing precision medicine frameworks in T2DM research\u003csup\u003e[10]\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn this systematic literature review and meta-analysis, we synthesize contemporary human studies published between 2021 and 2025 that employ single-cell transcriptomic technologies to investigate gene\u0026ndash;environment interactions in T2DM. Specifically, this review aims to: (1) characterize cell-type\u0026ndash;specific transcriptional alterations associated with T2DM; (2) evaluate how GWAS and PRS are integrated with single-cell data to localize genetic risk within cellular populations; (3) assess the influence of environmental and clinical exposures on genotype\u0026ndash;expression relationships; and (4) examine the role of machine learning and multiomics integration in advancing mechanistic understanding and precision medicine applications. By consolidating evidence across these domains, this review seeks to clarify how genetic and environmental factors converge at the cellular level to drive T2DM pathogenesis.\u0026nbsp;\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003e1. Objective and Scope\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe primary objective of this systematic review and meta-analysis was to identify, evaluate, and synthesize human research published from \u003cstrong\u003eJanuary 1, 2021 to December 31, 2025\u003c/strong\u003e that investigates \u003cstrong\u003egene\u0026ndash;environment interactions (G\u0026times;E)\u003c/strong\u003e in \u003cstrong\u003eType 2 diabetes mellitus (T2DM)\u003c/strong\u003e using \u003cstrong\u003esingle-cell transcriptomics\u003c/strong\u003e. Secondary goals included assessment of studies integrating \u003cstrong\u003eGWAS\u003c/strong\u003e and \u003cstrong\u003ePolygenic Risk Scores (PRS)\u003c/strong\u003e with single-cell data, leveraging \u003cstrong\u003ePLINK\u003c/strong\u003e software for genetic analyses, and highlighting multi-dimensional analytic frameworks such as \u003cstrong\u003emachine learning\u003c/strong\u003e, \u003cstrong\u003emulti-omics integration\u003c/strong\u003e, and \u003cstrong\u003egene regulatory network reconstruction\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e2. Eligibility Criteria\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.1. Study Design\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Original research articles, cohort studies, case\u0026ndash;control studies, cross-sectional analyses, and clinical observational studies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Meta-analyses and systematic reviews were surveyed to identify additional primary research but not included in quantitative synthesis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Preprints were considered only if subsequently published or peer-reviewed by December 31, 2025.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.2. Population\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Human subjects diagnosed with T2DM.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Studies exclusively using rodent or other non-human models were excluded.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.3. Technologies and Techniques\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSingle-cell transcriptomic approaches (e.g., scRNA-seq, single-cell multi-omics) applied to human tissues relevant to T2DM (e.g., pancreatic islets, adipose, liver, muscle, blood immune cells).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u0026nbsp;Studies that combined genetic association data (e.g., GWAS, PRS) with single-cell resolution.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2.4. Gene\u0026ndash;Environment Interaction\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Explicit or implicit investigation of environmental exposures (diet, lifestyle, pollutants, medication response) in interaction with genetic variation at the single-cell level.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Inclusion of analyses linking genotype with transcriptomic variation conditioned on environmental factors.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5. Language and Time Frame\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026bull; Publications in English from 2021 to 2025.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3. Information Sources and Search Strategy\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3.1. Databases\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSystematic searches were conducted in the following electronic databases:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; PubMed/MEDLINE\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; EMBASE\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Web of Science (Core Collection)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Scopus\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Google Scholar (for supplementary grey literature and citation chasing)\u003c/p\u003e\n\u003cp\u003e3.2. Search Terms and Boolean Strategy\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSearch strings were developed using combinations of controlled vocabulary (e.g., MeSH) and keywords related to:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eSingle-Cell Transcriptomics\u003c/strong\u003e (e.g., \u0026ldquo;single-cell RNA-seq\u0026rdquo;, \u0026ldquo;scRNA-seq\u0026rdquo;, \u0026ldquo;single cell transcriptomics\u0026rdquo;)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eType 2 Diabetes\u003c/strong\u003e (e.g., \u0026ldquo;T2D\u0026rdquo;, \u0026ldquo;Type 2 Diabetes Mellitus\u0026rdquo;)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGene-Environment Interaction\u003c/strong\u003e (e.g., \u0026ldquo;gene\u0026ndash;environment\u0026rdquo;, \u0026ldquo;G\u0026times;E\u0026rdquo;, \u0026ldquo;environmental exposure\u0026rdquo;)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGenetic Analysis Tools\u003c/strong\u003e (e.g., \u0026ldquo;GWAS\u0026rdquo;, \u0026ldquo;PLINK\u0026rdquo;, \u0026ldquo;polygenic risk score\u0026rdquo;, \u0026ldquo;PRS\u0026rdquo;)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eAnalytic Frameworks\u003c/strong\u003e (e.g., \u0026ldquo;machine learning\u0026rdquo;, \u0026ldquo;multi-omics\u0026rdquo;, \u0026ldquo;gene regulatory networks\u0026rdquo;, \u0026ldquo;cellular heterogeneity\u0026rdquo;, \u0026ldquo;precision medicine\u0026rdquo;)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExample combined Boolean search (adapted per database syntax):\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(\u0026ldquo;Type 2 diabetes\u0026rdquo; OR \u0026ldquo;T2DM\u0026rdquo;) AND \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(\u0026ldquo;single-cell transcriptomics\u0026rdquo; OR \u0026ldquo;scRNA-seq\u0026rdquo; OR \u0026ldquo;single cell\u0026rdquo;) AND \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(\u0026ldquo;gene-environment interaction\u0026rdquo; OR \u0026ldquo;GxE\u0026rdquo; OR \u0026ldquo;environment* expos*\u0026rdquo;) AND \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(\u0026ldquo;GWAS\u0026rdquo; OR \u0026ldquo;PLINK\u0026rdquo; OR \u0026ldquo;polygenic risk score\u0026rdquo; OR \u0026ldquo;PRS\u0026rdquo;) AND \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e(\u0026ldquo;2021/01/01\u0026rdquo;[Date \u0026ndash; Publication] : \u0026ldquo;2025/12/31\u0026rdquo;[Date \u0026ndash; Publication])\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3.3. Additional Search Techniques\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eCitation tracking\u003c/strong\u003e of key articles.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eManual review\u003c/strong\u003e of references in review articles and high-impact publications.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSearch alerts\u003c/strong\u003e for emerging 2025 publications during manuscript preparation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4. Study Selection and Screening\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4.1. Deduplication\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll retrieved records were exported into a reference manager (e.g., EndNote, Zotero) and deduplicated.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4.2. Screening Procedure\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eTitle and Abstract Screening:\u003c/strong\u003e Two independent reviewers screened all titles and abstracts.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eFull-Text Assessment:\u003c/strong\u003e Full texts of potentially eligible studies were retrieved and reviewed against the inclusion/exclusion criteria.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4.3. Discrepancy Resolution\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConflicts were resolved by consensus or, when necessary, a third reviewer adjudicated disagreements.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5. Data Extraction and Management\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5.1. Extracted Variables\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor each study meeting eligibility:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eStudy Characteristics:\u003c/strong\u003e Authors, year, country, cohort size, clinical characteristics.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eSingle-Cell Methods:\u003c/strong\u003e Technology platform (e.g., 10\u0026times; Genomics), tissue/cell type, quality control metrics.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGenetic Data:\u003c/strong\u003e Source of GWAS data, imputation reference, variant filtering, population stratification adjustments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003ePLINK Usage:\u003c/strong\u003e Specific analyses (e.g., genotype QC, GWAS, PRS computation).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eEnvironmental Factors:\u003c/strong\u003e Lifestyle, diet, exposures, clinical variables.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eOutcomes:\u003c/strong\u003e Expression signatures, cell clusters implicated, G\u0026times;E interactions, effect sizes, statistical significance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eAnalytic Tools:\u003c/strong\u003e Machine learning frameworks, network inference tools, multi-omics integration pipelines.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull;\u0026nbsp; Key Findings and Limitations\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5.2. Data Organization\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData were entered into standardized spreadsheets and a relational database to facilitate synthesis and meta-analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e6. Quality Assessment\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEach included study was evaluated for:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eMethodological Rigor:\u003c/strong\u003e Appropriateness of sequencing depth, cell filtering, and normalization.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGenetic Analysis Quality:\u003c/strong\u003e Adequacy of GWAS design, population structure control,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePRS validation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eReporting of Environmental Variables:\u003c/strong\u003e Clarity and quantification of exposures.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eReproducibility:\u003c/strong\u003e Transparency of analytic pipelines, code availability.\u003c/p\u003e\n\u003cp\u003eQuality was graded using modified tools suited for multi-omics and single-cell studies (e.g., adapted STROBE and QUADAS-2 frameworks).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e7. Meta-Analysis Methods\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e7.1. Quantitative Synthesis Criteria\u0026nbsp;\u003c/strong\u003eStudies were pooled if they:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Reported comparable effect measures (e.g., odds ratios, expression fold changes stratified by genotype/environment).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Included compatible cell types or shared analytical outcomes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e7.2. Statistical Aggregation\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eRandom-effects models\u003c/strong\u003e were used to account for heterogeneity.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Heterogeneity was quantified using \u003cstrong\u003eI\u0026sup2; statistics\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e7.3. Sensitivity and Subgroup Analyses\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; By environmental exposure category (e.g., diet, smoking).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; By tissue type (e.g., pancreatic vs peripheral).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; By genetic risk strata (high vs low PRS groups).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e7.4. Publication Bias\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEvaluated via funnel plots and Egger\u0026rsquo;s regression tests where appropriate.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8. Analytical Tools and Software\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.1. Genetic Analysis\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003ePLINK (v1.9 or later)\u003c/strong\u003e was used for genotype quality control, imputation checks,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGWAS, and \u003cstrong\u003ePolygenic Risk Score\u003c/strong\u003e calculations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Standard PLINK workflows ensured:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eo Removal of low-quality variants (e.g., call rate thresholds).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eo Population stratification adjustment (e.g., principal component inclusion).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eo PRS derivation across multiple thresholds and validation sets.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.2. Single-Cell Transcriptomics\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Data processed through commonly accepted pipelines (e.g., Cell Ranger, Seurat,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eScanpy).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Normalization, batch correction, and clustering adhered to best-practice guidelines to minimize technical biases.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.3. Machine Learning and Multi-omics Integration\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Algorithms (e.g., random forests, elastic net, deep learning models) applied where indicated to:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eo Predict phenotypes from integrated features.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eo Identify interaction signatures.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Tools for gene regulatory network inference (e.g., SCENIC, GRNBoost) supported network elucidation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.4. Statistical Software\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eR\u003c/strong\u003e and \u003cstrong\u003ePython\u003c/strong\u003e environments for data analysis and visualization.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; Meta-analysis computations performed with packages such as \u003cem\u003emetafor\u003c/em\u003e (R).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e9. Keywords for Retrieval and Indexing\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eSingle-Cell Transcriptomics\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003ePolygenic Risk Score\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eMachine Learning\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGene\u0026ndash;Environment Interaction\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eCellular Heterogeneity\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eMulti-omics Integration\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eEnvironmental Exposure\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGene Regulatory Networks\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003ePrecision Medicine\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003eGWAS\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026bull; \u003cstrong\u003ePLINK\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e10. Reporting Standards\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis review followed the \u003cstrong\u003ePRISMA 2020\u003c/strong\u003e (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, with a detailed flow diagram documenting screening and inclusion processes, and compliance with best practices for transparent reporting of multi-omics research.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eA total of seven original studies published between 2021 and 2025 were included in the final synthesis. These studies collectively explored genetic, transcriptomic, immunological, and polygenic risk mechanisms associated with type 2 diabetes mellitus (T2DM) using advanced methodologies such as single-cell RNA sequencing, Mendelian randomization, and largescale biobank analyses. The results are presented in a structured manner, including study characteristics, risk of bias assessment, quantitative synthesis, heterogeneity analysis, and graphical representation of pooled effects.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRisk of bias assessment demonstrated that all included studies were of overall low risk. This reflects robust study designs, well-defined populations, and transparent analytical pipelines.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eQuantitative Synthesis and Heterogeneity Analysis\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA meta-analytical framework was applied to synthesize the direction and magnitude of associations reported across studies. The pooled effect estimate demonstrated a significant association between genetic/transcriptomic dysregulation and T2DM-related phenotypes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHeterogeneity analysis revealed moderate between-study variability.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Cochran Q test indicated statistically significant heterogeneity (p \u0026lt; 0.05). The I\u0026sup2; statistic was calculated at approximately 48%, suggesting moderate heterogeneity, likely attributable to differences in study populations, tissues analyzed, and analytical platforms. Figure 2. Forest Plot\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe forest plot illustrates individual and pooled effect estimates from the included studies. All studies demonstrated effect sizes greater than unity, indicating a consistent positive association between molecular dysregulation and T2DM-related outcomes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFigure 3. Funnel Plot\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe funnel plot demonstrates approximate symmetry around the pooled effect estimate, suggesting a low likelihood of publication bias among the included studies.\u0026nbsp;\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis review integrates recent human research conducted between 2021 and 2025 that applies single-cell transcriptomic technologies to explore how genetic susceptibility and environmental influences jointly contribute to Type 2 diabetes mellitus (T2DM). By aggregating evidence from seven independent studies that combine cell-resolved gene expression data with genomic risk measures and environmental context, this analysis advances current understanding of the biological pathways through which inherited risk is translated into disease-relevant cellular dysfunction.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003eDisease Mechanisms Are Embedded in Cellular Subpopulations\u003c/h2\u003e \u003cp\u003eA defining theme across the reviewed literature is that molecular abnormalities associated with T2DM are concentrated within discrete cellular subgroups rather than distributed uniformly across tissues\u003csup\u003e[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e. High-resolution transcriptomic profiling consistently demonstrated that disease-associated gene expression changes are confined to specific cell identities and activation states. In pancreatic tissue, insulin-producing cells displayed multiple, distinct transcriptional phenotypes reflecting differential metabolic stress adaptation, inflammatory signaling, and functional decline\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e. These observations indicate that β-cell impairment in T2DM reflects heterogeneity in cellular responses rather than a single pathological program.\u003c/p\u003e \u003cp\u003eParallel findings in immune compartments further reinforce this concept. Single-cell analyses of circulating and tissue-resident immune cells revealed selective transcriptional reprogramming in defined leukocyte subsets, implicating targeted immune dysregulation in metabolic disease\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e. Such findings support emerging models in which immune-mediated inflammation in T2DM originates from cell-specific perturbations shaped by both genetic background and metabolic environment.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003eFunctional Localization of Genetic Susceptibility\u003c/h2\u003e \u003cp\u003eThe reviewed studies provide compelling evidence that polygenic risk for T2DM is preferentially concentrated within particular cellular populations\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. When polygenic risk scores were integrated with single-cell expression data, genetically vulnerable cell types emerged as transcriptionally distinct entities exhibiting dysregulation of pathways central to glucose metabolism, insulin secretion, and immune signaling\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e. This selective enrichment of genetic risk clarifies the functional relevance of noncoding risk variants identified by GWAS, which often lack clear biological interpretation in the absence of cellular context\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe consistent use of standardized genomic analysis pipelines, including well-established GWAS and PRS methodologies, lends credibility to these findings. Importantly, the observed cell-specific manifestation of genetic risk aligns with population-level evidence showing that genetic predisposition is modulated by behavioral, metabolic, and environmental factors\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e. Together, these data suggest that genetic susceptibility operates through conditional, celldependent mechanisms rather than fixed deterministic pathways\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003eCellular Context Defines Gene\u0026ndash;Environment Interactions\u003c/h2\u003e \u003cp\u003eA key contribution of this review is the demonstration that gene\u0026ndash;environment interactions in T2DM become apparent only when examined at single-cell resolution\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e. Environmental exposures\u0026mdash;such as nutritional excess, obesity-related metabolic stress, and inflammatory stimuli\u0026mdash;were shown to influence how genetic variants shape transcriptional outputs within specific cell populations\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e. In immune cells, environmental cues altered the magnitude and direction of genotype-associated expression patterns, while in pancreatic cells, metabolic conditions modified the expression of genes linked to insulin biosynthesis and cellular resilience\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe moderate variability detected across studies likely reflects authentic biological diversity arising from differences in tissue sources, genetic ancestry, disease stage, and environmental exposure rather than methodological shortcomings\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]\u003c/sup\u003e. This heterogeneity underscores the necessity of cell-resolved analytical strategies to capture the complexity of G\u0026times;E interactions in multifactorial diseases such as T2DM.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section2\"\u003e \u003ch2\u003eAnalytical Innovation through Integrative Modeling\u003c/h2\u003e \u003cp\u003eSeveral studies incorporated computational frameworks capable of integrating genetic, transcriptomic, and environmental data into unified analytical models\u003csup\u003e[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e. Machine learning approaches proved particularly effective in identifying interaction patterns that are difficult to detect using conventional statistical techniques. These models improved risk prediction accuracy when compared with single-modality analyses and highlighted the value of highdimensional data integration for complex disease research\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn addition, network-based analyses revealed that diverse genetic and environmental influences converge on shared regulatory nodes governing metabolic homeostasis, immune activation, and cellular stress pathways\u003csup\u003e[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e. Such insights move beyond descriptive associations, offering mechanistic hypotheses that may inform future therapeutic development and biomarker discovery.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003eImplications for Precision Medicine\u003c/h2\u003e \u003cp\u003eThe findings synthesized in this review have direct relevance for the development of personalized approaches to T2DM prevention and management\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e. The demonstration that genetic risk is expressed through defined cellular contexts suggests that future risk stratification models should incorporate cell-specific molecular information alongside traditional clinical and lifestyle factors. Single-cell\u0026ndash;derived transcriptional signatures and cell-informed PRS frameworks may ultimately enable earlier detection of disease susceptibility and more precise targeting of interventions.\u003c/p\u003e \u003cp\u003eMoreover, identifying how modifiable environmental factors interact with genetically sensitive cell populations provides a mechanistic foundation for individualized lifestyle and pharmacological strategies. Such approaches hold promise for mitigating disease risk even among individuals with substantial inherited susceptibility\u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudy Constraints\u003c/h3\u003e\n\u003cp\u003eDespite these advances, several limitations must be acknowledged. The relatively small number of eligible human studies reflects the technical complexity and resource demands of single-cell multiomics research. Differences in sequencing platforms, tissue sampling strategies, analytical workflows, and exposure assessment methods may contribute to residual variability. Furthermore, the predominance of cross-sectional designs limits insight into the temporal evolution of cellular G\u0026times;E interactions.\u003c/p\u003e \u003cp\u003ePopulation representation across studies was uneven, raising concerns about the transferability of PRS-based conclusions across diverse ancestries. Environmental exposures were often inferred rather than directly quantified, highlighting an important area for methodological improvement.\u003c/p\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eDirections for Future Research\u003c/h2\u003e \u003cp\u003eFuture research should emphasize longitudinal designs that integrate repeated single-cell profiling with comprehensive environmental and lifestyle measurements. Expanding representation across diverse populations and harmonizing analytical standards will be essential for robust inference. Advances in spatial transcriptomics and multi-modal single-cell technologies are expected to further illuminate how microenvironmental cues influence genetically susceptible cells within intact tissues.\u003c/p\u003e \u003c/div\u003e"},{"header":"References","content":"\u003cp\u003e1. \u0026nbsp; Mahajan A, Taliun D, Thurner M, Robertson NR, Torres JM, Rayner NW, et al. Finemapping type 2 diabetes loci to single-variant resolution using high-density imputation and islet-specific epigenome maps. \u003cem\u003eNat Genet\u003c/em\u003e. 2018;50(11):1505\u0026ndash;13.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e2. \u0026nbsp; Scott RA, Scott LJ, M\u0026auml;gi R, Marullo L, Gaulton KJ, Kaakinen M, et al. An expanded genome-wide association study of type 2 diabetes in Europeans. \u003cem\u003eDiabetes\u003c/em\u003e. 2017;66(11):2888\u0026ndash;902.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3. \u0026nbsp; Voight BF, Scott LJ, Steinthorsdottir V, Morris AP, Dina C, Welch RP, et al. Twelve type 2 diabetes susceptibility loci identified through large-scale association analysis. \u003cem\u003eNat Genet\u003c/em\u003e. 2010;42(7):579\u0026ndash;89.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e4. \u0026nbsp; Segerstolpe \u0026Aring;, Palasantza A, Eliasson P, Andersson AC, Andr\u0026eacute;asson AC, Sun X, et al. Single-cell transcriptome profiling of human pancreatic islets in health and type 2 diabetes. \u003cem\u003eCell Metab\u003c/em\u003e. 2016;24(4):593\u0026ndash;607.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e5.\u0026nbsp; \u0026nbsp;Baron M, Veres A, Wolock SL, Faust AL, Gaujoux R, Vetere A, et al. A single-cell transcriptomic map of the human and mouse pancreas reveals inter- and intra-cell population structure. \u003cem\u003eCell Syst\u003c/em\u003e. 2016;3(4):346\u0026ndash;60.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e6.\u0026nbsp; \u0026nbsp;Muraro MJ, Dharmadhikari G, Gr\u0026uuml;n D, Groen N, Dielen T, Jansen E, et al. A singlecell transcriptome atlas of the human pancreas. \u003cem\u003eCell Syst\u003c/em\u003e. 2016;3(4):385\u0026ndash;94.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e7.\u0026nbsp; \u0026nbsp;Lawlor N, George J, Bolisetty M, Kursawe R, Sun L, Sivakamasundari V, et al. Single-cell transcriptomes identify human islet cell signatures and reveal cell-type\u0026ndash; specific expression changes in type 2 diabetes. \u003cem\u003eGenome Res\u003c/em\u003e. 2017;27(2):208\u0026ndash;22.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e8.\u0026nbsp; \u0026nbsp;Fadista J, Vikman P, Laakso EO, Mollet IG, Esguerra JL, Taneera J, et al. Global genomic and transcriptomic analysis of human pancreatic islets reveals novel genes influencing glucose metabolism. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e. 2014;111(38):13924\u0026ndash;9.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e9.\u0026nbsp; \u0026nbsp;Khera AV, Chaffin M, Aragam KG, Haas ME, Roselli C, Choi SH, et al. Genomewide polygenic scores for common diseases identify individuals with risk equivalent to monogenic mutations. \u003cem\u003eNat Genet\u003c/em\u003e. 2018;50(9):1219\u0026ndash;24.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e10. Mars N, Koskela JT, Ripatti P, Kiiskinen T, Havulinna AS, Lindbohm JV, et al. Polygenic and clinical risk scores and their impact on age at onset and prediction of cardiometabolic diseases. \u003cem\u003eNat Med\u003c/em\u003e. 2020;26(4):549\u0026ndash;57.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e11. Duncan L, Shen H, Gelaye B, Meijsen J, Ressler K, Feldman M, et al. Analysis of polygenic risk score usage and performance in diverse human populations. \u003cem\u003eNat\u0026nbsp;\u003c/em\u003e\u003cem\u003eCommun\u003c/em\u003e. 2019;10(1):3328.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e12. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePLINK: rising to the challenge of larger and richer datasets. \u003cem\u003eGigaScience\u003c/em\u003e. 2015;4:7.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e13. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. \u003cem\u003eAm J\u0026nbsp;\u0026nbsp;\u003c/em\u003e\u003cem\u003eHum Genet\u003c/em\u003e. 2007;81(3):559\u0026ndash;75.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e14. Finucane HK, Reshef YA, Anttila V, Slowikowski K, Gusev A, Byrnes A, et al. Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types. \u003cem\u003eNat Genet\u003c/em\u003e. 2018;50(4):621\u0026ndash;9.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e15. Zhang Y, Hu Y, Wang J, Liu J, Chen L, Zhang S, et al. Cell-type-specific polygenic risk score analysis reveals cellular mechanisms of complex traits. \u003cem\u003eNat Genet\u003c/em\u003e. \u0026nbsp;2025;57(2):231\u0026ndash;42.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e16. Argelaguet R, Velten B, Arnol D, Dietrich S, Zenz T, Marioni JC, et al. Multi-omics factor analysis\u0026mdash;a framework for unsupervised integration of multi-omics data sets. \u003cem\u003eMol Syst Biol\u003c/em\u003e. 2018;14(6):e8124.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e17. Zhou W, Sailani MR, Contrepois K, Zhou Y, Ahadi S, Leopold SR, et al. Longitudinal multi-omics of host\u0026ndash;microbe dynamics in prediabetes. \u003cem\u003eNature\u003c/em\u003e. 2019;569(7758):663\u0026ndash; 71.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e18. Libbrecht MW, Noble WS. Machine learning applications in genetics and genomics. \u003cem\u003eNat Rev Genet\u003c/em\u003e. 2015;16(6):321\u0026ndash;32.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e19. Moore JH, Williams SM. Epistasis and its implications for personal genetics. \u003cem\u003eAm J\u0026nbsp;\u003c/em\u003e\u003cem\u003eHum Genet\u003c/em\u003e. 2009;85(3):309\u0026ndash;20.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e20. Franks PW, McCarthy MI. Exposing the exposures responsible for type 2 diabetes and obesity. \u003cem\u003eScience\u003c/em\u003e. 2016;354(6308):69\u0026ndash;73.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e21. Mart\u0026iacute;nez-L\u0026oacute;pez JA, Lindqvist A, Lopez-Pascual A, et al. Single-cell mRNAregulation analysis reveals cell type-specific mechanisms of type 2 diabetes. \u003cstrong\u003eNat\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eCommun.\u003c/strong\u003e 2025;16:9475. doi:10.1038/s41467-025-65060-z\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e22. Liu Y, Wang T, Wu R, et al. Single-cell transcriptome atlas and genome-wide Mendelian randomization reveal chemokine involvement in diverse immune cells in type 2 diabetes. \u003cstrong\u003eInt J Obes (Lond).\u003c/strong\u003e 2025;49:2213-2228. doi:10.1038/s41366-025-01846\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e23. Xiao M, Xu Z, Zhu X, Chen J, Wang R, Wang Y, Gao X, Luan S, Pu X. Immunological profiling in type 2 diabetes mellitus and type 2 diabetic kidney disease: insights from single-cell LacNAc sequencing. \u003cstrong\u003eFront Endocrinol\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e(Lausanne).\u003c/strong\u003e 2025;16:1550925. doi:10.3389/fendo.2025.1550925\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e24. Zhao J, Fang Z. Single-cell RNA sequencing reveals the dysfunctional characteristics of PBMCs in patients with type 2 diabetes mellitus. \u003cstrong\u003eFront Immunol.\u003c/strong\u003e 2024;15:1501660. doi:10.3389/fimmu.2024.1501660\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e25. Grenko CM, Taylor HJ, Bonnycastle LL, Xue D, Lee BN, Weiss Z, Yan T, Swift AJ, Mansell EC, Lee A, Robertson CC, Narisu N, Erdos MR, Chen S, Collins FS, Taylor DL. Single-cell transcriptomic profiling of human pancreatic islets reveals genes responsive to glucose exposure over 24 h. \u003cstrong\u003eDiabetologia.\u003c/strong\u003e 2024;67(10):2246-2259. doi:10.1007/s00125-024-06214-4\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e26. Yun JS, Jung SH, Shivakumar M, Xiao B, Khera AV, Won HH, Kim D. Polygenic risk for type 2 diabetes, lifestyle, metabolic health, and cardiovascular disease: a prospective UK Biobank study. \u003cstrong\u003eCardiovasc Diabetol.\u003c/strong\u003e 2022;21(1):131. doi:10.1186/s12933-022-01560-2\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e27. Zhang S, Shu H, Zhou J, Rubin-Sigler J, Yang X, Liu Y, Cooper-Knock J, Monte E,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eZhu C, Tu S, Li H, Tong M, Ecker JR, Ichida JK, Shen Y, Zeng J, Tsao PS, Snyder MP. Single-cell polygenic risk scores dissect cellular and molecular heterogeneity of complex human diseases. \u003cstrong\u003eNat Biotechnol.\u003c/strong\u003e 2025 Jul 25. doi:10.1038/s41587-025-02725-6\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e28. Zhang S, Shu H, Zhou J, Rubin-Sigler J, Yang X, Liu Y, Cooper-Knock J, Monte E,\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eZhu C, Tu S, Li H, Tong M, Ecker JR, Ichida JK, Shen Y, Zeng J, Tsao PS, Snyder MP. Deconvolution of polygenic risk score in single cells unravels cellular and molecular heterogeneity of complex human diseases. \u003cstrong\u003ebioRxiv\u003c/strong\u003e [Preprint]. 2024 May 14:2024.05.14.594252. doi:10.1101/2024.05.14.594252\u0026nbsp;\u003c/p\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 and 2 are available in the supplementary files section\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Indian Institute of Technology Jodhpur","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"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":"Type 2 diabetes mellitus, Single-cell transcriptomics, Gene–environment interaction, Polygenic risk score, Genome-wide association studies, Cellular heterogeneity, Multiomics integration, Machine learning, Precision medicine, Environmental exposure","lastPublishedDoi":"10.21203/rs.3.rs-9283816/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9283816/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eType 2 diabetes mellitus (T2DM) arises from the interplay between inherited genetic factors and environmental influences such as lifestyle, metabolic challenges, and inflammation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective:\u003c/strong\u003e To provide a systematic review and meta-analysis of human studies that use single-cell transcriptomic methods to examine gene–environment interactions in T2DM, focusing on the integration of genome-wide association studies (GWAS), polygenic risk scores (PRS), environmental exposures, and comprehensive multi-omics analysis frameworks. Methods: We conducted a systematic literature search of PubMed/MEDLINE, EMBASE, Web of Science, Scopus, and Google Scholar for eligible human studies published from January 2021 to December\u003c/p\u003e\n\u003cp\u003e2025. Studies were included if they applied single-cell transcriptomics to tissues relevant to T2DM and combined genetic risk assessment with environmental or clinical data. Study quality was evaluated using modified STROBE and QUADAS-2 tools. When suitable, random-effects meta-analyses were conducted, and heterogeneity was estimated using the I² statistic. \u003cstrong\u003eResults: \u003c/strong\u003eSeven studies of high methodological quality were included. Single-cell transcriptomic analyses consistently revealed significant cellular diversity in pancreatic, immune, and peripheral tissues, with disease-related gene expression changes restricted to specific cell populations. The integration of PRS showed that genetic risk was distributed unevenly across various cell types, identifying subsets of cells with heightened genetic susceptibility.\u003c/p\u003e\n\u003cp\u003eEnvironmental exposures were found to influence the relationship between genetic variations and gene expression, indicating genuine gene–environment interactions at the single-cell level. The meta-analysis revealed a significant overall association between molecular dysregulation and T2DM-related traits, with moderate heterogeneity (I² ≈ 48%).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eThe available evidence demonstrates that gene–environment interactions in T2DM are highly specific to individual cell types and are best elucidated through single-cell transcriptomic studies. Combining genetic risk, environmental exposures, and multi-omics analysis yields valuable mechanistic insights and informs the advancement of precision medicine for T2DM.\u003c/p\u003e","manuscriptTitle":"Gene-Environment Interaction through single-cell transcriptomics in Type 2 diabetes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-02 06:58:31","doi":"10.21203/rs.3.rs-9283816/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":"6602574a-959a-46b2-90ca-f555b390442b","owner":[],"postedDate":"April 2nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":65492748,"name":"Systems Biology"},{"id":65492749,"name":"Computational Biology"}],"tags":[],"updatedAt":"2026-04-02T06:58:31+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-02 06:58:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9283816","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9283816","identity":"rs-9283816","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}