DNA Encryption of Compartmentalized DNA Files

Project Information

DNACryp

Grant agreement ID: 101188053

DOI

10.3030/101188053

Project closed

EC signature date 17 July 2024

Start date 1 January 2025

End date 30 June 2026

Funded under

• European Research Council (ERC)

Total cost

No data

EU contribution

€ 150 000,00

Coordinated by

TECHNISCHE UNIVERSITEIT EINDHOVEN
Netherlands

Periodic Reporting for period 1 - DNACryp (DNA Encryption of Compartmentalized DNA Files)

Reporting period: 2025-01-01 to 2026-06-30

Summary of the context and overall objectives of the project

The amount of digital data the world produces is growing far faster than our ability to store it. Conventional media such as hard drives and magnetic tape are reaching hard limits in data density, lifespan and energy use, and they depend on scarce and sometimes toxic materials. Long-term forecasts suggest that within a couple of decades the global demand for storage could outstrip what these technologies can supply by a wide margin. A different kind of medium is needed for data that must be kept safely for a very long time.
DNA, the molecule that carries the genetic information of living things, is a remarkable candidate. It can hold enormous amounts of information in a tiny volume, stays stable for hundreds to thousands of years under the right conditions, and needs almost no energy to maintain. Over the past decade researchers have shown that digital files can be written into synthetic DNA and read back again, and much effort has gone into making these steps efficient and reliable.
One important piece has been missing. A complete storage system must not only store and retrieve information but also keep it safe from anyone who is not authorised to see it. This matters most for sensitive or confidential archives. Digital encryption can scramble a file before it is written into DNA, but once the DNA has been copied and read, the scrambled data can still be attacked with powerful computers, and future advances in computing could break protections that look secure today. Before this project there was no way to protect DNA-stored data at the physical level of the molecules themselves.
The overall objective was to develop and prove the first molecular-level protection method for data stored in DNA. Instead of scrambling the information mathematically, the approach controls whether the DNA can be physically copied at all. Reading DNA-stored data normally begins with a standard laboratory copying step called the polymerase chain reaction. In our method, small purpose-designed molecules called lockers attach to the strands that must be protected and block this copying step, so those strands cannot be read and the file cannot be reconstructed. A matching molecular password then removes the lockers and lets normal reading resume. Access is reversible and controlled by holding the correct molecular key, and because an attacker would have to find that key among an astronomically large number of possibilities, the protection strongly resists automated computer attacks.
The expected impact is to give DNA data storage a practical, reversible access-control layer that works with the reading and writing methods the field already uses. This helps make secure long-term DNA archiving viable, strengthens Europe's position in an emerging strategic technology, and opens routes to commercial exploitation and to wider molecular information security applications.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The project set out to turn the idea of molecular-level protection for DNA-stored data into a working, tested method.
The first activity was to design the protective molecules. The team created short pieces of DNA, called lockers, that bind tightly to the specific strands chosen for protection and carry a chemical stop at one end so the copying machinery cannot read through them. Because a locker attaches to its target more strongly than the small starter molecule the copying process normally relies on, it wins the competition and shuts down copying of that strand. To reverse this, molecular passwords were built by attaching the matching partner strand to tiny magnetic beads. Adding these beads captures the lockers, which are then pulled out of the sample with a magnet so that normal reading can resume.
This locking and unlocking behaviour was first demonstrated in a deliberately simple system containing only two DNA strands, one protected and one not. Measurements confirmed that the locker strongly and selectively blocked copying of the protected strand, and that the molecular password removed the block and restored balanced copying.
The method was then applied to real digital files. A dedicated encoding scheme was developed that reserves a small region in a chosen subset of strands to mark them as lockable, and that uses layered error correction so a file can only be rebuilt if enough of the protected strands are recovered. A short text file was encoded onto DNA in this way. When locked it could not be decoded, and after unlocking it decoded reliably every time. Sequencing, the step that reads the DNA back into digital form, together with realistic simulations of normal reading conditions, confirmed both outcomes.
Two further challenges were addressed. The first was scalability. The method was applied to a file of close to one megabyte, made up of roughly 65,000 DNA strands, and controlled access just as effectively as for the small file, with no loss of performance. The second was independent control of many files at once. An automated design algorithm was created to generate sets of locker and password pairs that act independently of one another, so individual files within a mixed collection can be locked or unlocked separately without affecting the rest. The password design was also improved so files can be locked and unlocked repeatedly over several cycles while keeping performance high, and the approach was shown to add only a negligible amount of extra DNA and cost at archive scale.
The outcome of the action is a complete and experimentally validated method for reversible, molecular-level access control of DNA-stored data. It works with the standard writing, copying and reading steps already used in the field, scales from small files up to megabyte files, allows independent protection of many files held together, and can be applied repeatedly. This provides a practical foundation for keeping long-term DNA archives secure

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The project delivered the first demonstrated method for reversible protection of DNA-stored data at the level of the molecules themselves. Where earlier work in the field had concentrated on writing and reading information efficiently, no physical way existed to control who could actually retrieve it. This project closed that gap.
The main results are a set of clear, repeatable demonstrations. Protected DNA strands could be locked strongly and selectively, and then unlocked on demand with the matching molecular password. The method worked not only on simple test molecules but on genuine encoded files, and it scaled without loss of performance from a few hundred bytes up to a file of close to one megabyte containing about 65,000 strands. Several files stored together could be locked and unlocked independently of one another, and the same file could be locked and unlocked repeatedly. The protection adds only a negligible amount of extra DNA and cost, and this overhead becomes smaller as archives grow larger.
The potential impact is significant for the long-term, secure storage of sensitive information such as medical, financial, governmental and cultural records. Because the method blocks physical retrieval rather than only scrambling the information mathematically, it protects data even if a physical sample falls into the wrong hands, and it works alongside conventional digital encryption to provide layered, defence-in-depth security. As it is compatible with the reading and writing steps the field already uses, it can be adopted without rebuilding existing DNA storage pipelines. The same design also extends naturally to schemes that require several molecular keys held by different parties before access is granted, which is useful wherever shared authorisation is needed.
Realising this potential will require further steps. Additional research and demonstration are needed to test the method across a wider range of storage conditions and larger, more complex libraries, to characterise unlocking speed and error rates, and to integrate it with the automated laboratory workflows used at scale. The method is protected by a granted patent, with international applications pending, giving a firm basis for licensing or a spin-out. Further engagement with companies and end users, together with investment and access to markets, will be important to move from laboratory proof of concept towards a product. As DNA data storage matures more broadly, common standards, interoperability and continued international collaboration would help a molecular security layer of this kind be adopted consistently.
In summary, the project produced a complete and validated technology for reversible, molecular-level access control of DNA-stored data. It is selective, scalable to megabyte files, capable of independent control of many files, repeatable and low in cost, and it integrates directly with established DNA storage methods. Together these results establish the first practical foundation for keeping long-term DNA archives secure and open a route towards commercial molecular information security.