A Characterization of the DNA Data Storage Channel

Abstract

Owing to its longevity and enormous information density, DNA, the molecule encoding biological information, has emerged as a promising archival storage medium. However, due to technological constraints, data can only be written onto many short DNA molecules that are stored in an unordered way, and can only be read by sampling from this DNA pool. Moreover, imperfections in writing (synthesis), reading (sequencing), storage, and handling of the DNA, in particular amplification via PCR, lead to a loss of DNA molecules and induce errors within the molecules. In order to design DNA storage systems, a qualitative and quantitative understanding of the errors and the loss of molecules is crucial. In this paper, we characterize those error probabilities by analyzing data from our own experiments as well as from experiments of two different groups. We find that errors within molecules are mainly due to synthesis and sequencing, while imperfections in handling and storage lead to a significant loss of sequences. The aim of our study is to help guide the design of future DNA data storage systems by providing a quantitative and qualitative understanding of the DNA data storage channel.

Improved read/write cost tradeoff in DNA-based data storage using LDPC codes

“…On average, most of the experiments had total error rate around 1.3% (substitution: 0.4%, deletion: 0.85%, insertion: 0.05%). Based on the typical error rates for Illumina sequencing and experiments on paired-end data as in [9], the substitution errors are primarily caused by the sequencing and the insertion and deletion errors are primarily from the synthesis. Figure 10 shows the histogram of the coverage of the oligonucleotides for experiments #2 and #5 when the mean coverage was set to 5 by subsampling.…”

“…Since the RS code is applied on very short sequences, the scheme suffers from limitations of short block length codes [13]. The scheme also ignores reads with insertions and deletions, which usually comprise around 30-50% of the reads [9].…”

“…We assume that the binary data is a uniformly random stream, which can be obtained by lossless compression and encryption of the actual data. The currently standard synthesis process generates millions of copies of each oligonucleotide [9], possibly with errors -substitutions, insertions and deletions. The first step of the reading process involves amplification of the synthesized DNA sequences using PCR [3].…”

Overcoming High Nanopore Basecaller Error Rates for DNA Storage Via Basecaller-Decoder Integration and Convolutional Codes

“…1 shows a typical DNA storage system. Binary data is encoded into short DNA sequences (oligonucleotides, or oligos for short) with lengths limited to around 150 bases/nucleotides for current scalable synthesis technologies [6]. Note that each DNA nucleotide belongs to the set {A, C, G, T}.…”

Multidimensional Data Organization and Random Access in Large-Scale DNA Storage Systems

“…We do not discuss the details on experimental implementations of DNA storage systems, which involve other important aspects such as data encoding, error correction, and system integration and automation. [11][12][13][14] In design A (Figure 1), the pool of data oligos is virtually organized in the form blocks, tables, rows, and columns. Each data-encoding strand is composed of several domains, including a data payload block surrounded by three address blocks (i.e., PCR primer targets) on both sides.…”