Correcting deletion errors in DNA data storage with enzymatic synthesis

Abstract

DNA is considered a promising alternative to traditional storage media because of advantages such as high data density, longevity, and ease of generating copies. One of the major drawbacks of DNA data storage however is that DNA synthesis is costly and resource intensive. A newly proposed enzymatic method has the potential to decrease the cost of synthesis but has the disadvantage that the number of times a base is repeated cannot be precisely controlled. The method is also prone to deletion of runs. Existing encoding approaches for this synthesis method either have a low rate, specifically, $\leq\log_{2} 3$ per run, or cannot protect against deletion errors. The current paper proposes a new error-correcting code and a synchronization algorithm that can combat deletions and achieve a code rate higher than $\log_{2} 3$ bits per unit time.