Mapping the South Korean Genome for Global Precision Medicine

The project will generate the first large-scale, telomere-to-telomere reference genome for the South Korean population, capturing its unique genomic diversity. Currently, scientists’ understanding of the human genome is largely based on the reference genome known as GRCh38, which is built from a composite of fragments of DNA from multiple individuals. While many of these contributors declared themselves to be of Northern European descent, many did not disclose their ancestry at all. As a result, it is difficult to pinpoint which genomic regions of the reference carry ancestry-specific variants, limiting GRCh38’s reliability as a true baseline.

No single reference genome can represent or capture the full spectrum of human diversity. Pangenomes, like the one we’re creating with South Korea, offer an alternative way of studying human biology that is more population-specific. They are built from the genomes of many individuals in a population to represent both the core genes shared across humanity and the variable genes found locally.

By uncovering population-specific variants, the project will deepen understanding of diseases more common in South Korea and support more precise diagnostics and therapies. Beyond its national significance, the project could serve as a model for other countries to build their own pangenomes, especially those underrepresented in current datasets.

PacBio’s technology was instrumental in sequencing the first-ever telomere-to-telomere human genome in 2022. Prior to this, the human genome was only 92% complete because the missing 8% included some of the most complex and repetitive regions of human DNA, which other sequencing technologies either misassembled or missed.

This breakthrough was made possible due to PacBio’s HiFi long-read sequencing method, which sequences DNA in long fragments that span thousands of base pairs, compared to other technologies, which only sequence a few hundred at a time. This means HiFi is significantly more effective at sequencing longer, more complex variants and regions of the genome, such as repeat expansions. Such enhanced accuracy and completeness are essential when building whole-genome datasets so that government-funded organizations, such as the Korea Disease Control and Prevention Agency and the National Institute of Health, can be confident they are obtaining the most actionable genomic data possible.

Since completing the first genome in 2022, HiFi technology has advanced even further, offering both accuracy and improved cost-efficiency, making it well-suited for national-scale genomics projects.

Understanding human biology and disease requires more than just looking at DNA alone – it calls for a multi-dimensional view of how genetic information is organized, expressed and regulated. While whole-genome analysis provides the blueprint of our genetic code, RNA analysis reveals how that code is actually read and executed within the body. By capturing both DNA and RNA information together, researchers can uncover how changes at the genomic level influence gene expression, protein production and disease development. 

Using PacBio’s full technology suite makes this multidimensional view possible. HiFi whole-genome sequencing delivers a highly accurate picture of DNA – the whole genome – including regions that are challenging to sequence. Meanwhile, Kinnex RNA analysis enables precise transcriptome profiling, revealing gene expression and alternative splicing patterns that help link DNA changes to their effects on the body. CiFi chromosome-scale mapping adds another layer of context, allowing researchers to detect large structural variants and complex genomic features that are difficult to resolve with HiFi sequencing alone.

Together, these technologies provide a more complete picture of human biology – one that neither DNA nor RNA sequencing alone could achieve. This integrated approach accelerates the discovery and interpretation of population-specific variants, offering new insights into how genetic diversity drives health and disease.

These cost reductions are being achieved by enabling multiple runs per SMRT Cell, the chip used in PacBio’s sequencing, while maintaining output per run. This approach improves efficiency, reduces waste and preserves accuracy and the richness of the data.

The latest updates to the Revio and Vega platforms reflect our continued commitment to making highly accurate long-read sequencing solutions accessible at scale for population genomics, clinical research and production-scale environments. By reducing costs without compromising quality, these advances enable larger, more diverse genomic studies that were previously cost-prohibitive.

HiFi sequencing can create comprehensive benchmarking datasets, such as the Platinum Pedigree, which are essential for training and validating AI models for genome interpretation. By providing benchmarks that accurately characterize variants, even in complex and challenging regions, PacBio helps improve the reliability of AI models in identifying and classifying genetic variants.

For example, our Platinum Pedigree benchmark was used to retrain Google’s DeepVariant, resulting in a 34% reduction in erroneously called variants. These rich, high-quality datasets lay the foundation for advancing precision medicine by allowing AI tools to better identify genetic variants that can guide diagnosis and inform personalized treatment plans.

At ASHG, our key message was the need for researchers to go beyond the genome and embrace a multiomic approach. While industry understanding of the genome has grown significantly in recent years, there is still a world of insight to uncover across the transcriptome, methylome and chromatin epigenome. PacBio’s sequencing technology allows researchers to capture data from multiple ‘omes in a single sequencing run, providing richer data while reducing costs and the need for patients to undergo multiple tests.