{"paper_id":"2e5850f9-903f-4b2a-b54b-ea8345d9e151","body_text":"An epigenetic clock for chronological age estimation in East Asian populations | 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 An epigenetic clock for chronological age estimation in East Asian populations Kuan Chen Lu, Po-Hsiu Kuo, Amrita Chattopadhyay, Tzu-Pin Lu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7327305/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: The rapid rise in the older population over the past decade risks significant burden on healthcare, families, and society. This has drawn significant attention to aging research. DNA methylation is a heritable epigenetic alteration that is known to be linked with developmental processes via physiological and disease-associated changes. Hence, methylation changes are potential representatives of the natural aging process and age-related phenotypes, and are used as a predictor of chronological age. In this study, we explored the relationship between aging and changes in DNA methylation specifically among East-Asian (EAS) cohorts from Taiwan, Japan, and China to develop an epigenetic clock. Methods : Following quality control, methylation data from EAS samples were used to develop a predictive model, east-Asian epigenetic clock (EAS clock). A stepwise multivariate regression model with forward-selection and Bayesian Information Criteria (BIC) was implemented to conduct variable selection. EAS clock’s performance was validated through rigorous statistical evaluation. Subgroup analyses across age intervals were conducted to assess age-specific efficacy. Additionally, functional enrichment analysis using Ingenuity Pathway Analysis (IPA) was performed to investigate the biological relevance of the selected CpG sites. Results : Correlation analysis between predicted and actual chronological age showed strong positive correlations in both training (r = 0.71, p < 0.0001) and testing (r = 0.68, p < 0.0001) sets. Difference between estimated age by the EAS clock and chronological age showed an approximate median and mean value of zero. Subgroup analysis implied that epigenetic aging may vary across the lifespan, especially at age extremes. Functional annotation revealed enrichment of CpG-associated genes in age-related pathways, including neurodegeneration, musculoskeletal disorders, and immune regulation. Compared with other methylation clocks, EAS clock demonstrated tighter residual clustering around zero, indicating improved accuracy. Conclusion : EAS clock, a robust and accurate epigenetic clock tailored to East Asian populations was developed. Early and precise epigenetic age prediction may support timely anti-aging interventions and disease management, potentially mitigating the individual and healthcare burden of aging. Epigenetic clock East-Asian DNA-methylation chronological age Epigenetic age Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction There has been a rapid growth of the older population over the past decade, with people over the age of 60 years projected to constitute greater than 22% of the global population by 2050 ( 1 ). This trend raises concerns, as increased longevity is not matched by reduced chronic disease burden ( 2 ), placing significant burden on healthcare, families, and society. As a result, aging research has attracted growing attention in recent years. One goal has been to establish methods that can enable accurate measurement of biological aging, risk prediction and identification, and exploration of effective interventions. As opposed to chronological age, which is the total number of years a person has lived, biological age represents the true health and functional status of the body, that are prone to variations based on factors like lifestyle, genetics and epigenetics. Biological age is increasingly recognized as being more accurate than chronological age in determining chronic health outcomes. Biological aging is often accelerated compared to chronological aging, and accurate measurement of biological age early on can be used to identify a population at high risk for adverse health outcomes and who may be a target for clinical interventions. On the other hand, estimation of chronological age has application into fields such as forensic science for identifying individuals, especially in scenarios where age is unknown or disputed, and for conducting research for clinical uses by comprehending how age-related changes occur at a molecular level towards developing interventions, and potentially predicting health outcomes. DNA methylation is a heritable epigenetic alteration that is known to be linked with developmental processes in several eukaryotes ( 3 ). It is a process by which a methyl group is added to cytosines in the DNA molecule, resulting in the formation of 5-methylcytosine, which leads to modification of DNA activity without any sequence alteration ( 4 ). The landscape of DNA methylation is dynamic and is subject to physiological and disease-associated changes ( 5 ). While, massive changes in methylation are characteristic of early stages of development, epigenetic alterations in adult somatic tissue may be indicative of aging-associated deleterious events ( 6 ). Aging is characterized by a gradual decline of physiological, functional, and biological efficiency, the biological component of which has been explained through a variety of markers including genomic damage such as chromosomal instability and telomere shortening, mitochondrial damage leading to reduced energy production, stem cell depletion, accumulation of damaged proteins, or modifications of the epigenome ( 7 ). Many of the age-related diseases such as cancer ( 8 ), neurodegenerative diseases ( 9 ), atherosclerosis ( 10 ), and inflammation ( 11 ) are a result of the deregulation of pathways and modification in transcriptomes caused by alterations in DNA methylation. The nine “hallmarks of aging” as enumerated by Lopez-Otín et al. have also been suggested to be responsible ( 12 ). On the other hand, prior studies have hypothesized and demonstrated that there exists partial overlap of methylation changes with regions that harbor changes in histone modifications with age, and that approximately one-third of the methylation sites in the genome are affected by age ( 13 – 15 ).The above findings indicate that methylation changes could be potential representatives of the natural aging process and age-related phenotypes, and thereby could be used as a predictor of chronological age. Methylation patterns, furthermore, have been observed to vary across populations, where they not only impact externally observed phenotypes but also have an effect on underlying health disparities ( 16 ). Moreover, studying methylation profiles of subjects with identical ethnicity but living in different geographical locations would provide knowledge on conserved methylation patterns within populations. Investigation and evaluation of the epigenome is necessary to understand the ethnicity-specific effects of DNA methylation on aging. Epigenetic clocks have recently emerged as a promising tool for predicting both biological and chronological age ( 17 ). There exist several studies that have utilized differentially methylated regions (DMRs) as markers of age; however, they have mostly been directed towards Western populations. In this study, we explored the relationship between aging and changes in DNA methylation specifically among East-Asian cohorts from Taiwan, Japan, and China. A model was proposed for predicting chronological age among East Asians. An accurate epigenetic clock can enable measurement of true chronological age, whether accelerated or not, and estimate a person’s lifespan, allowing interventions to slow the rate of aging and maximize a person’s years of good health. Methods Datasets This study utilized methylation data from individuals of East Asian ancestry. A total of 3,637 subjects were included, out of which 2,090 were of Taiwanese origin obtained from the Taiwan Biobank (TWB), and the remaining 1,546 subjects of Chinese, Japanese, or Korean origin were obtained from the Gene Expression Omnibus (GEO) database (see Supplementary Table S1 for dataset details). Epigenetic data were profiled using DNA from whole blood samples utilizing the Illumina HumanMethylation450 BeadChip array, Infinium MethylationEPIC array, or Illumina HumanMethylation27 BeadChip array. The 27K array data were used solely for initial quality inspections and were excluded from downstream model training due to limited probe overlap. The TWB dataset initially consisted of 865,917 CpG sites across all samples. Datasets from both TWB and GEO were integrated by retaining 398,296 CpG sites that were common to the Illumina 450K and EPIC arrays. Samples from the 27K platform were excluded from this merging step. Raw signal intensity data were processed using standard pipelines described in the published studies associated with each dataset (see Supplementary Table S1 for references and accession numbers) Beta values (ratios of methylated to unmethylated probes for a given CpG site) were used for this analysis instead of M values (standardized Beta values –logit transformed) due to the challenges associated with it, such as the potential for infinite or undefined results. Beta values range from 0 to 1 and thus offer a more stable and interpretable methylation level for each CpG site. The individuals in the merged dataset were randomly allocated into two subsets: 2,546 in the training set, which was used to develop the model; and 1,091 in the testing set, which was used for evaluating the model’s performance, ensuring a comprehensive and representative sample for our epigenetic analysis. A total of 398,296 CpG targets that were common to all the datasets were included for analysis. Quality control A series of quality control steps were performed. CpG sites with detection p-values ≥ 0.01 and missing rates ≥ 5% were excluded. Normalization using Illumina GenomeStudio V2011.1 was conducted before merging the TWB with the GEO datasets to eliminate batch effects. Toward that end, coefficient of variation (CV) analysis was conducted on all CpG targets. The target with the lowest CV was designated as the reference and used as the baseline to normalize each CpG target by subtracting the reference value from each target. After normalization, a simple linear regression was performed with age as the dependent variable and each CpG site as an independent variable. CpG sites that did not meet a Bonferroni-corrected significance threshold were excluded from further analysis. Outlier CpG sites were identified by ranking adjusted coefficient of determination (Adj R²) values from linear regression; sites with greater than the third quartile (Q3) plus 0.15 times the interquartile range (IQR) (Adj R² >Q3 + 0.15×IQR) were excluded from further analysis. Construction of an epigenetic clock for East Asian populations: EAS clock The CpG targets that passed quality control and statistical filtering were used to construct the chronological age prediction model. A stepwise multivariate linear regression model with forward selection and Bayesian Information Criteria (BIC) was implemented to conduct variable selection for incorporating CpG targets as predictors, into the final prediction model. The final model (EAS clock) was selected as the one with the lowest BIC, ensuring optimal predictor selection while minimizing overfitting. This approach is a widely used statistical method for variable selection that balances model complexity with goodness of fit, and has been commonly applied across various fields including genomics and epidemiology. Training and testing Subjects from the merged dataset were randomly allocated into two subsets with an approximate training to testing ratio of 2:1. The training set was used to estimate regression coefficients, and the derived prediction equation was applied to the testing set to evaluate EAS clock’s performance, using the Pearson correlation coefficient (r) between predicted and actual chronological age, as well as the mean absolute error (MAE) and coefficient of determination (R²) to evaluate both accuracy and explained variance. A further subgroup analysis for different age groups (0–20 years, > 20–40 years, > 40–60 years, > 60–80 years, and > 80–100 years) were conducted towards understanding the EAS clock’s efficacy for specific age intervals. Finally, a comparison analysis was conducted with existing epigenetic clocks from EstimAge ( https://estimage.iac.rm.cnr.it/tutorial ). EstimAge allows identification of specific CpGs in the genome that are globally correlated to chronological age in any tissue or cell type through their methylation state ( 18 ). Functional Enrichment Analysis via IPA To investigate the biological functions associated with age-informative methylation markers, we conducted functional enrichment analysis using Ingenuity Pathway Analysis (IPA, QIAGEN Inc.) ( 19 ). Genes annotated to the CpG sites included in our final epigenetic clock model were analyzed in the IPA Core Analysis module, using default parameters with the Ingenuity Knowledge Base (Genes Only) as the reference background. Significance of enrichment was assessed using Fisher’s exact test, and disease or function terms with p -values < 0.01 were considered significantly enriched. Results Sample characteristics Quality control measures were implemented on an initial 2,090 subjects from TWB and 1,546 samples from other East Asian countries, providing a total of 398,296 CpG islands across 3,637 individuals for analysis. A total of 2,546 samples were allocated as the training dataset and 1,091 as the testing dataset. The age ranged between 6.40 and 84.00 years, with a mean of 49.98 years (SD = 12.93) for the training samples, while the testing set comprised subjects of age 1.34–85.65 years, with a mean of 47.92 years (SD = 14.77) (Fig. 1 ). The summary statistics for age are provided in Table 1 . Table 1 Summary statistics of age for training and testing datasets Datasets Minimum First quartile Median Mean Third Quartile Maximum Std. Dev Training 6.40 41.50 51.00 49.98 60.00 84.00 12.93 Testing 1.34 38.81 50.08 47.92 58.77 85.65 14.77 Regression analysis for model development The distribution of the CV calculated across all 398,296 CpG sites is demonstrated in Fig. 2 A, ranging from 0.96 to 242.90, with a mean (SD) of 51.72 (37.49). The CpG site cg10192265, located at chromosome 20 and at base pair 30,220,446 (GRCh38.p13), corresponding to the SNP rs1979233980, had the lowest CV and was selected as the reference for normalization of the dataset to correct for batch effects, if any, prior to downstream analysis. The Beta value distribution for cg10192265 as shown in Fig. 2 B, demonstrates a range of 0.9095 to 0.9941 and a mean (SD) of 0.9638 (0.009). Each CpG site was individually tested using a simple linear regression with chronological age as the dependent variable, and 134,635 CpG sites showed significant associations (Bonferroni-adjusted p < 1.25E-07). Among these, only 2,087 sites were retained, after eliminating outliers, based on the Adj R 2 values, to be used as potential predictors of chronological age. These candidate sites were then entered into a stepwise multivariate regression with forward selection, where the Bayesian Information Criterion (BIC) guided the inclusion of 38 CpG sites in the final model (EAS clock). Details of the selected CpG sites and their estimated regression coefficients (based on the training dataset) are provided in Supplementary Table S2 . Performance of the model (EAS clock) with the training and testing datasets In addition to internal validation using the training data, the performance of the fitted regression model (EAS clock) was also evaluated on an independent testing dataset. Pearson correlation analysis between predicted and actual chronological age showed significant positive correlations of r = 0.71 (p < 0.0001) in the training set and r = 0.68 (p < 0.0001) in the testing set (Table 2 , Fig. 3 ). Furthermore, the distribution of residuals (i.e., estimated age minus chronological age) showed that both the mean and median values were approximately zero in both the training and testing sets (Fig. 4 ), indicating no systematic bias. In addition, the mean absolute error (MAE) was 7.48 years for the training set and 6.22 years for the testing set. These results collectively demonstrated that our epigenetic clock, EAS clock, achieved strong predictive performance, with precision and accuracy comparable to or exceeding other models reported in population-level epigenetic aging studies. Table 2 Correlation between actual age and age estimated using methylation targets as predictors Data sets Correlation Estimate Standard Error T-statistics P-value Training 0.71 1.00 1.95E-02 51.20 < 0.0001 Testing 0.68 0.94 3.16E-02 29.80 < 0.0001 Age-subgroup analysis As shown in Fig. 5 , individuals exhibited distinct clustering patterns based on their epigenetic profiles, which aligning closely with chronological age. Notably, the distribution suggested a nonlinear relationship between DNA methylation and age, particularly during early childhood and advanced age, indicating that the epigenetic aging process may follow different trajectories across life stages. Comparison with EstimAge Six samples with the largest prediction errors using our proposed model were fed into the EstimAge platform, which provides epigenetic age estimates based on CpG sites strongly correlated with chronological age across various tissues and cell types. A comparison of the predicted age by our method (EAS Clock) and other existing models including Epigenetic Pacemaker (EPM), Hannum 13, Horvath 13 and18, PhenoAge, Zhang Enpred, and Zhang Blupred were evaluated using the actual chronological age as the reference. Table 3 lists the differences between the predicted age and the actual age for each of the methods, with red and blue indicating over- and underestimation, respectively. While deviations were observed in our model, similar or greater discrepancies were also present in the other methods, highlighting the challenge of age prediction in these outlier cases. The EPM method was found to perform the best, likely due to its use of 20,031 CpG sites. In contrast, our model relies on only 38 CpG sites, offering a far more interpretable and efficient alternative while maintaining competitive performance. Table 3. Comparative analysis of the deviations of predicted age from chronological age for different epigenetic clocks and EAS clock Values represent residuals calculated as (Chronological Age − Predicted Age). Blue text indicates negative residuals (i.e., predicted age older than actual age), and red text indicates positive residuals (i.e., predicted age younger than actual age) To further evaluate EAS clock’s performance, 273 samples randomly selected from the testing dataset were submitted to the EstimAge platform for comparative prediction analysis. The residual distributions of the EstimAge models were compared against those of the EAS epigenetic clock across the 273 samples. Residuals were calculated as the difference between the estimated age and the actual chronological age. Figure 6 A demonstrates the residual distributions of estimated age predicted by EPM_0.65, Hannum's 2013 model, Horvath's 2013 and 2018 models, EAS clock, PhenoAge, Zhang's Enpred, Blupred models, and GrimAge. Notably, the EPM_0.65 model exhibited the widest spread of residuals, with several predictions deviating substantially from zero—suggesting less consistent performance across samples compared to our model (EAS clock). Figure 6 B demonstrates the comparison after excluding the EPM_0.65 model to provide a cleaner and more focused picture of the residual distributions among the remaining epigenetic clock models. Notably, EAS clock stood out with residuals tightly clustered around zero, indicating high precision and accuracy in age estimation. This suggests that the clock is well-calibrated to the underlying biological characteristics of the dataset. Furthermore, the median residual of EAS clock was closest to zero among all methods, underscoring its robust fit for East Asian populations. Functional enrichment analysis To explore the biological functions associated with the 38 CpG sites in our EAS clock, we submitted their annotated genes to IPA’s Core Analysis and examined the top 25 enriched Disease & Function terms (Fisher’s exact test, p < 0.01; Supplementary Table S3 ). The most significant annotations fell into several coherent categories. Neurodegenerative and neurodevelopmental processes were heavily represented, with terms such as “Grade 3–4 glioma cancer” (p = 4.22 × 10⁻⁶), “Grade 4 astrocytoma” (p = 6.83 × 10⁻⁶), “brain astrocytoma” (p = 7.00 × 10⁻⁶) and “dementia” (p = 1.10 × 10⁻⁵). Musculoskeletal decline emerged via enrichments like “curvature of spine” (p = 2.80 × 10⁻⁵) and “osteoporosis” (p = 4.00 × 10⁻⁵). Immune and inflammatory signatures appeared in terms such as “macrophage activation” (p = 6.20 × 10⁻⁵) and “cytokine signaling” (p = 8.10 × 10⁻⁵). Finally, cancer-related pathways beyond the central nervous system—e.g. “skin cancer” (p = 2.10 × 10⁻⁵) and “polycystic ovary disease” (p = 5.50 × 10⁻⁵)—were also enriched. Across these categories, several genes recurred, most notably ELOVL2 , DNMT3A , and CHRNA9 , each implicated in multiple enriched annotations. Together, these results indicate that the CpG markers driving our East Asian clock are functionally embedded in key biological pathways underlying systemic aging. Discussion Given the well-established role of age-associated epigenetic modifications, particularly the global decline of DNA methylation with age, methylation is widely accepted as a reliable proxy for estimating chronological age ( 20 ). In this study, we specifically focused on East Asian ancestry and constructed an epigenetic clock, EAS clock, based on methylation profiles, demonstrating accurate and reliable prediction of chronological age. Rigorous evaluations showed that the EAS clock performed in a robust manner and predicted age with high accuracy and precision. We further utilized the methylation patterns to conduct a subgroup analysis for different age groups to evaluate the EAS clock’s efficacy for specific age intervals which demonstrated that the rate and pattern of epigenetic changes doesn’t progress uniformly through the lifespan. Residuals, defined as actual age minus predicted age, showed an increase in variance at the extremes of the age spectrum, particularly among older adults. This indicated potential overfitting or that age-related methylation dynamics possibly reaches a plateau or exhibits nonlinear trajectories in older individuals, limiting model resolution at advanced ages. To further elucidate the biological relevance of age-associated methylation changes, we performed functional enrichment analysis using IPA. Results revealed that CpG-associated genes in our model are enriched for aging-related functions, including neurodegeneration ( e.g., dementia ), musculoskeletal decline ( e.g., curvature of spine ), and immune/inflammatory pathways. Although cancer-related terms were also enriched, these likely reflect biological overlap with aging rather than model bias. For clarity, we prioritized age- and development-related functions in our interpretation. The complete list of top 25 annotations is available in Supplementary Table S3 . Notably, several genes recurred across multiple enriched terms, suggesting they may play a central role in aging-related biological pathways. Among these, ELOVL2 is one of the most well-documented age-associated loci: its promoter methylation levels rise consistently with age in various tissues and it has been widely used as a single-locus biomarker for epigenetic aging. Experimental studies have shown that restoring ELOVL2 function in aged mice can reverse certain vision impairments, implicating a potentially causal role in age-related functional decline ( 21 ). DNMT3A , a de novo DNA methyltransferase, is involved in the establishment of methylation patterns during development and hematopoietic stem cell renewal. Loss-of-function mutations in DNMT3A have been linked to clonal hematopoiesis and accelerated epigenetic aging ( 22 ). Finally, CHRNA9 , a subunit of nicotinic acetylcholine receptors, appeared in terms related to auditory cell morphology and neurobiology. Dysregulation of CHRNA9 has been implicated in hearing loss and neuroinflammation, both of which are common features of aging ( 23 ). These findings support the notion that the CpG sites selected for our East Asian epigenetic clock are not only predictive of chronological age but are also functionally embedded in pathways associated with systemic aging and tissue degeneration. Their recurrence in age-related annotations reinforces the biological interpretability and relevance of our model’s features. The non-linear relationship between methylation and age, especially in early life and advanced age, as shown in our subgroup analysis (Fig. 5 ), may stem from rapid methylation changes during early development, leading to greater inter-individual variability, and a biological plateau in older age, reducing sensitivity to age differences. These dynamics likely contribute to reduced model correlations at age extremes and underscore a key challenge for epigenetic clocks. Future models could improve accuracy by incorporating age-specific trajectories or flexible non-linear approaches such as splines or Gaussian processes to better capture these complex patterns across the lifespan. Numerous epigenetic clocks have been developed, among which Horvath’s clock is the most widely cited and extensively validated. It was developed using methylation data from diverse tissues and conditions, including cancer ( 24 , 25 ), Alzheimer’s disease ( 26 ), aging ( 27 ), and lifestyle factor ( 28 ). Notably, it was the first age estimator to leverage methylation profiles across multiple human tissues and developmental stages. The model was constructed using elastic net regression, which automatically selected 353 age-associated CpG sites—193 positively and 160 negatively correlated with age. The correlation between predicted age and chronological age, for Horvath's clock, was 0.96 with a median absolute difference of 3.6 years ( 29 , 30 ). Another one is Hannum’s epigenetic clock, which is a single-tissue DNA methylation-based age predictor, which was created by training an elastic net regression model utilizing 71 age-related CpGs from 482 Caucasian and 174 Hispanic adults ( 31 ). In Hannum’s epigenetic clock, the correlation between predicted and chronological age was 0.96, with a median absolute difference of 3.9 years; however, this clock has some bias in estimation when applied to non-blood tissues ( 17 , 32 , 33 ). Other established epigenetic clocks includes the epigenetic pacemaker (EPM) ( 34 ), PhenoAge ( 35 ), and BluPred, Enpred ( 36 ), and GrimAge ( 37 ). The age predicted by these models is referred to as “epigenetic age”, which often deviates from an individual’s chronological age. This difference, often termed as the residual, indicates “epigenetic age acceleration” when epigenetic age exceeds chronological age, and “epigenetic age deceleration” when epigenetic age is lower than chronological age. Exploring this gap between chronological age and epigenetic age is a major research focus in the field of aging. We compared our proposed epigenetic clock with the aforementioned models, and found that it produced the smallest residuals, tightly clustered around zero. This suggested superior accuracy of EAS clock in age estimation, particularly for East Asian individuals. As multiple models may yield similar results for easily predictable samples, we specifically evaluated the performance of EAS clock as opposed to other existing epigenetic clocks using more challenging or borderline cases. Additionally, we emphasized comparisons with population-matched models—those trained on East Asian datasets—to ensure fair and interpretable benchmarking, particularly for samples that may be misestimated by clocks trained on other ancestries. Furthermore, EAS clock, as opposed to other existing epigenetic clocks, was intentionally designed to use a minimal number of CpG targets (38 only), selected based on statistical significance and model fit criteria. This design choice reflects a trade-off between model complexity and practical utility. While methods like EPM achieved higher accuracy by incorporating tens of thousands of CpG sites (e.g., 20,031 in EPM), they require extensive data and computational resources, which may limit their feasibility in routine clinical or screening settings. In contrast, our EAS Clock model provides a more parsimonious alternative, potentially allowing for easier implementation, lower cost, and better generalizability in real-world applications, despite a modest reduction in predictive accuracy. We believe this balance makes EAS clock particularly suitable for scalable use in health examination contexts. In this study, we compared our EAS clock with several widely used epigenetic clocks that estimate chronological age, including GrimAge, Horvath’s 2013 and 2018 clocks, Hannum’s clock, PhenoAge, EPM, an Zhang’s Enpred and Blupred clocks. Among these, GrimAge was successfully applied to our data where applicable, which exhibited residual distributions highly similar to those of PhenoAge (Fig. 6 ), suggesting potential convergence in the biological aging dimensions they capture. Another popular model, DunedinPACE ( 38 ), which estimates the pace of aging rather than accumulated age, was not included in this study for comparison, due to its conceptual difference and data requirements with that of the other epigenetic clocks including EAS clock. While it offers valuable insights into the rate of biological aging, it is not directly comparable to our model’s aim of estimating chronological age. Future studies may incorporate DunedinPACE to explore longitudinal aging trajectories in East Asian populations. Epigenetic clocks play important roles in aging research. For one, they are used for evaluating the efficacy of interventions for aging. These interventions may reverse epigenetic age, thereby mitigating the physical manifestations of aging. Promising approaches that target the epigenome to promote rejuvenation include cellular reprogramming, pharmacological treatments, and lifestyle modifications. To gain a better understanding of the connection between epigenetic age and chronological age, it is essential to conduct longitudinal cohort studies, which follow individuals over extended timeframes, providing valuable insights into the phenotypic changes that occur as they age and the factors that cause diseases ( 39 ). A meta-analysis by Marioni et al. revealed that epigenetic aging progresses slightly more slowly than chronological aging over the life course, especially in older adults ( 40 ). Other studies have also reported a non-linear, logarithmic pattern of epigenetic aging during adolescence ( 30 , 41 ). Longitudinal cohort studies can potentially identify deviations between epigenetic and chronological age, enabling early-life interventions ( 17 ). However, the influence of genetics and signaling pathways on DNA methylation across different stages of aging remains unclear ( 42 – 45 ). A key limitation of current epigenetic clocks lies in their variability across CpG targets, Illumina array platforms, tissue sources, and population ancestry. Such inconsistencies contribute to considerable heterogeneity across studies and constrain the generalizability and applicability of epigenetic clocks. Utilizing tissue-specific CpGs to construct clocks can potentially lead to more accurate and robust predictions. However, adjusting for cell type heterogeneity remains essential when estimating age from DNA obtained from mixed or diverse cell populations ( 36 ). Future studies will be conducted leveraging multiple accessible DNA sources to produce more accurate and comprehensive age estimates ( 39 ). Moreover, additional biological markers will be integrated alongside DNA methylation to enhance the current model’s (EAS clock) ability to estimate biological age. Although the TWB dataset contains limited clinical and lifestyle variables, our analytic strategy required merging it with GEO datasets, which do not provide such annotations. As a result, we were unable to explore the associations between epigenetic age acceleration and clinical outcomes or environmental exposures. This limitation underscores the importance of using well-annotated clinical cohorts in future research to assess the broader biological relevance and potential health applications of our model in aging and precision medicine. Conclusion This study presents a robust and accurate epigenetic clock, EAS clock, tailored to East Asian populations for estimating chronological age. While it performs well in most scenarios, its use may be limited when biological age diverges significantly from chronological age. Nonetheless, such discrepancies may themselves offer insights into age acceleration or deceleration. In the current era of increasing life expectancy, early and accurate prediction of epigenetic age could facilitate timely anti-aging interventions and targeted treatments for age-related diseases. Such efforts may ultimately ease the aging process for individuals and reduce the burden on healthcare systems. Declarations Ethics approval and consent to participate The study was approved by the institutional review board (IRB # is: 201506095RINC) of the National Taiwan University Hospital. All subjects provided written consent for participation. Consent for publication Not applicable Availability of data and materials The data s that support the findings of this study belong to Taiwan Biobank, which requires permission for access. The data is available from the corresponding author on reasonable request, upon permission obtained from Taiwan Biobank. Competing interests The authors declare that they have no competing interests. Funding This work was partly supported by National Science and Technology Council, Taiwan (MOST-109-2314-B-002-151-MY3, NSTC-113-2314-B-002-170-MY3, and 114-2314-B-002-056-) and Population Health and Welfare Research Center from Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan (grant number NTU-113 L9004).The funders had no role in the study design, data collection and analysis, the decision to publish, or preparation of the manuscript. Authors' contributions T.P.L. conceived and designed the study. P.H.K. conducted data acquisition. T.P.L and A.C. provided the resources and administrative support. A.C., and T.P.L supervised the analysis. K.C.L did the formal analysis. K.C.L. and A.C. interpreted data. A.C. drafted the work. K.C.L, P.H.K., T.P.L, and A.C. reviewed and approved the manuscript. Acknowledgements We thank Dr. Melissa Stauffer for English editing our manuscript. References Newgard CB, Sharpless NE. Coming of age: molecular drivers of aging and therapeutic opportunities. The Journal of clinical investigation. 2013;123(3):946-50. Partridge L, Deelen J, Slagboom P. Facing up to the global challenges of ageing. Nature [Internet]. 2018; 561 (7721): 45–56. Greenberg MV, Bourc’his D. The diverse roles of DNA methylation in mammalian development and disease. Nature reviews Molecular cell biology. 2019;20(10):590-607. Singal R, Ginder GD. DNA methylation. Blood, The Journal of the American Society of Hematology. 1999;93(12):4059-70. Andersen GB, Tost J. A summary of the biological processes, disease-associated changes, and clinical applications of DNA methylation. DNA Methylation Protocols. 2018:3-30. Jung M, Pfeifer GP. Aging and DNA methylation. BMC Biol. 2015;13:1-8. Almeida AJPOd, Ribeiro TP, Medeiros IAd. Aging: molecular pathways and implications on the cardiovascular system. Oxid Med Cell Longev. 2017;2017(1):7941563. Jones PA. DNA methylation and cancer. Oncogene. 2002;21(35):5358-60. Fodder K, de Silva R, Warner TT, Bettencourt C. The contribution of DNA methylation to the (dys) function of oligodendroglia in neurodegeneration. Acta Neuropathologica Communications. 2023;11(1):106. Tabaei S, Tabaee SS. DNA methylation abnormalities in atherosclerosis. Artificial cells, nanomedicine, and biotechnology. 2019;47(1):2031-41. Ligthart S, Marzi C, Aslibekyan S, Mendelson MM, Conneely KN, Tanaka T, et al. DNA methylation signatures of chronic low-grade inflammation are associated with complex diseases. Genome Biol. 2016;17:1-15. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194-217. Johansson Å, Enroth S, Gyllensten U. Continuous aging of the human DNA methylome throughout the human lifespan. PLoS One. 2013;8(6):e67378. Torres IO, Fujimori DG. Functional coupling between writers, erasers and readers of histone and DNA methylation. Curr Opin Struct Biol. 2015;35:68-75. Saul D, Kosinsky RL. Epigenetics of aging and aging-associated diseases. International journal of molecular sciences. 2021;22(1):401. Giuliani C, Sazzini M, Bacalini MG, Pirazzini C, Marasco E, Fontanesi E, et al. Epigenetic variability across human populations: a focus on DNA methylation profiles of the KRTCAP3, MAD1L1 and BRSK2 genes. Genome Biol Evol. 2016;8(9):2760-73. Duan R, Fu Q, Sun Y, Li Q. Epigenetic clock: A promising biomarker and practical tool in aging. Ageing Res Rev. 2022;81:101743. Di Lena P, Sala C, Nardini C. Estimage: a webserver hub for the computation of methylation age. Nucleic Acids Res. 2021;49(W1):W199-W206. Inc. Q. Ingenuity Pathway Analysis (IPA). Version 01-23-01 ed. Redwood City, CA: QIAGEN Inc.; 2025. Hillary RF, Stevenson AJ, McCartney DL, Campbell A, Walker RM, Howard DM, et al. Epigenetic measures of ageing predict the prevalence and incidence of leading causes of death and disease burden. Clin Epigenetics. 2020;12:1-12. Belsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022;11:e73420. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman Peter V, Mar Brenton G, et al. Age-Related Clonal Hematopoiesis Associated with Adverse Outcomes. New England Journal of Medicine.371(26):2488-98. Elgoyhen AB, Katz E, Fuchs PA. The nicotinic receptor of cochlear hair cells: a possible pharmacotherapeutic target? Biochem Pharmacol. 2009;78(7):712-9. Levine ME, Hosgood HD, Chen B, Absher D, Assimes T, Horvath S. DNA methylation age of blood predicts future onset of lung cancer in the women's health initiative. Aging. 2015;7(9):690. Sehl ME, Henry JE, Storniolo AM, Ganz PA, Horvath S. DNA methylation age is elevated in breast tissue of healthy women. Breast Cancer Res Treat. 2017;164:209-19. McCartney DL, Stevenson AJ, Walker RM, Gibson J, Morris SW, Campbell A, et al. Investigating the relationship between DNA methylation age acceleration and risk factors for Alzheimer's disease. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2018;10:429-37. Declerck K, Berghe WV. Back to the future: epigenetic clock plasticity towards healthy aging. Mech Ageing Dev. 2018;174:18-29. Ryan J, Wrigglesworth J, Loong J, Fransquet PD, Woods RL. A systematic review and meta-analysis of environmental, lifestyle, and health factors associated with DNA methylation age. The Journals of Gerontology: Series A. 2020;75(3):481-94. Oblak L, van der Zaag J, Higgins-Chen AT, Levine ME, Boks MP. A systematic review of biological, social and environmental factors associated with epigenetic clock acceleration. Ageing Res Rev. 2021;69:101348. Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14:1-20. Hannum G, Guinney J, Zhao L, Zhang L, Hughes G, Sadda S, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell. 2013;49(2):359-67. Simpkin AJ, Hemani G, Suderman M, Gaunt TR, Lyttleton O, Mcardle WL, et al. Prenatal and early life influences on epigenetic age in children: a study of mother–offspring pairs from two cohort studies. Human molecular genetics. 2016;25(1):191-201. Simpkin AJ, Howe LD, Tilling K, Gaunt TR, Lyttleton O, McArdle WL, et al. The epigenetic clock and physical development during childhood and adolescence: longitudinal analysis from a UK birth cohort. International journal of epidemiology. 2017;46(2):549-58. Farrell C, Snir S, Pellegrini M. The Epigenetic Pacemaker: modeling epigenetic states under an evolutionary framework. Bioinformatics. 2020;36(17):4662-3. Levine ME, Lu AT, Quach A, Chen BH, Assimes TL, Bandinelli S, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging. 2018;10(4):573. Zhang Q, Vallerga CL, Walker RM, Lin T, Henders AK, Montgomery GW, et al. Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing. Genome medicine. 2019;11:1-11. Lu AT, Binder AM, Zhang J, Yan Q, Reiner AP, Cox SR, et al. DNA methylation GrimAge version 2. Aging (Albany NY). 2022;14(23):9484-549. Belsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. Elife. 2022;11. Horvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nature reviews genetics. 2018;19(6):371-84. Marioni RE, Suderman M, Chen BH, Horvath S, Bandinelli S, Morris T, et al. Tracking the epigenetic clock across the human life course: a meta-analysis of longitudinal cohort data. The Journals of Gerontology: Series A. 2019;74(1):57-61. Alisch RS, Barwick BG, Chopra P, Myrick LK, Satten GA, Conneely KN, et al. Age-associated DNA methylation in pediatric populations. Genome Res. 2012;22(4):623-32. Gaunt TR, Shihab HA, Hemani G, Min JL, Woodward G, Lyttleton O, et al. Systematic identification of genetic influences on methylation across the human life course. Genome Biol. 2016;17:1-14. Van Dongen J, Nivard MG, Willemsen G, Hottenga J-J, Helmer Q, Dolan CV, et al. Genetic and environmental influences interact with age and sex in shaping the human methylome. Nature communications. 2016;7(1):11115. Bell CG, Gao F, Yuan W, Roos L, Acton RJ, Xia Y, et al. Obligatory and facilitative allelic variation in the DNA methylome within common disease-associated loci. Nature communications. 2018;9(1):8. Hannon E, Knox O, Sugden K, Burrage J, Wong CC, Belsky DW, et al. Characterizing genetic and environmental influences on variable DNA methylation using monozygotic and dizygotic twins. PLoS genetics. 2018;14(8):e1007544. Supplementary Tables Supplementary tables S1-S3 are not available with this version. Additional Declarations No competing interests reported. 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-7327305\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":true,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":517820350,\"identity\":\"a2f99c25-cfcd-47f7-a900-16648d06dd46\",\"order_by\":0,\"name\":\"Kuan Chen Lu\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Taiwan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Kuan\",\"middleName\":\"Chen\",\"lastName\":\"Lu\",\"suffix\":\"\"},{\"id\":517820351,\"identity\":\"4a28701e-b4c2-444e-a730-1a6cfa1d3110\",\"order_by\":1,\"name\":\"Po-Hsiu Kuo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Taiwan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Po-Hsiu\",\"middleName\":\"\",\"lastName\":\"Kuo\",\"suffix\":\"\"},{\"id\":517820352,\"identity\":\"c1efc2f8-e6da-4905-8612-c1eab9d6ad8b\",\"order_by\":2,\"name\":\"Amrita Chattopadhyay\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"National Taiwan University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Amrita\",\"middleName\":\"\",\"lastName\":\"Chattopadhyay\",\"suffix\":\"\"},{\"id\":517820353,\"identity\":\"439461c9-35cd-466d-a6f6-042f114f7fa1\",\"order_by\":3,\"name\":\"Tzu-Pin Lu\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAqklEQVRIiWNgGAWjYFAC5gaGDxCWAXEaeBgYGxhnQFQDiQQitTDzkKTFXrqxTdp2x5/EBvbmbRKMPw4TYYvMwTbp3DMGiQ08x8okGBKI0SKRCNTSBtQikWMG1HKbSC2WIC3yb0jRwgi2hYdYLXcONlv2thkbt/GkFVskpP0nrIV9dvPBGz/b5GT72Q9vvPHBJo2wFgYJKM0GIhKI0IDQMgpGwSgYBaMAJwAArqoyk7VOjSwAAAAASUVORK5CYII=\",\"orcid\":\"\",\"institution\":\"National Taiwan University\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Tzu-Pin\",\"middleName\":\"\",\"lastName\":\"Lu\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-08-08 12:38:23\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-7327305/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-7327305/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":91845398,\"identity\":\"a5f6e58b-2fb8-481b-a602-bef5f21b91cd\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:06\",\"extension\":\"docx\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":4769663,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"newsubmissioneditedKC0729amrita0730Upload0807.docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/fd2b17798318ca02262b1556.docx\"},{\"id\":91846321,\"identity\":\"558e8dd5-a12c-43fc-99fa-1815b30f1202\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:19:01\",\"extension\":\"json\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":7605,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"1d6f8d49ecf24c7cb06e709955c95b38.json\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/414c993898de844655b64e5a.json\"},{\"id\":91845383,\"identity\":\"5cf3502e-3f7d-4570-8175-d93ef6fc603d\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"xml\",\"order_by\":2,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":118255,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"1d6f8d49ecf24c7cb06e709955c95b381enriched.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/a7865c671cb1f50be3f27c75.xml\"},{\"id\":91845356,\"identity\":\"36867b4b-fb62-4072-9312-7c4c278d0570\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:02\",\"extension\":\"png\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":344407,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/9b448c0292cff27b454734e3.png\"},{\"id\":91845352,\"identity\":\"eb3ba8fc-ca06-4d7f-bdbe-6f4bb51fe688\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:01\",\"extension\":\"png\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":39765,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/6e1bc919cb2d71f66bd87931.png\"},{\"id\":91845350,\"identity\":\"0f547b5b-dcd7-40b6-a0d6-3efe27b5cdee\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:10:59\",\"extension\":\"png\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":205819,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/b0eda5924aa732303cb57573.png\"},{\"id\":91845400,\"identity\":\"37223d01-15b9-4291-bc20-5bdcb1ff36b5\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:07\",\"extension\":\"png\",\"order_by\":6,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":132362,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/a50edb40ee3b77196cb4b1a9.png\"},{\"id\":91845390,\"identity\":\"d5ea3409-28a3-4279-8836-665973b688e8\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"jpeg\",\"order_by\":7,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":3714202,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage5.jpeg\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/f2423851be6b7f940de193af.jpeg\"},{\"id\":91845391,\"identity\":\"6b7ea8a2-6a67-4971-bf96-eadc958efa61\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"png\",\"order_by\":8,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":237754,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"floatimage6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/7a025697f99f31e6443692b3.png\"},{\"id\":91845382,\"identity\":\"a665c9eb-42a2-4266-afc7-e978f28d28a8\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"png\",\"order_by\":9,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":30647,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/db7359815cb1f3872c0fee81.png\"},{\"id\":91845393,\"identity\":\"8500f0ee-3aa4-40a0-b8ae-7e800e1a4815\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:05\",\"extension\":\"png\",\"order_by\":10,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":8830,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/9bd5edf4ff4277c2b4d608e0.png\"},{\"id\":91847734,\"identity\":\"20367410-abb4-4957-84e9-7d928a6a753e\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:27:04\",\"extension\":\"png\",\"order_by\":11,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":46113,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/b54a39aa88eb1eed6c435d05.png\"},{\"id\":91845355,\"identity\":\"1f16ef54-6fa0-440a-b70e-c79753e5407e\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:02\",\"extension\":\"png\",\"order_by\":12,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":28993,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/46f5a27724d79a304165457e.png\"},{\"id\":91846333,\"identity\":\"1caf0e4f-590a-4246-9d1a-b58bfbe9e4bf\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:19:04\",\"extension\":\"png\",\"order_by\":13,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":14331,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/15ffffcbfe3767a514a6e2eb.png\"},{\"id\":91845394,\"identity\":\"51af5006-8095-4b26-982a-aea81d2bb4be\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:05\",\"extension\":\"png\",\"order_by\":14,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":32844,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Onlinefloatimage6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/d0aa6c1f484c115c0ffe48cd.png\"},{\"id\":91845389,\"identity\":\"988bb451-4170-483f-873c-55f1d96decfa\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"xml\",\"order_by\":15,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":117869,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"1d6f8d49ecf24c7cb06e709955c95b381structuring.xml\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/e6bf2aabb3cd2cde13e95c6c.xml\"},{\"id\":91845388,\"identity\":\"feb8242f-3a9b-4484-af32-218bc5932fce\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:04\",\"extension\":\"html\",\"order_by\":16,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"acdc-reference\",\"size\":126732,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"earlyproof.html\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/9e8c5f2bf5aa10149693c031.html\"},{\"id\":91845395,\"identity\":\"b364b481-d2a8-4efd-93eb-05af0bcfbbe3\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:05\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":73621,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eAge distribution of study subjects in the training and testing datasets.\\u003c/strong\\u003e\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/ba3021489f5597f0bf838c36.png\"},{\"id\":91845353,\"identity\":\"730d791a-78b1-4dec-89a3-1c82c3e2f968\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:01\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":20998,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eDistribution of (A) coefficient of variation among all samples and (B) Beta values of methylation for the reference target. \\u003c/strong\\u003eThe coefficient of variation across 398,296 CpG islands was used to select the reference target (cg10192265).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/0905115f14fbf7cfe58e8341.png\"},{\"id\":91845365,\"identity\":\"0bc0e3d4-ff2c-4156-8442-d6725e1f902d\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:02\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":81475,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eCorrelation plots between actual chronological age and predicted chronological age for (A) the training dataset and (B) the testing dataset. Red dashed line represents the Y = X line that is included in the plot as a reference. Black dots represent individual samples, while red dots indicate outliers, defined as values falling outside the median ± 3 interquartile ranges (IQR) of the difference between actual and predicted age.\\u003c/strong\\u003e\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/e3afc043320de1bca08af294.png\"},{\"id\":91845396,\"identity\":\"675c75f0-4ba8-45c3-9096-f352fabaedce\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:05\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":45820,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eDensity plots of the difference between predicted and actual chronological age for (A) training and (B) testing datasets. The x-axis represents the difference between the actual (chronological) age and the age predicted by the EAS clock (in years), with values closer to zero indicating more accurate predictions. The y-axis represents the density (probability distribution) of these differences across individuals in the respective datasets. Each histogram is overlaid with a blue kernel density estimation (KDE) curve, providing a smoothed representation of the distribution. The vertical red line indicates the median of the residuals (Real Age − Estimated Age), while the horizontal orange rug plot at the bottom shows the individual data points. The boxplot at the top illustrates the distribution of residuals, showing the median, interquartile range (IQR), and potential outliers.\\u003c/strong\\u003e\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/4af8220535947b44babcbe77.png\"},{\"id\":91845351,\"identity\":\"808f3ff3-ef05-4219-8f04-79c0ea8bb434\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:10:59\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":21139,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eAge subgroups identified using DNA methylation profiles in East Asian populations.\\u003c/strong\\u003e \\u003cstrong\\u003eThe x-axis represents chronological age categories: 0–19 years, 20–39 years, 40–59 years, 60–79 years, and 80–99 years. The y-axis shows residuals, calculated as the difference between actual chronological age and predicted epigenetic age (Real Age − Estimated Age). Each colored boxplot displays the distribution of residuals for individuals within the corresponding age range. The central line in each box indicates the median, the box denotes the interquartile range (IQR), and the whiskers extend to 1.5 times the IQR. Data points beyond the whiskers are considered outliers and are shown as individual black dots. A fitted regression line is overlaid to illustrate the trend in residuals across age groups, revealing deviations at the youngest and oldest age extremes that may reflect nonlinear methylation dynamics.\\u003c/strong\\u003e\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/025bb7717cb2397a1a889de2.png\"},{\"id\":91845357,\"identity\":\"550d99d0-528c-43cb-b80b-8f4c17db1c8a\",\"added_by\":\"auto\",\"created_at\":\"2025-09-22 10:11:02\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":80780,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eComparison of the residuals (differences between estimated age and actual chronological age) of the epigenetic clocks from EstimAge and the proposed epigenetic clock for East Asian populations, EAS clock, using a randomly selected 273 samples from the testing dataset. \\u003c/strong\\u003e(A) Residual comparison of all models against epigenetic clock for East Asians (EAS clock). (B) Similar comparison with EPM_0.65 excluded.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/620b6da165094b69ecb4bd51.png\"},{\"id\":95801704,\"identity\":\"98b4e486-37e2-4182-a803-50ee35a7e35d\",\"added_by\":\"auto\",\"created_at\":\"2025-11-13 08:26:00\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":2196946,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7327305/v1/04878145-f499-40bb-af7f-75eaff8b5fe4.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"An epigenetic clock for chronological age estimation in East Asian populations\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003eThere has been a rapid growth of the older population over the past decade, with people over the age of 60 years projected to constitute greater than 22% of the global population by 2050 (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e). This trend raises concerns, as increased longevity is not matched by reduced chronic disease burden (\\u003cspan citationid=\\\"CR2\\\" class=\\\"CitationRef\\\"\\u003e2\\u003c/span\\u003e), placing significant burden on healthcare, families, and society. As a result, aging research has attracted growing attention in recent years. One goal has been to establish methods that can enable accurate measurement of biological aging, risk prediction and identification, and exploration of effective interventions. As opposed to chronological age, which is the total number of years a person has lived, biological age represents the true health and functional status of the body, that are prone to variations based on factors like lifestyle, genetics and epigenetics. Biological age is increasingly recognized as being more accurate than chronological age in determining chronic health outcomes. Biological aging is often accelerated compared to chronological aging, and accurate measurement of biological age early on can be used to identify a population at high risk for adverse health outcomes and who may be a target for clinical interventions. On the other hand, estimation of chronological age has application into fields such as forensic science for identifying individuals, especially in scenarios where age is unknown or disputed, and for conducting research for clinical uses by comprehending how age-related changes occur at a molecular level towards developing interventions, and potentially predicting health outcomes.\\u003c/p\\u003e\\u003cp\\u003eDNA methylation is a heritable epigenetic alteration that is known to be linked with developmental processes in several eukaryotes (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e). It is a process by which a methyl group is added to cytosines in the DNA molecule, resulting in the formation of 5-methylcytosine, which leads to modification of DNA activity without any sequence alteration (\\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e). The landscape of DNA methylation is dynamic and is subject to physiological and disease-associated changes (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e). While, massive changes in methylation are characteristic of early stages of development, epigenetic alterations in adult somatic tissue may be indicative of aging-associated deleterious events (\\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e). Aging is characterized by a gradual decline of physiological, functional, and biological efficiency, the biological component of which has been explained through a variety of markers including genomic damage such as chromosomal instability and telomere shortening, mitochondrial damage leading to reduced energy production, stem cell depletion, accumulation of damaged proteins, or modifications of the epigenome (\\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e7\\u003c/span\\u003e). Many of the age-related diseases such as cancer (\\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e8\\u003c/span\\u003e), neurodegenerative diseases (\\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e9\\u003c/span\\u003e), atherosclerosis (\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e), and inflammation (\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e) are a result of the deregulation of pathways and modification in transcriptomes caused by alterations in DNA methylation. The nine \\u0026ldquo;hallmarks of aging\\u0026rdquo; as enumerated by Lopez-Ot\\u0026iacute;n \\u003cem\\u003eet al.\\u003c/em\\u003e have also been suggested to be responsible (\\u003cspan citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e). On the other hand, prior studies have hypothesized and demonstrated that there exists partial overlap of methylation changes with regions that harbor changes in histone modifications with age, and that approximately one-third of the methylation sites in the genome are affected by age (\\u003cspan additionalcitationids=\\\"CR14\\\" citationid=\\\"CR13\\\" class=\\\"CitationRef\\\"\\u003e13\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e).The above findings indicate that methylation changes could be potential representatives of the natural aging process and age-related phenotypes, and thereby could be used as a predictor of chronological age.\\u003c/p\\u003e\\u003cp\\u003eMethylation patterns, furthermore, have been observed to vary across populations, where they not only impact externally observed phenotypes but also have an effect on underlying health disparities (\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e). Moreover, studying methylation profiles of subjects with identical ethnicity but living in different geographical locations would provide knowledge on conserved methylation patterns within populations. Investigation and evaluation of the epigenome is necessary to understand the ethnicity-specific effects of DNA methylation on aging. Epigenetic clocks have recently emerged as a promising tool for predicting both biological and chronological age (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). There exist several studies that have utilized differentially methylated regions (DMRs) as markers of age; however, they have mostly been directed towards Western populations. In this study, we explored the relationship between aging and changes in DNA methylation specifically among East-Asian cohorts from Taiwan, Japan, and China. A model was proposed for predicting chronological age among East Asians. An accurate epigenetic clock can enable measurement of true chronological age, whether accelerated or not, and estimate a person\\u0026rsquo;s lifespan, allowing interventions to slow the rate of aging and maximize a person\\u0026rsquo;s years of good health.\\u003c/p\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e\\u003ch2\\u003eDatasets\\u003c/h2\\u003e\\u003cp\\u003eThis study utilized methylation data from individuals of East Asian ancestry. A total of 3,637 subjects were included, out of which 2,090 were of Taiwanese origin obtained from the Taiwan Biobank (TWB), and the remaining 1,546 subjects of Chinese, Japanese, or Korean origin were obtained from the Gene Expression Omnibus (GEO) database (see \\u003cb\\u003eSupplementary Table S1\\u003c/b\\u003e for dataset details). Epigenetic data were profiled using DNA from whole blood samples utilizing the Illumina HumanMethylation450 BeadChip array, Infinium MethylationEPIC array, or Illumina HumanMethylation27 BeadChip array. The 27K array data were used solely for initial quality inspections and were excluded from downstream model training due to limited probe overlap. The TWB dataset initially consisted of 865,917 CpG sites across all samples. Datasets from both TWB and GEO were integrated by retaining 398,296 CpG sites that were common to the Illumina 450K and EPIC arrays. Samples from the 27K platform were excluded from this merging step. Raw signal intensity data were processed using standard pipelines described in the published studies associated with each dataset (see \\u003cb\\u003eSupplementary Table S1\\u003c/b\\u003e for references and accession numbers)\\u003c/p\\u003e\\u003cp\\u003eBeta values (ratios of methylated to unmethylated probes for a given CpG site) were used for this analysis instead of M values (standardized Beta values \\u0026ndash;logit transformed) due to the challenges associated with it, such as the potential for infinite or undefined results. Beta values range from 0 to 1 and thus offer a more stable and interpretable methylation level for each CpG site. The individuals in the merged dataset were randomly allocated into two subsets: 2,546 in the training set, which was used to develop the model; and 1,091 in the testing set, which was used for evaluating the model\\u0026rsquo;s performance, ensuring a comprehensive and representative sample for our epigenetic analysis. A total of 398,296 CpG targets that were common to all the datasets were included for analysis.\\u003c/p\\u003e\\u003c/div\\u003e\\n\\u003ch3\\u003eQuality control\\u003c/h3\\u003e\\n\\u003cp\\u003eA series of quality control steps were performed. CpG sites with detection p-values\\u0026thinsp;\\u0026ge;\\u0026thinsp;0.01 and missing rates\\u0026thinsp;\\u0026ge;\\u0026thinsp;5% were excluded. Normalization using Illumina GenomeStudio V2011.1 was conducted before merging the TWB with the GEO datasets to eliminate batch effects. Toward that end, coefficient of variation (CV) analysis was conducted on all CpG targets. The target with the lowest CV was designated as the reference and used as the baseline to normalize each CpG target by subtracting the reference value from each target. After normalization, a simple linear regression was performed with age as the dependent variable and each CpG site as an independent variable. CpG sites that did not meet a Bonferroni-corrected significance threshold were excluded from further analysis. Outlier CpG sites were identified by ranking adjusted coefficient of determination (Adj R\\u0026sup2;) values from linear regression; sites with greater than the third quartile (Q3) plus 0.15 times the interquartile range (IQR) (Adj R\\u0026sup2; \\u0026gt;Q3\\u0026thinsp;+\\u0026thinsp;0.15\\u0026times;IQR) were excluded from further analysis.\\u003c/p\\u003e\\n\\u003ch3\\u003eConstruction of an epigenetic clock for East Asian populations: EAS clock\\u003c/h3\\u003e\\n\\u003cp\\u003eThe CpG targets that passed quality control and statistical filtering were used to construct the chronological age prediction model. A stepwise multivariate linear regression model with forward selection and Bayesian Information Criteria (BIC) was implemented to conduct variable selection for incorporating CpG targets as predictors, into the final prediction model. The final model (EAS clock) was selected as the one with the lowest BIC, ensuring optimal predictor selection while minimizing overfitting. This approach is a widely used statistical method for variable selection that balances model complexity with goodness of fit, and has been commonly applied across various fields including genomics and epidemiology.\\u003c/p\\u003e\\n\\u003ch3\\u003eTraining and testing\\u003c/h3\\u003e\\n\\u003cp\\u003eSubjects from the merged dataset were randomly allocated into two subsets with an approximate training to testing ratio of 2:1. The training set was used to estimate regression coefficients, and the derived prediction equation was applied to the testing set to evaluate EAS clock\\u0026rsquo;s performance, using the Pearson correlation coefficient (r) between predicted and actual chronological age, as well as the mean absolute error (MAE) and coefficient of determination (R\\u0026sup2;) to evaluate both accuracy and explained variance. A further subgroup analysis for different age groups (0\\u0026ndash;20 years, \\u0026gt;\\u0026thinsp;20\\u0026ndash;40 years, \\u0026gt;\\u0026thinsp;40\\u0026ndash;60 years, \\u0026gt;\\u0026thinsp;60\\u0026ndash;80 years, and \\u0026gt;\\u0026thinsp;80\\u0026ndash;100 years) were conducted towards understanding the EAS clock\\u0026rsquo;s efficacy for specific age intervals. Finally, a comparison analysis was conducted with existing epigenetic clocks from EstimAge (\\u003cspan class=\\\"ExternalRef\\\"\\u003e\\u003cspan class=\\\"RefSource\\\"\\u003ehttps://estimage.iac.rm.cnr.it/tutorial\\u003c/span\\u003e\\u003cspan address=\\\"https://estimage.iac.rm.cnr.it/tutorial\\\" targettype=\\\"URL\\\" class=\\\"RefTarget\\\"\\u003e\\u003c/span\\u003e\\u003c/span\\u003e). EstimAge allows identification of specific CpGs in the genome that are globally correlated to chronological age in any tissue or cell type through their methylation state (\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e).\\u003c/p\\u003e\\n\\u003ch3\\u003eFunctional Enrichment Analysis via IPA\\u003c/h3\\u003e\\n\\u003cp\\u003eTo investigate the biological functions associated with age-informative methylation markers, we conducted functional enrichment analysis using Ingenuity Pathway Analysis (IPA, QIAGEN Inc.) (\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e). Genes annotated to the CpG sites included in our final epigenetic clock model were analyzed in the IPA Core Analysis module, using default parameters with the Ingenuity Knowledge Base (Genes Only) as the reference background. Significance of enrichment was assessed using Fisher\\u0026rsquo;s exact test, and disease or function terms with \\u003cem\\u003ep\\u003c/em\\u003e-values\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01 were considered significantly enriched.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec9\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eSample characteristics\\u003c/h2\\u003e\\n \\u003cp\\u003eQuality control measures were implemented on an initial 2,090 subjects from TWB and 1,546 samples from other East Asian countries, providing a total of 398,296 CpG islands across 3,637 individuals for analysis. A total of 2,546 samples were allocated as the training dataset and 1,091 as the testing dataset. The age ranged between 6.40 and 84.00 years, with a mean of 49.98 years (SD\\u0026thinsp;=\\u0026thinsp;12.93) for the training samples, while the testing set comprised subjects of age 1.34\\u0026ndash;85.65 years, with a mean of 47.92 years (SD\\u0026thinsp;=\\u0026thinsp;14.77) (Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). The summary statistics for age are provided in Table\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e.\\u003c/p\\u003e\\n \\u003cdiv class=\\\"gridtable\\\"\\u003e\\n \\u003ctable id=\\\"Tab1\\\" border=\\\"1\\\"\\u003e\\n \\u003ccaption language=\\\"En\\\"\\u003e\\n \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 1\\u003c/div\\u003e\\n \\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\n \\u003cp\\u003eSummary statistics of age for training and testing datasets\\u003c/p\\u003e\\n \\u003c/div\\u003e\\n \\u003c/caption\\u003e\\n \\u003cthead\\u003e\\n \\u003ctr\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eDatasets\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eMinimum\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eFirst quartile\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eMedian\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eMean\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eThird Quartile\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eMaximum\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eStd. Dev\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/thead\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eTraining\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e6.40\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e41.50\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e51.00\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e49.98\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e60.00\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e84.00\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e12.93\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eTesting\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e1.34\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e38.81\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e50.08\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e47.92\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e58.77\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e85.65\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e14.77\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n \\u003c/table\\u003e\\n \\u003c/div\\u003e\\n\\u003c/div\\u003e\\n\\u003ch3\\u003eRegression analysis for model development\\u003c/h3\\u003e\\n\\u003cp\\u003eThe distribution of the CV calculated across all 398,296 CpG sites is demonstrated in Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eA, ranging from 0.96 to 242.90, with a mean (SD) of 51.72 (37.49). The CpG site cg10192265, located at chromosome 20 and at base pair 30,220,446 (GRCh38.p13), corresponding to the SNP rs1979233980, had the lowest CV and was selected as the reference for normalization of the dataset to correct for batch effects, if any, prior to downstream analysis. The Beta value distribution for cg10192265 as shown in Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eB, demonstrates a range of 0.9095 to 0.9941 and a mean (SD) of 0.9638 (0.009).\\u003c/p\\u003e\\n\\u003cp\\u003eEach CpG site was individually tested using a simple linear regression with chronological age as the dependent variable, and 134,635 CpG sites showed significant associations (Bonferroni-adjusted p\\u0026thinsp;\\u0026lt;\\u0026thinsp;1.25E-07). Among these, only 2,087 sites were retained, after eliminating outliers, based on the Adj R\\u003csup\\u003e2\\u003c/sup\\u003e values, to be used as potential predictors of chronological age. These candidate sites were then entered into a stepwise multivariate regression with forward selection, where the Bayesian Information Criterion (BIC) guided the inclusion of 38 CpG sites in the final model (EAS clock). Details of the selected CpG sites and their estimated regression coefficients (based on the training dataset) are provided in \\u003cstrong\\u003eSupplementary Table S2\\u003c/strong\\u003e.\\u003c/p\\u003e\\n\\u003cdiv id=\\\"Sec11\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003ePerformance of the model (EAS clock) with the training and testing datasets\\u003c/h2\\u003e\\n \\u003cp\\u003eIn addition to internal validation using the training data, the performance of the fitted regression model (EAS clock) was also evaluated on an independent testing dataset. Pearson correlation analysis between predicted and actual chronological age showed significant positive correlations of r\\u0026thinsp;=\\u0026thinsp;0.71 (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.0001) in the training set and r\\u0026thinsp;=\\u0026thinsp;0.68 (p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.0001) in the testing set (Table \\u003cspan class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003e, Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e). Furthermore, the distribution of residuals (i.e., estimated age minus chronological age) showed that both the mean and median values were approximately zero in both the training and testing sets (Fig.\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e4\\u003c/span\\u003e), indicating no systematic bias. In addition, the mean absolute error (MAE) was 7.48 years for the training set and 6.22 years for the testing set. These results collectively demonstrated that our epigenetic clock, EAS clock, achieved strong predictive performance, with precision and accuracy comparable to or exceeding other models reported in population-level epigenetic aging studies.\\u003c/p\\u003e\\n \\u003cdiv class=\\\"gridtable\\\"\\u003e\\n \\u003ctable id=\\\"Tab2\\\" border=\\\"1\\\"\\u003e\\n \\u003ccaption language=\\\"En\\\"\\u003e\\n \\u003cdiv class=\\\"CaptionNumber\\\"\\u003eTable 2\\u003c/div\\u003e\\n \\u003cdiv class=\\\"CaptionContent\\\"\\u003e\\n \\u003cp\\u003eCorrelation between actual age and age estimated using methylation targets as predictors\\u003c/p\\u003e\\n \\u003c/div\\u003e\\n \\u003c/caption\\u003e\\n \\u003cthead\\u003e\\n \\u003ctr\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eData sets\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eCorrelation\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eEstimate\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eStandard Error\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eT-statistics\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003cth align=\\\"left\\\"\\u003e\\n \\u003cp\\u003eP-value\\u003c/p\\u003e\\n \\u003c/th\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/thead\\u003e\\n \\u003ctbody\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eTraining\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e0.71\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e1.00\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e1.95E-02\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e51.20\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.0001\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003ctr\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eTesting\\u003c/strong\\u003e\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e0.68\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e0.94\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"left\\\"\\u003e\\n \\u003cp\\u003e3.16E-02\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e29.80\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003ctd align=\\\"char\\\"\\u003e\\n \\u003cp\\u003e\\u0026lt;\\u0026thinsp;0.0001\\u003c/p\\u003e\\n \\u003c/td\\u003e\\n \\u003c/tr\\u003e\\n \\u003c/tbody\\u003e\\n \\u003c/table\\u003e\\n \\u003c/div\\u003e\\n\\u003c/div\\u003e\\n\\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eAge-subgroup analysis\\u003c/h2\\u003e\\n \\u003cp\\u003eAs shown in Fig. \\u003cspan class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e, individuals exhibited distinct clustering patterns based on their epigenetic profiles, which aligning closely with chronological age. Notably, the distribution suggested a nonlinear relationship between DNA methylation and age, particularly during early childhood and advanced age, indicating that the epigenetic aging process may follow different trajectories across life stages.\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003cdiv id=\\\"Sec13\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eComparison with EstimAge\\u003c/h2\\u003e\\n \\u003cp\\u003eSix samples with the largest prediction errors using our proposed model were fed into the EstimAge platform, which provides epigenetic age estimates based on CpG sites strongly correlated with chronological age across various tissues and cell types. A comparison of the predicted age by our method (EAS Clock) and other existing models including Epigenetic Pacemaker (EPM), Hannum 13, Horvath 13 and18, PhenoAge, Zhang Enpred, and Zhang Blupred were evaluated using the actual chronological age as the reference. Table\\u0026nbsp;\\u003cspan class=\\\"InternalRef\\\"\\u003e3\\u003c/span\\u003e lists the differences between the predicted age and the actual age for each of the methods, with red and blue indicating over- and underestimation, respectively. While deviations were observed in our model, similar or greater discrepancies were also present in the other methods, highlighting the challenge of age prediction in these outlier cases. The EPM method was found to perform the best, likely due to its use of 20,031 CpG sites. In contrast, our model relies on only 38 CpG sites, offering a far more interpretable and efficient alternative while maintaining competitive performance.\\u003c/p\\u003e\\n \\u003cp\\u003e\\u003cstrong\\u003eTable 3. Comparative analysis of the deviations of predicted age from chronological age for different epigenetic clocks and EAS clock\\u0026nbsp;\\u003c/strong\\u003e\\u0026nbsp;\\u003c/p\\u003e\\n \\u003cdiv class=\\\"gridtable\\\"\\u003e\\n \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\"\\u003e\\u003cimg src=\\\"data:image/png;base64,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\\\"\\u003e\\u003c/div\\u003e\\n \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\"\\u003e\\u003cbr\\u003e\\u003c/div\\u003e\\n \\u003cdiv align=\\\"left\\\" class=\\\"colspec\\\"\\u003e\\u003cem\\u003eValues represent residuals calculated as (Chronological Age\\u0026thinsp;\\u0026minus;\\u0026thinsp;Predicted Age). Blue text indicates negative residuals (i.e., predicted age older than actual age), and red text indicates positive residuals (i.e., predicted age younger than actual age)\\u003c/em\\u003e\\u003c/div\\u003e\\n \\u003c/div\\u003e\\n \\u003cp\\u003eTo further evaluate EAS clock\\u0026rsquo;s performance, 273 samples randomly selected from the testing dataset were submitted to the EstimAge platform for comparative prediction analysis. The residual distributions of the EstimAge models were compared against those of the EAS epigenetic clock across the 273 samples. Residuals were calculated as the difference between the estimated age and the actual chronological age. Figure \\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eA demonstrates the residual distributions of estimated age predicted by EPM_0.65, Hannum\\u0026apos;s 2013 model, Horvath\\u0026apos;s 2013 and 2018 models, EAS clock, PhenoAge, Zhang\\u0026apos;s Enpred, Blupred models, and GrimAge. Notably, the EPM_0.65 model exhibited the widest spread of residuals, with several predictions deviating substantially from zero\\u0026mdash;suggesting less consistent performance across samples compared to our model (EAS clock). Figure \\u003cspan class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003eB demonstrates the comparison after excluding the EPM_0.65 model to provide a cleaner and more focused picture of the residual distributions among the remaining epigenetic clock models. Notably, EAS clock stood out with residuals tightly clustered around zero, indicating high precision and accuracy in age estimation. This suggests that the clock is well-calibrated to the underlying biological characteristics of the dataset. Furthermore, the median residual of EAS clock was closest to zero among all methods, underscoring its robust fit for East Asian populations.\\u003c/p\\u003e\\n\\u003c/div\\u003e\\n\\u003cdiv id=\\\"Sec14\\\" class=\\\"Section2\\\"\\u003e\\n \\u003ch2\\u003eFunctional enrichment analysis\\u003c/h2\\u003e\\n \\u003cp\\u003eTo explore the biological functions associated with the 38 CpG sites in our EAS clock, we submitted their annotated genes to IPA\\u0026rsquo;s Core Analysis and examined the top 25 enriched Disease \\u0026amp; Function terms (Fisher\\u0026rsquo;s exact test, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.01; \\u003cstrong\\u003eSupplementary Table S3\\u003c/strong\\u003e). The most significant annotations fell into several coherent categories. Neurodegenerative and neurodevelopmental processes were heavily represented, with terms such as \\u0026ldquo;Grade 3\\u0026ndash;4 glioma cancer\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;4.22 \\u0026times; 10⁻⁶), \\u0026ldquo;Grade 4 astrocytoma\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;6.83 \\u0026times; 10⁻⁶), \\u0026ldquo;brain astrocytoma\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;7.00 \\u0026times; 10⁻⁶) and \\u0026ldquo;dementia\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;1.10 \\u0026times; 10⁻⁵). Musculoskeletal decline emerged via enrichments like \\u0026ldquo;curvature of spine\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;2.80 \\u0026times; 10⁻⁵) and \\u0026ldquo;osteoporosis\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;4.00 \\u0026times; 10⁻⁵). Immune and inflammatory signatures appeared in terms such as \\u0026ldquo;macrophage activation\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;6.20 \\u0026times; 10⁻⁵) and \\u0026ldquo;cytokine signaling\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;8.10 \\u0026times; 10⁻⁵). Finally, cancer-related pathways beyond the central nervous system\\u0026mdash;e.g. \\u0026ldquo;skin cancer\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;2.10 \\u0026times; 10⁻⁵) and \\u0026ldquo;polycystic ovary disease\\u0026rdquo; (p\\u0026thinsp;=\\u0026thinsp;5.50 \\u0026times; 10⁻⁵)\\u0026mdash;were also enriched. Across these categories, several genes recurred, most notably \\u003cem\\u003eELOVL2\\u003c/em\\u003e, \\u003cem\\u003eDNMT3A\\u003c/em\\u003e, and \\u003cem\\u003eCHRNA9\\u003c/em\\u003e, each implicated in multiple enriched annotations. Together, these results indicate that the CpG markers driving our East Asian clock are functionally embedded in key biological pathways underlying systemic aging.\\u003c/p\\u003e\\n\\u003c/div\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eGiven the well-established role of age-associated epigenetic modifications, particularly the global decline of DNA methylation with age, methylation is widely accepted as a reliable proxy for estimating chronological age (\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). In this study, we specifically focused on East Asian ancestry and constructed an epigenetic clock, EAS clock, based on methylation profiles, demonstrating accurate and reliable prediction of chronological age. Rigorous evaluations showed that the EAS clock performed in a robust manner and predicted age with high accuracy and precision. We further utilized the methylation patterns to conduct a subgroup analysis for different age groups to evaluate the EAS clock\\u0026rsquo;s efficacy for specific age intervals which demonstrated that the rate and pattern of epigenetic changes doesn\\u0026rsquo;t progress uniformly through the lifespan. Residuals, defined as actual age minus predicted age, showed an increase in variance at the extremes of the age spectrum, particularly among older adults. This indicated potential overfitting or that age-related methylation dynamics possibly reaches a plateau or exhibits nonlinear trajectories in older individuals, limiting model resolution at advanced ages.\\u003c/p\\u003e\\u003cp\\u003eTo further elucidate the biological relevance of age-associated methylation changes, we performed functional enrichment analysis using IPA. Results revealed that CpG-associated genes in our model are enriched for aging-related functions, including neurodegeneration (\\u003cem\\u003ee.g., dementia\\u003c/em\\u003e), musculoskeletal decline (\\u003cem\\u003ee.g., curvature of spine\\u003c/em\\u003e), and immune/inflammatory pathways. Although cancer-related terms were also enriched, these likely reflect biological overlap with aging rather than model bias. For clarity, we prioritized age- and development-related functions in our interpretation. The complete list of top 25 annotations is available in \\u003cb\\u003eSupplementary Table S3\\u003c/b\\u003e. Notably, several genes recurred across multiple enriched terms, suggesting they may play a central role in aging-related biological pathways. Among these, \\u003cem\\u003eELOVL2\\u003c/em\\u003e is one of the most well-documented age-associated loci: its promoter methylation levels rise consistently with age in various tissues and it has been widely used as a single-locus biomarker for epigenetic aging. Experimental studies have shown that restoring \\u003cem\\u003eELOVL2\\u003c/em\\u003e function in aged mice can reverse certain vision impairments, implicating a potentially causal role in age-related functional decline (\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e). \\u003cem\\u003eDNMT3A\\u003c/em\\u003e, a de novo DNA methyltransferase, is involved in the establishment of methylation patterns during development and hematopoietic stem cell renewal. Loss-of-function mutations in \\u003cem\\u003eDNMT3A\\u003c/em\\u003e have been linked to clonal hematopoiesis and accelerated epigenetic aging (\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e). Finally, \\u003cem\\u003eCHRNA9\\u003c/em\\u003e, a subunit of nicotinic acetylcholine receptors, appeared in terms related to auditory cell morphology and neurobiology. Dysregulation of \\u003cem\\u003eCHRNA9\\u003c/em\\u003e has been implicated in hearing loss and neuroinflammation, both of which are common features of aging (\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). These findings support the notion that the CpG sites selected for our East Asian epigenetic clock are not only predictive of chronological age but are also functionally embedded in pathways associated with systemic aging and tissue degeneration. Their recurrence in age-related annotations reinforces the biological interpretability and relevance of our model\\u0026rsquo;s features. The non-linear relationship between methylation and age, especially in early life and advanced age, as shown in our subgroup analysis (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig5\\\" class=\\\"InternalRef\\\"\\u003e5\\u003c/span\\u003e), may stem from rapid methylation changes during early development, leading to greater inter-individual variability, and a biological plateau in older age, reducing sensitivity to age differences. These dynamics likely contribute to reduced model correlations at age extremes and underscore a key challenge for epigenetic clocks. Future models could improve accuracy by incorporating age-specific trajectories or flexible non-linear approaches such as splines or Gaussian processes to better capture these complex patterns across the lifespan.\\u003c/p\\u003e\\u003cp\\u003eNumerous epigenetic clocks have been developed, among which Horvath\\u0026rsquo;s clock is the most widely cited and extensively validated. It was developed using methylation data from diverse tissues and conditions, including cancer (\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e), Alzheimer\\u0026rsquo;s disease (\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e), aging (\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e), and lifestyle factor (\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e). Notably, it was the first age estimator to leverage methylation profiles across multiple human tissues and developmental stages. The model was constructed using elastic net regression, which automatically selected 353 age-associated CpG sites\\u0026mdash;193 positively and 160 negatively correlated with age. The correlation between predicted age and chronological age, for Horvath's clock, was 0.96 with a median absolute difference of 3.6 years (\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e). Another one is Hannum\\u0026rsquo;s epigenetic clock, which is a single-tissue DNA methylation-based age predictor, which was created by training an elastic net regression model utilizing 71 age-related CpGs from 482 Caucasian and 174 Hispanic adults (\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e). In Hannum\\u0026rsquo;s epigenetic clock, the correlation between predicted and chronological age was 0.96, with a median absolute difference of 3.9 years; however, this clock has some bias in estimation when applied to non-blood tissues (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e). Other established epigenetic clocks includes the epigenetic pacemaker (EPM) (\\u003cspan citationid=\\\"CR34\\\" class=\\\"CitationRef\\\"\\u003e34\\u003c/span\\u003e), PhenoAge (\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e), and BluPred, Enpred (\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e), and GrimAge (\\u003cspan citationid=\\\"CR37\\\" class=\\\"CitationRef\\\"\\u003e37\\u003c/span\\u003e). The age predicted by these models is referred to as \\u0026ldquo;epigenetic age\\u0026rdquo;, which often deviates from an individual\\u0026rsquo;s chronological age. This difference, often termed as the residual, indicates \\u0026ldquo;epigenetic age acceleration\\u0026rdquo; when epigenetic age exceeds chronological age, and \\u0026ldquo;epigenetic age deceleration\\u0026rdquo; when epigenetic age is lower than chronological age. Exploring this gap between chronological age and epigenetic age is a major research focus in the field of aging. We compared our proposed epigenetic clock with the aforementioned models, and found that it produced the smallest residuals, tightly clustered around zero. This suggested superior accuracy of EAS clock in age estimation, particularly for East Asian individuals. As multiple models may yield similar results for easily predictable samples, we specifically evaluated the performance of EAS clock as opposed to other existing epigenetic clocks using more challenging or borderline cases. Additionally, we emphasized comparisons with population-matched models\\u0026mdash;those trained on East Asian datasets\\u0026mdash;to ensure fair and interpretable benchmarking, particularly for samples that may be misestimated by clocks trained on other ancestries. Furthermore, EAS clock, as opposed to other existing epigenetic clocks, was intentionally designed to use a minimal number of CpG targets (38 only), selected based on statistical significance and model fit criteria. This design choice reflects a trade-off between model complexity and practical utility. While methods like EPM achieved higher accuracy by incorporating tens of thousands of CpG sites (e.g., 20,031 in EPM), they require extensive data and computational resources, which may limit their feasibility in routine clinical or screening settings. In contrast, our EAS Clock model provides a more parsimonious alternative, potentially allowing for easier implementation, lower cost, and better generalizability in real-world applications, despite a modest reduction in predictive accuracy. We believe this balance makes EAS clock particularly suitable for scalable use in health examination contexts.\\u003c/p\\u003e\\u003cp\\u003eIn this study, we compared our EAS clock with several widely used epigenetic clocks that estimate chronological age, including GrimAge, Horvath\\u0026rsquo;s 2013 and 2018 clocks, Hannum\\u0026rsquo;s clock, PhenoAge, EPM, an Zhang\\u0026rsquo;s Enpred and Blupred clocks. Among these, GrimAge was successfully applied to our data where applicable, which exhibited residual distributions highly similar to those of PhenoAge (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig6\\\" class=\\\"InternalRef\\\"\\u003e6\\u003c/span\\u003e), suggesting potential convergence in the biological aging dimensions they capture. Another popular model, DunedinPACE (\\u003cspan citationid=\\\"CR38\\\" class=\\\"CitationRef\\\"\\u003e38\\u003c/span\\u003e), which estimates the pace of aging rather than accumulated age, was not included in this study for comparison, due to its conceptual difference and data requirements with that of the other epigenetic clocks including EAS clock. While it offers valuable insights into the rate of biological aging, it is not directly comparable to our model\\u0026rsquo;s aim of estimating chronological age. Future studies may incorporate DunedinPACE to explore longitudinal aging trajectories in East Asian populations.\\u003c/p\\u003e\\u003cp\\u003eEpigenetic clocks play important roles in aging research. For one, they are used for evaluating the efficacy of interventions for aging. These interventions may reverse epigenetic age, thereby mitigating the physical manifestations of aging. Promising approaches that target the epigenome to promote rejuvenation include cellular reprogramming, pharmacological treatments, and lifestyle modifications. To gain a better understanding of the connection between epigenetic age and chronological age, it is essential to conduct longitudinal cohort studies, which follow individuals over extended timeframes, providing valuable insights into the phenotypic changes that occur as they age and the factors that cause diseases (\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e). A meta-analysis by Marioni et al. revealed that epigenetic aging progresses slightly more slowly than chronological aging over the life course, especially in older adults (\\u003cspan citationid=\\\"CR40\\\" class=\\\"CitationRef\\\"\\u003e40\\u003c/span\\u003e). Other studies have also reported a non-linear, logarithmic pattern of epigenetic aging during adolescence (\\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR41\\\" class=\\\"CitationRef\\\"\\u003e41\\u003c/span\\u003e). Longitudinal cohort studies can potentially identify deviations between epigenetic and chronological age, enabling early-life interventions (\\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e). However, the influence of genetics and signaling pathways on DNA methylation across different stages of aging remains unclear (\\u003cspan additionalcitationids=\\\"CR43 CR44\\\" citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e).\\u003c/p\\u003e\\u003cp\\u003eA key limitation of current epigenetic clocks lies in their variability across CpG targets, Illumina array platforms, tissue sources, and population ancestry. Such inconsistencies contribute to considerable heterogeneity across studies and constrain the generalizability and applicability of epigenetic clocks. Utilizing tissue-specific CpGs to construct clocks can potentially lead to more accurate and robust predictions. However, adjusting for cell type heterogeneity remains essential when estimating age from DNA obtained from mixed or diverse cell populations (\\u003cspan citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e). Future studies will be conducted leveraging multiple accessible DNA sources to produce more accurate and comprehensive age estimates (\\u003cspan citationid=\\\"CR39\\\" class=\\\"CitationRef\\\"\\u003e39\\u003c/span\\u003e). Moreover, additional biological markers will be integrated alongside DNA methylation to enhance the current model\\u0026rsquo;s (EAS clock) ability to estimate biological age. Although the TWB dataset contains limited clinical and lifestyle variables, our analytic strategy required merging it with GEO datasets, which do not provide such annotations. As a result, we were unable to explore the associations between epigenetic age acceleration and clinical outcomes or environmental exposures. This limitation underscores the importance of using well-annotated clinical cohorts in future research to assess the broader biological relevance and potential health applications of our model in aging and precision medicine.\\u003c/p\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eThis study presents a robust and accurate epigenetic clock, EAS clock, tailored to East Asian populations for estimating chronological age. While it performs well in most scenarios, its use may be limited when biological age diverges significantly from chronological age. Nonetheless, such discrepancies may themselves offer insights into age acceleration or deceleration. In the current era of increasing life expectancy, early and accurate prediction of epigenetic age could facilitate timely anti-aging interventions and targeted treatments for age-related diseases. Such efforts may ultimately ease the aging process for individuals and reduce the burden on healthcare systems.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eEthics approval and consent to participate\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe study was approved by the institutional review board (IRB # is: 201506095RINC) of the National Taiwan University Hospital. All subjects provided written consent for participation.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConsent for publication\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eNot applicable\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and materials\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe data s that support the findings of this study belong to Taiwan Biobank, which requires permission for access. The data is available from the corresponding author on reasonable request, upon permission obtained from Taiwan Biobank.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare that they have no competing interests.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis work was partly supported by National Science and Technology Council, Taiwan (MOST-109-2314-B-002-151-MY3, NSTC-113-2314-B-002-170-MY3, and 114-2314-B-002-056-) and Population Health and Welfare Research Center from Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan (grant number NTU-113 L9004).The funders had no role in the study design, data collection and analysis, the decision to publish, or preparation of the manuscript.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors' contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eT.P.L. conceived and designed the study. P.H.K. conducted data acquisition. T.P.L and A.C. provided the resources and administrative support. A.C., and T.P.L supervised the analysis. K.C.L did the formal analysis. K.C.L. and A.C. interpreted data. A.C. drafted the work. K.C.L, P.H.K., T.P.L, and A.C. reviewed and approved the manuscript.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eWe thank Dr. Melissa Stauffer for English editing our manuscript.\\u0026nbsp;\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n \\u003cli\\u003eNewgard CB, Sharpless NE. Coming of age: molecular drivers of aging and therapeutic opportunities. The Journal of clinical investigation. 2013;123(3):946-50.\\u003c/li\\u003e\\n \\u003cli\\u003ePartridge L, Deelen J, Slagboom P. Facing up to the global challenges of ageing. Nature [Internet]. 2018; 561 (7721): 45\\u0026ndash;56.\\u003c/li\\u003e\\n \\u003cli\\u003eGreenberg MV, Bourc\\u0026rsquo;his D. The diverse roles of DNA methylation in mammalian development and disease. Nature reviews Molecular cell biology. 2019;20(10):590-607.\\u003c/li\\u003e\\n \\u003cli\\u003eSingal R, Ginder GD. DNA methylation. Blood, The Journal of the American Society of Hematology. 1999;93(12):4059-70.\\u003c/li\\u003e\\n \\u003cli\\u003eAndersen GB, Tost J. A summary of the biological processes, disease-associated changes, and clinical applications of DNA methylation. DNA Methylation Protocols. 2018:3-30.\\u003c/li\\u003e\\n \\u003cli\\u003eJung M, Pfeifer GP. Aging and DNA methylation. BMC Biol. 2015;13:1-8.\\u003c/li\\u003e\\n \\u003cli\\u003eAlmeida AJPOd, Ribeiro TP, Medeiros IAd. Aging: molecular pathways and implications on the cardiovascular system. Oxid Med Cell Longev. 2017;2017(1):7941563.\\u003c/li\\u003e\\n \\u003cli\\u003eJones PA. DNA methylation and cancer. Oncogene. 2002;21(35):5358-60.\\u003c/li\\u003e\\n \\u003cli\\u003eFodder K, de Silva R, Warner TT, Bettencourt C. The contribution of DNA methylation to the (dys) function of oligodendroglia in neurodegeneration. Acta Neuropathologica Communications. 2023;11(1):106.\\u003c/li\\u003e\\n \\u003cli\\u003eTabaei S, Tabaee SS. DNA methylation abnormalities in atherosclerosis. Artificial cells, nanomedicine, and biotechnology. 2019;47(1):2031-41.\\u003c/li\\u003e\\n \\u003cli\\u003eLigthart S, Marzi C, Aslibekyan S, Mendelson MM, Conneely KN, Tanaka T, et al. DNA methylation signatures of chronic low-grade inflammation are associated with complex diseases. Genome Biol. 2016;17:1-15.\\u003c/li\\u003e\\n \\u003cli\\u003eL\\u0026oacute;pez-Ot\\u0026iacute;n C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194-217.\\u003c/li\\u003e\\n \\u003cli\\u003eJohansson \\u0026Aring;, Enroth S, Gyllensten U. Continuous aging of the human DNA methylome throughout the human lifespan. PLoS One. 2013;8(6):e67378.\\u003c/li\\u003e\\n \\u003cli\\u003eTorres IO, Fujimori DG. Functional coupling between writers, erasers and readers of histone and DNA methylation. Curr Opin Struct Biol. 2015;35:68-75.\\u003c/li\\u003e\\n \\u003cli\\u003eSaul D, Kosinsky RL. Epigenetics of aging and aging-associated diseases. International journal of molecular sciences. 2021;22(1):401.\\u003c/li\\u003e\\n \\u003cli\\u003eGiuliani C, Sazzini M, Bacalini MG, Pirazzini C, Marasco E, Fontanesi E, et al. Epigenetic variability across human populations: a focus on DNA methylation profiles of the KRTCAP3, MAD1L1 and BRSK2 genes. Genome Biol Evol. 2016;8(9):2760-73.\\u003c/li\\u003e\\n \\u003cli\\u003eDuan R, Fu Q, Sun Y, Li Q. Epigenetic clock: A promising biomarker and practical tool in aging. Ageing Res Rev. 2022;81:101743.\\u003c/li\\u003e\\n \\u003cli\\u003eDi Lena P, Sala C, Nardini C. Estimage: a webserver hub for the computation of methylation age. Nucleic Acids Res. 2021;49(W1):W199-W206.\\u003c/li\\u003e\\n \\u003cli\\u003eInc. Q. Ingenuity Pathway Analysis (IPA). Version 01-23-01 ed. Redwood City, CA: QIAGEN Inc.; 2025.\\u003c/li\\u003e\\n \\u003cli\\u003eHillary RF, Stevenson AJ, McCartney DL, Campbell A, Walker RM, Howard DM, et al. Epigenetic measures of ageing predict the prevalence and incidence of leading causes of death and disease burden. Clin Epigenetics. 2020;12:1-12.\\u003c/li\\u003e\\n \\u003cli\\u003eBelsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022;11:e73420.\\u003c/li\\u003e\\n \\u003cli\\u003eJaiswal S, Fontanillas P, Flannick J, Manning A, Grauman Peter V, Mar Brenton G, et al. Age-Related Clonal Hematopoiesis Associated with Adverse Outcomes. New England Journal of Medicine.371(26):2488-98.\\u003c/li\\u003e\\n \\u003cli\\u003eElgoyhen AB, Katz E, Fuchs PA. The nicotinic receptor of cochlear hair cells: a possible pharmacotherapeutic target? Biochem Pharmacol. 2009;78(7):712-9.\\u003c/li\\u003e\\n \\u003cli\\u003eLevine ME, Hosgood HD, Chen B, Absher D, Assimes T, Horvath S. DNA methylation age of blood predicts future onset of lung cancer in the women\\u0026apos;s health initiative. Aging. 2015;7(9):690.\\u003c/li\\u003e\\n \\u003cli\\u003eSehl ME, Henry JE, Storniolo AM, Ganz PA, Horvath S. DNA methylation age is elevated in breast tissue of healthy women. Breast Cancer Res Treat. 2017;164:209-19.\\u003c/li\\u003e\\n \\u003cli\\u003eMcCartney DL, Stevenson AJ, Walker RM, Gibson J, Morris SW, Campbell A, et al. Investigating the relationship between DNA methylation age acceleration and risk factors for Alzheimer\\u0026apos;s disease. Alzheimer\\u0026apos;s \\u0026amp; Dementia: Diagnosis, Assessment \\u0026amp; Disease Monitoring. 2018;10:429-37.\\u003c/li\\u003e\\n \\u003cli\\u003eDeclerck K, Berghe WV. Back to the future: epigenetic clock plasticity towards healthy aging. Mech Ageing Dev. 2018;174:18-29.\\u003c/li\\u003e\\n \\u003cli\\u003eRyan J, Wrigglesworth J, Loong J, Fransquet PD, Woods RL. A systematic review and meta-analysis of environmental, lifestyle, and health factors associated with DNA methylation age. The Journals of Gerontology: Series A. 2020;75(3):481-94.\\u003c/li\\u003e\\n \\u003cli\\u003eOblak L, van der Zaag J, Higgins-Chen AT, Levine ME, Boks MP. A systematic review of biological, social and environmental factors associated with epigenetic clock acceleration. Ageing Res Rev. 2021;69:101348.\\u003c/li\\u003e\\n \\u003cli\\u003eHorvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14:1-20.\\u003c/li\\u003e\\n \\u003cli\\u003eHannum G, Guinney J, Zhao L, Zhang L, Hughes G, Sadda S, et al. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell. 2013;49(2):359-67.\\u003c/li\\u003e\\n \\u003cli\\u003eSimpkin AJ, Hemani G, Suderman M, Gaunt TR, Lyttleton O, Mcardle WL, et al. Prenatal and early life influences on epigenetic age in children: a study of mother\\u0026ndash;offspring pairs from two cohort studies. Human molecular genetics. 2016;25(1):191-201.\\u003c/li\\u003e\\n \\u003cli\\u003eSimpkin AJ, Howe LD, Tilling K, Gaunt TR, Lyttleton O, McArdle WL, et al. The epigenetic clock and physical development during childhood and adolescence: longitudinal analysis from a UK birth cohort. International journal of epidemiology. 2017;46(2):549-58.\\u003c/li\\u003e\\n \\u003cli\\u003eFarrell C, Snir S, Pellegrini M. The Epigenetic Pacemaker: modeling epigenetic states under an evolutionary framework. Bioinformatics. 2020;36(17):4662-3.\\u003c/li\\u003e\\n \\u003cli\\u003eLevine ME, Lu AT, Quach A, Chen BH, Assimes TL, Bandinelli S, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging. 2018;10(4):573.\\u003c/li\\u003e\\n \\u003cli\\u003eZhang Q, Vallerga CL, Walker RM, Lin T, Henders AK, Montgomery GW, et al. Improved precision of epigenetic clock estimates across tissues and its implication for biological ageing. Genome medicine. 2019;11:1-11.\\u003c/li\\u003e\\n \\u003cli\\u003eLu AT, Binder AM, Zhang J, Yan Q, Reiner AP, Cox SR, et al. DNA methylation GrimAge version 2. Aging (Albany NY). 2022;14(23):9484-549.\\u003c/li\\u003e\\n \\u003cli\\u003eBelsky DW, Caspi A, Corcoran DL, Sugden K, Poulton R, Arseneault L, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. Elife. 2022;11.\\u003c/li\\u003e\\n \\u003cli\\u003eHorvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nature reviews genetics. 2018;19(6):371-84.\\u003c/li\\u003e\\n \\u003cli\\u003eMarioni RE, Suderman M, Chen BH, Horvath S, Bandinelli S, Morris T, et al. Tracking the epigenetic clock across the human life course: a meta-analysis of longitudinal cohort data. The Journals of Gerontology: Series A. 2019;74(1):57-61.\\u003c/li\\u003e\\n \\u003cli\\u003eAlisch RS, Barwick BG, Chopra P, Myrick LK, Satten GA, Conneely KN, et al. Age-associated DNA methylation in pediatric populations. Genome Res. 2012;22(4):623-32.\\u003c/li\\u003e\\n \\u003cli\\u003eGaunt TR, Shihab HA, Hemani G, Min JL, Woodward G, Lyttleton O, et al. Systematic identification of genetic influences on methylation across the human life course. Genome Biol. 2016;17:1-14.\\u003c/li\\u003e\\n \\u003cli\\u003eVan Dongen J, Nivard MG, Willemsen G, Hottenga J-J, Helmer Q, Dolan CV, et al. Genetic and environmental influences interact with age and sex in shaping the human methylome. Nature communications. 2016;7(1):11115.\\u003c/li\\u003e\\n \\u003cli\\u003eBell CG, Gao F, Yuan W, Roos L, Acton RJ, Xia Y, et al. Obligatory and facilitative allelic variation in the DNA methylome within common disease-associated loci. Nature communications. 2018;9(1):8.\\u003c/li\\u003e\\n \\u003cli\\u003eHannon E, Knox O, Sugden K, Burrage J, Wong CC, Belsky DW, et al. Characterizing genetic and environmental influences on variable DNA methylation using monozygotic and dizygotic twins. PLoS genetics. 2018;14(8):e1007544.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"},{\"header\":\"Supplementary Tables\",\"content\":\"Supplementary tables S1-S3 are not available with this version. \"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":true,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"Epigenetic clock, East-Asian, DNA-methylation, chronological age, Epigenetic age\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-7327305/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-7327305/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eBackground:\\u003c/strong\\u003e The rapid rise in the older population over the past decade risks significant burden on healthcare, families, and society. This has drawn significant attention to aging research. DNA methylation is a heritable epigenetic alteration that is known to be linked with developmental processes via physiological and disease-associated changes. Hence, methylation changes are potential representatives of the natural aging process and age-related phenotypes, and are used as a predictor of chronological age. In this study, we explored the relationship between aging and changes in DNA methylation specifically among East-Asian (EAS) cohorts from Taiwan, Japan, and China to develop an epigenetic clock.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eMethods\\u003c/strong\\u003e: Following quality control, methylation data from EAS samples were used to develop a predictive model, east-Asian epigenetic clock (EAS clock). A stepwise multivariate regression model with forward-selection and Bayesian Information Criteria (BIC) was implemented to conduct variable selection. EAS clock’s performance was validated through rigorous statistical evaluation. Subgroup analyses across age intervals were conducted to assess age-specific efficacy. Additionally, functional enrichment analysis using Ingenuity Pathway Analysis (IPA) was performed to investigate the biological relevance of the selected CpG sites.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults\\u003c/strong\\u003e: Correlation analysis between predicted and actual chronological age showed strong positive correlations in both training (r = 0.71, p \\u0026lt; 0.0001) and testing (r = 0.68, p \\u0026lt; 0.0001) sets. Difference between estimated age by the EAS clock and chronological age showed an approximate median and mean value of zero. Subgroup analysis implied that epigenetic aging may vary across the lifespan, especially at age extremes. Functional annotation revealed enrichment of CpG-associated genes in age-related pathways, including neurodegeneration, musculoskeletal disorders, and immune regulation. Compared with other methylation clocks, EAS clock demonstrated tighter residual clustering around zero, indicating improved accuracy.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusion\\u003c/strong\\u003e: EAS clock, a robust and accurate epigenetic clock tailored to East Asian populations was developed. Early and precise epigenetic age prediction may support timely anti-aging interventions and disease management, potentially mitigating the individual and healthcare burden of aging.\\u003c/p\\u003e\",\"manuscriptTitle\":\"An epigenetic clock for chronological age estimation in East Asian populations\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-09-22 10:10:13\",\"doi\":\"10.21203/rs.3.rs-7327305/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"a842fc74-3f13-429a-a2b8-dea59ea38b01\",\"owner\":[],\"postedDate\":\"September 22nd, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"posted\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2025-11-12T18:23:33+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-09-22 10:10:13\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-7327305\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-7327305\",\"identity\":\"rs-7327305\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}