Genome-based medicine in Korea: the Korea National Institute of Health infrastructure for precision medicine

Abstract

Since the completion of the Human Genome Project, genome-based medicine has progressed from a predominantly research-driven endeavor to a field of increasing clinical relevance. In Korea, the Korea National Institute of Health (KNIH) has played a central role in the establishment of the necessary research infrastructure that supports the secure and responsible use of genomic and clinical data. These efforts have enabled the generation of comprehensive genomic datasets representative of the Korean population and, together with the Korea Biobank Array optimized for population-specific variants, have strengthened discovery-driven research and accelerated advances in disease gene identification and risk prediction. More recently, KNIH has expanded analyses based on whole-genome sequencing data to support clinical translation, enabling more comprehensive variant detection and facilitating the application of genomic information to disease diagnosis and precision medicine research. These national genomic resources provide an important foundation for improving the diagnosis and management of genetically mediated conditions, including pediatric kidney diseases, where early etiologic diagnosis can substantially influence clinical decision-making and long-term outcomes. Further strengthening of institutional and regulatory frameworks will be essential to support routine clinical implementation and maximize the public health impact of genomics in Korea.

Introduction

Genomic medicine has undergone rapid transformation in the past two decades, evolving from a predominantly discovery-driven scientific field to a clinically actionable discipline. Advances in next-generation sequencing technologies have substantially reduced both the cost and turnaround time of whole-exome and whole-genome sequencing (WGS), facilitating their widespread application in research and increasing integration into clinical practice [1,2]. These developments have had a particularly significant impact on rare diseases and oncology, where genomic analyses allow etiologic diagnosis, improve understanding of disease mechanisms, and support the implementation of targeted therapeutic strategies. This progress is especially relevant for pediatric nephrology, as a substantial proportion of kidney diseases in children have a monogenic or strong genetic basis. It is estimated that approximately 30%–50% of chronic kidney disease in pediatric populations is attributable to genetic causes, positioning genomic testing as an essential tool for precise diagnosis, prognostic assessment, and individualized clinical management [3,4]. Furthermore, the clinical and public health value of genomic medicine increasingly depends on the ability to interpret genomic variation within well-characterized population contexts, to link genomic findings with longitudinal clinical outcomes, and to integrate genomic data with complementary modalities including imaging, environmental exposures, and real-world health data. Together, these advances are reshaping the diagnostic and therapeutic landscape of pediatric kidney disease and underscore the growing importance of genomics in advancing precision medicine in pediatric nephrology.

International experience demonstrates that successful implementation of genome-based medicine requires sustained governmental commitment and coordinated large-scale infrastructure. In the United Kingdom, Genomics England—embedded within the National Health Service—illustrates systematic integration of genomic medicine into a public healthcare system, while the UK Biobank serves as a globally influential population resource linking genomic data with deep phenotyping and longitudinal health information [5,6]. In the United States, long-term federal investment has enabled a population-inclusive precision medicine infrastructure through the National Institutes of Health (NIH) All of Us Research Program, which integrates genomic, clinical, behavioral, and environmental data in a controlled-access research ecosystem [7]. These models underscore a shared lesson: genomic medicine becomes scalable and clinically meaningful when supported by a national infrastructure that enables high-quality data generation, standardized governance, and responsible data sharing.

In Korea, national genomic research was formally initiated in 2000 by the Korea NIH (KNIH) through the launch of the Korean Genome and Epidemiology Study (KoGES). KoGES became the foundational population-based program that allowed systematic cohort establishment, longitudinal phenotyping, and biospecimen collection on a large scale [8]. In parallel, the National Biobank of Korea (NBK) established a standardized infrastructure for biospecimen storage, management, quality assurance, and distribution [9]. Together, these initiatives enabled the generation of population-representative genomic datasets, the development of population-optimized genotyping tools, and the implementation of data-sharing mechanisms designed to promote broad scientific use while maintaining governance and privacy protections. Collectively, this integrated infrastructure constitutes the national foundation built under the leadership of KNIH to support genome-based research and emerging clinical applications. This review synthesizes this infrastructure—cohorts, biobanks, genomic platforms, sequencing-based translation, multi-omics expansion, and data-sharing systems—and highlights the need to strengthen institutional frameworks that allow for a wider and more systematic clinical implementation of genome-based medicine in Korea.

Large-scale population and disease-specific genomic cohorts

Large-scale genomic cohorts are essential for the implementation of precision medicine because they capture genetic diversity alongside environmental and lifestyle variation on the population scale, increasing statistical power for gene–disease association and allowing longitudinal observation of disease trajectories throughout the life course [10]. Cohorts also provide a structural backbone for integrating multi-modal data—including genomics, clinical records, imaging, digital health measures, and environmental exposures—thereby supporting data-intensive analytics and artificial intelligence (AI)–enabled precision healthcare [11]. For these reasons, population-based cohorts have become a central strategy in national precision medicine initiatives worldwide, not only for discovery, but also for risk prediction, validation, and implementation science.

In Korea, this strategic rationale motivated the development of national cohorts beginning in 2000. KoGES was launched as the first population-based genomic cohort and has since matured into one of the most comprehensive longitudinal cohort resources in Asia. KoGES has enrolled more than 200,000 participants in community-based and health examinee cohorts, with a subset of participants undergoing long-term follow-up extending beyond two decades [8]. Genome-wide genotyping and sequencing datasets have been generated for large subsets and made available to the research community, allowing genomic discovery, the development of risk prediction models, and population health research grounded in the genetic architecture of the Korean population. As a national cornerstone, KoGES has supported studies of cardiometabolic traits, chronic disease susceptibility, and complex trait biology, while also providing reference data for calibrating and validating models intended for clinical translation.

Beyond KoGES, KNIH operates and supports a broad portfolio of more than 30 population-based and disease-specific cohorts and registries, and the well-known cohorts and registries are specified in Table 1. These include cohorts of the general population, occupational and vulnerable subgroups, and disease-focused programs that include cardiovascular disease, metabolic disorders, chronic respiratory disease, organ transplantation, and infectious diseases. Collectively, these coordinated infrastructures enable systematic acquisition of biospecimens, longitudinal phenotyping, and generation of population-representative genomic and health datasets. This breadth supports research on the interaction between gene and the environment, improves the generation of pathophysiological insight, and enables the scalable development of predictive models and stratification tools that can be evaluated for clinical utility. Importantly, the national cohort ecosystem provides a platform for harmonization efforts, cross-cohort integration, and future linkage to healthcare delivery and implementation pathways.

Table 1.

Current status of KNIH-operated national cohorts and disease registries

Cohort/Registry name	Start year	No. of participants
KoGES Rural (Ansung)	2001	5,018
KoGES Urban (Ansan)	2001	5,012
KoGES Rural (CAVAS)	2004	28,337
KoGES-Health Examinee (HEXA)a)	2004	173,195
Korea HIV/AIDS Cohort	2006	1,771
Korea HCV Cohort	2007	2,163
Korea Acute Heart Failure Registry (KorAHF)a)	2010	5,625
Korea Acute Myocardial Infarction Registry–National Institutes of Health (KAMIR-NIH)a)	2011	13,707
Korea Urban and Rural Elderly Study (KURE)a)	2011	3,517
Cardiovascular & Metabolic Etiology Research Cohort (CMERC)	2013	11,365
Korea Neonatal Network (KNN)	2013	26,900
Korea Nurses’ Health Study (KNHS)	2013	25,729
The Korean Pregnancy Outcome Study (KPOS)a)	2013	4,195
Korea HBV Cohort	2014	2,657
Korean Transplant Registry (KOTRY)	2014	48,864
Prospective Cohort Study of Menopausal Women’s Health	2014	5,246
Korea COPD Subgroup Study (KOCOSS) Cohort	2016	3,932
Korea Frailty and Aging Cohort Study (KFACS)	2016	3,011
Korean Childhood Asthma Study	2016	958
Korean Stroke Registry (KSR)a)	2017	25,515
Resistant Hypertension Cohort	2018	1,796
Community-based Dementia Cohort	2019	4,789
Tuberculosis Cohorta)	2019	2,284
Early-onset Dementia Cohort	2021	699
Hospital-based Dementia Cohort	2021	829
Parkinson Disease Registry	2021	1,003
Korean Severe Asthma Registry	2022	865

Sample sizes are based on projections as of December 2025 and may change with continued participant enrollment.

KNIH, Korea National Institute of Health; oGES, Korean Genome and Epidemiology Study.

^a)Denotes a completed study. For registries with ongoing data collection, reported numbers are subject to continuous updates. Abbreviations are provided in parentheses.

Building on this foundation, Korea launched the National Integrated Bio Big Data Project in 2024 as a next-generation population-scale cohort initiative that aims to enroll one million participants [12]. Supported jointly by multiple ministries and national agencies—including the Ministry of Health and Welfare, the Ministry of Science and ICT, the Korea Disease Control and Prevention Agency, and the Ministry of Trade, Industry, and Energy—this program reflects a coordinated national commitment to advance precision medicine. By integrating genomic data with longitudinal clinical records, digital health information, and environmental data, the initiative is intended to function as a core national platform for the next-generation discovery, validation, and future clinical implementation of precision health strategies.

National biobank infrastructure enabling genomic and multi-omics research

Sustained generation of genomic and multi-omics data for biomarker discovery and clinical translation depends on reliable access to high-quality, well-annotated human biospecimens derived from population-representative and disease-specific cohorts [13]. To secure and govern such national research resources, KNIH established the NBK as a centralized infrastructure for systematic collection, preservation, distribution, and quality management of biospecimens obtained through government-supported cohort and registry programs [9]. Beyond functioning as a repository, NBK serves as a national platform that operationalizes standardized biospecimen practices, enabling reproducible downstream analyses and long-term sustainability of omics data generation.

Most biospecimens archived within NBK originate from national population cohorts, disease registries, and health examination surveys, including KoGES and the Korea National Health and Nutrition Examination Survey. The central biobank holds biospecimens from approximately 470,000 individuals, covering a broad demographic and disease representation. In parallel, a nationwide biobank network linking major hospitals and regional collection centers supports continued acquisition of clinically derived samples, expanding cumulative biospecimen collections to more than 760,000 individuals. Through integrated resource-sharing systems, researchers can identify available biospecimens and request materials through transparent procedures, supporting equitable access while maintaining governance oversight.

The utilization of NBK resources has expanded steadily, supporting more than 5,000 approved research projects and generating more than 2,000 documented research results, including peer-reviewed publications, biomarker identification studies, and disease risk prediction models. Critically, NBK operates in critical control and quality assurance frameworks, including standardized operating procedures for collection and processing, controlled storage environments, continuous tracking and chain-of-custody systems, and periodic internal and external quality assessments. These measures protect biospecimen integrity and ensure suitability for next-generation omics technologies, which increasingly demand high-quality input for robust molecular inference [14]. Together, NBK and its network provide an essential national foundation for genome-based research, accelerate biomarker development, and strengthen Korea’s capacity to translate omics discoveries into clinically meaningful applications.

Korea Biobank Array

Large-scale genomic research requires cost-effective generation of genome-wide data with dense variant coverage to support association testing, fine-mapping, and imputation. Although SNP microarrays have historically enabled population-scale genotyping, their performance is strongly dependent on ancestry-specific allele frequencies and linkage disequilibrium patterns [15]. Arrays optimized for European populations may, therefore, yield reduced imputation accuracy and limited transferability in East Asian datasets. To address this challenge, KNIH developed the Korea Biobank Array (KoreanChip; K-Chip), a population-optimized genotyping platform designed using Korean WGS reference data and designed to enrich variants that are common, low-frequency and functionally relevant in Koreans [16,17].

The first version of K-Chip incorporated more than 830,000 markers, including a substantial number of rare-frequency and functional variants derived from Korean sequencing resources, improving coverage of coding and clinically actionable variation relative to non-optimized platforms. Large-scale deployment enabled generation of genome-wide data for more than 300,000 participants, creating one of the largest population-scale genotyping resources in East Asia [16]. Studies using K-Chip have demonstrated its scientific and translational utility in identifying genetic determinants of metabolic, cardiovascular, and immune-related traits in Korean and broader East Asian populations, supporting both discovery and improved ancestry-relevant imputation.

The updated Korea Biobank Array (K-Chip 2) has been distributed while retaining legacy content for continuity (https://www.koreanchip.org/project). The revised array incorporates expanded modules that target clinically actionable genes, pharmacogenomics, and immune-related disease loci, including curated components related to drug metabolism and response, specifically genes involved in drug absorption, distribution, metabolism, and excretion (Table 2). These enhancements allow the investigation of interindividual variability in drug response, susceptibility to immune-mediated diseases, and improved downstream imputation performance, thus strengthening the link between population genomics and clinical translation. The pharmacogenomic variants included in K-Chip can also support research on the variability in drug response among pediatric patients with kidney disease, including children undergoing kidney transplantation or those treated for nephrotic syndrome, where the prediction of drug efficacy, toxicity, and optimal dosing of immunosuppressive or adjunctive therapies is clinically important. In addition, K-Chip 2 includes clinical and disease-associated variants curated from ClinVar and other clinical and disease-related public databases, enabling its application in research on monogenic and genetically influenced pediatric kidney diseases. However, as a genotyping array, K-Chip captures selected known variants rather than providing comprehensive sequencing coverage; therefore, its primary value lies in supporting population-based and translational research, variant prioritization, and hypothesis generation in pediatric kidney disease rather than serving as a standalone diagnostic platform.

Table 2.

Composition and functional categories of variants included in the KBA

Category	Sub-category	KBA v1.1	KBA v2.0
Clinical and disease related	Pathogenic variants (ClinVar database)	1,106	120,579
	Disease-associated variants (clinical and disease-related public database)	66,160	280,181
	ACMG (73 genes)a)	3,630	43,875
	Pharmacogenetic variants (ADME-related)	2,845	18,765
Functional, protein coding	Protein function–altering variants (nonsyn., indels)	170,639	517,515
Immune related	HLA and KIR region-specific variants	11,338	69,996
Genome-wide tagging	Tagging variants optimized for Korean genomes (imputation-aware)b)	584,364	954,248
Non-autosomes	X, Y, and MT chromosomes	2,815	35,428
Total		827,926	1,663,847

KBA v2.0 includes all variants of KBA v1.1

KBA, Korea Biobank Array; ACMG, American College of Medical Genetics and Genomics; ADME, absorption, distribution, metabolism, and excretion; HLA, human leukocyte antigen; KIR, killer-cell immunoglobulin-like receptors; MT, mitochondrial.

^a)ACMG has actionable 73 genes for incidental findings.

^b)Imputation: statistical estimation of untyped genotypes.

Importantly, KNIH has pursued nationwide adoption through structured technology transfer and capacity-building programs. Standardized laboratory protocols, analytical pipelines, and training workshops have been distributed to academic and clinical research institutions, allowing independent implementation of K-Chip–based studies and fostering a coordinated national network for genome-based research. In this way, K-Chip functions not only as a genotyping tool but also as a mechanism for standardization and interoperability in the genomic research ecosystem in Korea.

WGS for clinical translation

In parallel with array-based genotyping, KNIH has expanded WGS capacity through national cohort resources and the National Integrated Bio Big Data Pilot Project, a national initiative targeting rare diseases and designed to generate comprehensive genomic and clinical datasets that support precision medicine. WGS provides more complete variant detection than array-based approaches, including rare, structural, and non-coding variants, and serves as a critical platform for clinical translation, particularly in rare diseases where etiologic heterogeneity is substantial [5,18]. Through national sequencing initiatives, WGS data have been generated and are increasingly positioned as foundational resources for both research and clinical interpretation. In this context, WGS has demonstrated improved diagnostic performance compared to conventional panel-based analyses; within the National Bio Big Data Pilot Project, the overall diagnostic yield reached approximately 35%, with a diagnostic rate of about 34% observed in kidney and urinary tract disorders (unpublished data). These findings support the clinical utility of comprehensive genomic sequencing in genetically heterogeneous disease groups and are consistent with international evidence showing that systematic implementation of WGS improves diagnostic rates and can inform clinical management. Furthermore, such national resources contribute to improved variant interpretation in Korean populations by strengthening reference allele frequency datasets and enabling iterative refinement of analytic pipelines.

Beyond rare disease diagnosis, KNIH has advanced urgent-care genomic applications through rapid WGS pilot studies for critically ill neonates. These studies established streamlined workflows capable of providing sequencing and interpretation in days, demonstrating feasibility in neonatal intensive care settings. Significant diagnostic results have been observed in newborns with severe unexplained conditions, and genomic findings have directly guided therapeutic decisions in a significant subset of diagnosed cases. This experience is important because rapid genomic diagnosis can reduce the time to targeted care, limit unnecessary investigations, and improve clinical decision-making in time-sensitive environments. Building on these early achievements, KNIH plans to expand participating hospitals and clinical networks to accumulate broader evidence, validate clinical utility at scale, and establish standardized pathways for nationwide implementation of rapid genomic diagnostics.

Genomic medicine also extends beyond diagnosis to preventive strategies such as genome-based newborn screening [19-21]. Internationally, multiple countries have launched pilot programs that evaluate genomic technologies for newborn screening, including collaborative efforts through global consortiums [22]. In Korea, a national newborn genomic screening pilot has been initiated with a focus on identifying early-onset actionable conditions, developing informed consent and clinical decision-support frameworks, establishing secure long-term data governance and integrating genomic results with health records and longitudinal follow-up systems. Beginning this year, the program will transition from the development of preparatory infrastructure to implementation in the real world by enrolling and sequencing actual newborns, generating empirical evidence on feasibility, clinical utility, and ethical governance in routine healthcare settings. Collectively, these efforts position KNIH at the forefront of translating WGS into practice and advancing precision medicine throughout the life course.

Multi-omics beyond genomics

Although genomic data provide a foundational layer for genome-based medicine, effective precision health increasingly depends on the integration of additional biological and phenotypic dimensions. KNIH is therefore expanding national infrastructure beyond genomics to generate multi-omics and real-world health datasets linked to cohorts and biobanked specimens. Ongoing cohort programs incorporate brain magnetic resonance imaging, life log and wearable-device data, and longitudinal clinical information, enabling deep phenotyping of health and disease trajectories [23]. In parallel, microbiome initiatives are being developed for selected participants to support investigation of host–microbiome interactions in disease susceptibility and therapeutic response [24]. Furthermore, preserved biospecimens stored in NBK are being utilized to generate proteomic, metabolomic, and epigenomic datasets, creating a comprehensive molecular resource suitable for translational research and biomarker development. International experiences from large cohorts that integrate imaging, multi-omics and digital health data underscore the added value of multi-modal integration for risk prediction, disease stratification, and discovery of clinically deployable biomarkers. By establishing a national multi-omics ecosystem connected to longitudinal cohorts and biobanked specimens, KNIH is strengthening Korea’s capacity to accelerate biomarker discovery, refine disease classification, and support precision medicine clinical translation.

Clinical omics and data archives as central repositories for data sharing

Data sharing is a foundational requirement for genome-based medicine. Large-scale discovery often depends on the aggregation of datasets across studies to achieve sufficient power, particularly for rare variants and complex traits with modest effect sizes. Shared data enable reproducibility, independent validation, cross-cohort meta-analyses, and the development of robust predictive models. In addition, integration of genomic data with clinical, imaging, and environmental information is central to AI-driven precision healthcare, further increasing the need for well-curated, interoperable data resources [25,26]. Consequently, many countries have developed centralized repositories as national infrastructures for biomedical data sharing, supported by governance frameworks that balance scientific value with privacy protection [6,7,27].

International precedents illustrate that effective data-sharing ecosystems require not only technical platforms but also mature policies and governance. The UK Biobank’s analysis environment links genomic and longitudinal data under clear access policies and standardized curation pipelines. The NIH All of Us Researcher Workbench similarly provides controlled-access, cloud-based analytics governed by structured data management and security protocols. Major funding agencies have increasingly required data management plans and the deposition of research outputs in designated repositories as conditions of funding. These approaches highlight a consistent principle: voluntary sharing alone rarely produces comprehensive coverage, and sustainable data ecosystems require enforceable expectations, standardized quality frameworks, and incentive structures [28].

In Korea, KNIH operates the Clinical & Omics Data Archive (CODA) as a national central repository designed to enable systematic sharing and reuse of clinical and multi-omics datasets (https://coda.nih.go.kr/). CODA aggregates genomics, transcriptomics, proteomics, metabolomics, and associated clinical data generated through government-funded programs, national cohorts, and disease registries, while also accepting submissions from academic and clinical investigators. Following submission, CODA conducts data curation, quality control, metadata annotation, and harmonization to improve interoperability and research readiness. Investigators can search datasets and request controlled access through established procedures, and CODA is being expanded toward secure cloud-based in situ analysis environments aligned with international best practices.

Despite these advances, important challenges remain. Although regulations and research management practices have evolved, deposition in national repositories is not yet universally mandatory, and responsibilities for data sharing across publicly funded programs remain inconsistently defined. As a result, some publicly funded outputs are not systematically deposited, and sharing practices vary across institutions and projects, limiting discoverability and constraining integration. Another challenge concerns data quality and standardization: high-value clinical-genomic datasets require consistent metadata structures, harmonized phenotype definitions, and rigorous quality assurance to ensure downstream usability. Best-practice frameworks, such as the FAIR (Findable, Accessible, Interoperable and Reusable) principles, emphasize early-stage planning, standardized data models, and continuous quality monitoring as prerequisites for effective sharing [28].

Looking ahead, consolidating Korea’s national data-sharing ecosystem will require several strategic actions. Data deposition into certified repositories should become a standard expectation for publicly funded research with clear timelines and compliance monitoring. Harmonized national data standards—including common data models, phenotype ontologies, and metadata frameworks—should be adopted across cohorts, registries, and hospital-based studies to allow interoperability. Sustained investment in data stewardship expertise is essential, including dedicated support units spanning curation, bioinformatics, clinical informatics, and regulatory compliance that work with investigators from the inception of the project to ensure alignment with downstream sharing requirements. Finally, transparent governance, secure access controls, and robust privacy protections must remain central to maintaining public trust and long-term participation. In summary, while CODA and related platforms provide a strong technological foundation, the next phase must focus on mandates, standards, stewardship, and governance capacity to fully leverage national resources for reproducible research, AI-enabled healthcare innovation, and accelerated clinical translation.

Conclusions

During the past two decades, KNIH has played a central role in the establishment of genome-based medicine as a national public health infrastructure in Korea. Through the development of population- and disease-specific cohorts, the NBK and its network, the Korea Biobank Array, the expansion of WGS for clinical translation, multi-omics integration, and the establishment of CODA as a national data archive, KNIH has built an integrated platform supporting genomic discovery, multi-omics research, and responsible data sharing. These coordinated efforts have strengthened national research capacity, accelerated genomic R&D, and broadened access to high-quality biospecimens and data.

In pediatric nephrology, the role of genetic diagnosis is expected to continue to expand as the genetic basis of kidney disease becomes increasingly recognized. National genomic resources established through coordinated initiatives are anticipated to support the broader clinical implementation of genomic diagnostics, particularly in children with early-onset disease, familial clustering, or syndromic features suggestive of monogenic etiologies. Integration of genomic data with longitudinal clinical information can facilitate more accurate molecular diagnosis and improved disease characterization, and it is hoped that these national infrastructures will further promote the translation of genomic knowledge into clinical practice. Ultimately, such resources are expected to contribute to the wider adoption of genetic diagnosis in pediatric kidney disease, supporting genetic counseling, individualized management strategies, and improved clinical care.

Building on this foundation, Korea is now poised to advance precision medicine as a key component of the future healthcare strategy. Integration of genomics and multi-omics with longitudinal clinical data can enable more accurate risk assessment, earlier diagnosis, and individualized therapeutic decision-making. Realizing this potential at scale will require continued leadership in strengthening data governance, quality management, and equitable access, along with collaboration between healthcare systems, academia, industry, and international partners. By consolidating standards and stewardship while maintaining public trust, Korea can further develop a globally connected precision medicine ecosystem that delivers meaningful and equitable health benefits to the population.

Funding Statement

None.

Data availability statement

Data sharing is not applicable as no new data were created or analyzed in this study.