Adaptive Sampling for Storage of Progressive Images on DNA
New technique uses Nanopore sequencing to selectively read only the DNA needed for lower-resolution image previews.
A team of researchers has published a paper introducing an innovative method for storing images in DNA that significantly reduces the cost of retrieving them. The system, developed by Xavier Pic, Nimesh Pinnamaneni, and Raja Appuswamy, tackles a major bottleneck in DNA data storage: the impracticality and high cost of retrieving a single file from a mixed pool of DNA molecules (oligos). Their solution leverages the progressive decoding functionality of codecs like JPEG2000, encoding different image resolution layers into separate DNA oligos. This structure enables a breakthrough in access.
The technical core combines the JPEG DNA VM codec for reliable encoding with the adaptive sampling capability of Nanopore sequencers. When a user needs only a thumbnail or lower-resolution preview, the sequencer can be instructed to read only the specific oligos containing that data layer, skipping the rest. This selective, PCR-free random access means not all oligos in a pool need to be sequenced and decoded to retrieve a target file, slashing sequencing time and cost. The work represents a critical step toward making DNA storage—a medium prized for its extreme density and longevity—technologically and economically viable for practical archival applications.
- Integrates progressive image codecs (JPEG2000) with DNA encoding (JPEG DNA VM) to store images in resolvable layers.
- Uses Nanopore sequencers' adaptive sampling to enable selective, PCR-free reading of only the DNA needed for a desired resolution.
- Aims to solve the random access problem in mixed oligo pools, drastically cutting read costs for partial file retrieval.
Why It Matters
Makes ultra-long-term DNA data storage more practical by solving the costly 'needle-in-a-haystack' retrieval problem for archived files.