HKU Researchers Develop Blockchain-Inspired Codec to Extend DNA Data Storage Lifespan to Centuries

Researchers at the Faculty of Engineering at The University of Hong Kong have developed an open-source DNA storage codec that uses blockchain-inspired computing techniques to recover digital information from heavily degraded molecular strands, according to a report by TechXplore (https://techxplore.com/news/2026-09-blockchain-dna-storage-codec-recovers.html).

The framework, named Gungnir, was created by a research team led by Ruibang Luo of HKU's School of Computing and Data Science and Can Li of the Department of Electrical and Computer Engineering. The peer-reviewed study, lead-authored by Jingcheng Zhang, was published in Nature Communications.

DNA storage offers high information density and zero-energy physical preservation over time, but molecular sequences degrade and accumulate damage that impairs data recovery. Existing DNA codecs operate reliably on newly synthesized sequences but lose efficacy as storage errors accumulate over extended periods.

Gungnir alters that tradeoff by applying compute-heavy verification methods derived from blockchain architectures. The codec applies computational power to generate candidate reconstructions of the underlying sequence and verify them until the correct file is validated. Because error correction scales with available processing capacity, the system can tolerate error rates of up to 20 percent, representing a fourfold improvement over previous decoding techniques.

In extreme stress tests, the researchers demonstrated lossless data retrieval from DNA sequences subjected to damage levels equivalent to roughly 400 years of archival degradation. The team noted that this capability could expand the operational lifespan of DNA-based storage media from less than a decade to centuries, with further hardware scaling potentially extending recovery windows to millennia.

The Gungnir codec has been released as an open-source tool on GitHub to support archival storage research.