Deep DNA Storage: Scalable and Robust DNA Storage via Coding Theory and\n Deep Learning

Abstract

DNA-based storage is an emerging technology that enables digital information\nto be archived in DNA molecules. This method enjoys major advantages over\nmagnetic and optical storage solutions such as exceptional information density,\nenhanced data durability, and negligible power consumption to maintain data\nintegrity. To access the data, an information retrieval process is employed,\nwhere some of the main bottlenecks are the scalability and accuracy, which have\na natural tradeoff between the two. Here we show a modular and holistic\napproach that combines Deep Neural Networks (DNN) trained on simulated data,\nTensor-Product (TP) based Error-Correcting Codes (ECC), and a safety margin\nmechanism into a single coherent pipeline. We demonstrated our solution on\n3.1MB of information using two different sequencing technologies. Our work\nimproves upon the current leading solutions by up to x3200 increase in speed,\n40% improvement in accuracy, and offers a code rate of 1.6 bits per base in a\nhigh noise regime. In a broader sense, our work shows a viable path to\ncommercial DNA storage solutions hindered by current information retrieval\nprocesses.\n