Error profile of the DNA storage channel

For DNA to become a realistic and safe option for digital information storage we will need to address a number of challenges. One such challenge is the need for appropriately tailored error correction strategies. To do so we require an understanding of the noise, or error rates and types, within the DNA storage process. At present information to be stored in DNA progresses through a physical noisy channel beginning with DNA synthesis (writing), followed by a form of amplification (copying) ending with DNA sequencing (reading). This channel, termed the DNA storage channel, has a unique error profile.This study contains Illumina sequencing data from a proof-of-concept of a workflow aiming to define the error profile of the DNA storage channel. We created 29,040 synthetic DNA sequences consisting of random and custom sequences designed to probe what DNA sequences can and cannot be synthesised, amplified and sequenced. These synthetic DNA sequences were synthesised as large-oligo pools using phosphoramidite chemistry performed under different conditions of fidelity termed and then subjected to differing PCR amplification regimes prior to Illumina MiSeq sequencing.