Analysis of the influence of errors in DNA-based image coding

In the last decade, DNA has been increasingly investigated as an alternative medium for cold data storage, presenting several advantages over standard hard drives such as a higher density, longer lifespan and lower energy consumption. However, such coding methods are limited by biochemical constraints that elevate the probability of errors being added to the coded nucleotides during synthesis, storage, and sequencing. Although such errors can be limited by carefully designing the produced strands, it is unfeasible to avoid them completely. In this paper, we explore the impact of naturally induced errors on the performance of a DNA-based image coding by means of realistic simulations, demonstrating that the quality of the decoded images is severely impacted. We also propose an error correction scheme based on Reed-Solomon codes and Blawat encoding, which successfully removes the produced artifacts.