Progressive decoding of DNA-stored JPEG data with on-the-fly error correction

Abstract

DNA storage is a developing field that uses DNA to archive digital data owing to its superior information density and stability. Although DNA storage has been performed on a significant scale, challenges arise from the synthesis and sequencing of data-encoded oligonucleotides. Synthesis of DNA introduces significant noise into the process. Consequently, high-read-quality sequencers are often required, making the process expensive and lack scalability. Error correction codes are used within the DNA storage pipeline to provide resilience to noise at a cost of additional redundancy and decoding complexity. Given such constraints and challenges, the main objective we seek to deliver is a time- and storage-efficient image coding strategy. We introduce a novel DNA-based progressive JPEG decoder with on-the-fly error-correcting & rendering capability. This system can progressively decode an image while also correcting for errors as they occur. It modifies standard JPEG encoding methods to store data in localized chunks and uses adapted Raptor error-correction codes to improve the speed and quality of partial decoding. We optimize and evaluated the method under varying levels of simulated errors, as well as show how different parts of the pipeline improve real-time decoding capability. We also test the pipeline in a real-world wet-lab experiment. We present the first progressive image decoding schema aimed at realizing real-time rendering of DNA-stored images.