DNA Storage Systems Research Articles

Levy Sooty Tern Optimization Algorithm Builds DNA Storage Coding Sets for Random Access.

DNA molecules, as a storage medium, possess unique advantages. Not only does DNA storage exhibit significantly higher storage density compared to electromagnetic storage media, but it also features low energy consumption and extremely long storage times. However, the integration of DNA storage into daily life remains distant due to challenges such as low storage density, high latency, and inevitable errors during the storage process. Therefore, this paper proposes constructing a DNA storage coding set based on the Levy Sooty Tern Optimization Algorithm (LSTOA) to achieve an efficient random-access DNA storage system. Firstly, addressing the slow iteration speed and susceptibility to local optima of the Sooty Tern Optimization Algorithm (STOA), this paper introduces Levy flight operations and propose the LSTOA. Secondly, utilizing the LSTOA, this paper constructs a DNA storage encoding set to facilitate random access while meeting combinatorial constraints. To demonstrate the coding performance of the LSTOA, this paper consists of analyses on 13 benchmark test functions, showcasing its superior performance. Furthermore, under the same combinatorial constraints, the LSTOA constructs larger DNA storage coding sets, effectively reducing the read-write latency and error rate of DNA storage.

Data recovery methods for DNA storage based on fountain codes

Today's digital data storage systems typically offer advanced data recovery solutions to address the problem of catastrophic data loss, such as software-based disk sector analysis or physical-level data retrieval methods for conventional hard disk drives. However, DNA-based data storage currently relies solely on the inherent error correction properties of the methods used to encode digital data into strands of DNA. Any error that cannot be corrected utilizing the redundancy added by DNA encoding methods results in permanent data loss. To provide data recovery for DNA storage systems, we present a method to automatically reconstruct corrupted or missing data stored in DNA using fountain codes. Our method exploits the relationships between packets encoded with fountain codes to identify and rectify corrupted or lost data. Furthermore, we present file type-specific and content-based data recovery methods for three file types, illustrating how a fusion of fountain encoding-specific redundancy and knowledge about the data can effectively recover information in a corrupted DNA storage system, both in an automatic and in a guided manual manner. To demonstrate our approach, we introduce DR4DNA, a software toolkit that contains all methods presented. We evaluate DR4DNA using both in-silico and in-vitro experiments.

GradHC: highly reliable gradual hash-based clustering for DNA storage systems.

As data storage challenges grow and existing technologies approach their limits, synthetic DNA emerges as a promising storage solution due to its remarkable density and durability advantages. While cost remains a concern, emerging sequencing and synthetic technologies aim to mitigate it, yet introduce challenges such as errors in the storage and retrieval process. One crucial task in a DNA storage system is clustering numerous DNA reads into groups that represent the original input strands. In this paper, we review different methods for evaluating clustering algorithms and introduce a novel clustering algorithm for DNA storage systems, named Gradual Hash-based clustering (GradHC). The primary strength of GradHC lies in its capability to cluster with excellent accuracy various types of designs, including varying strand lengths, cluster sizes (including extremely small clusters), and different error ranges. Benchmark analysis demonstrates that GradHC is significantly more stable and robust than other clustering algorithms previously proposed for DNA storage, while also producing highly reliable clustering results. https://github.com/bensdvir/GradHC.

DNA-QLC: an efficient and reliable image encoding scheme for DNA storage

BackgroundDNA storage has the advantages of large capacity, long-term stability, and low power consumption relative to other storage mediums, making it a promising new storage medium for multimedia information such as images. However, DNA storage has a low coding density and weak error correction ability.ResultsTo achieve more efficient DNA storage image reconstruction, we propose DNA-QLC (QRes-VAE and Levenshtein code (LC)), which uses the quantized ResNet VAE (QRes-VAE) model and LC for image compression and DNA sequence error correction, thus improving both the coding density and error correction ability. Experimental results show that the DNA-QLC encoding method can not only obtain DNA sequences that meet the combinatorial constraints, but also have a net information density that is 2.4 times higher than DNA Fountain. Furthermore, at a higher error rate (2%), DNA-QLC achieved image reconstruction with an SSIM value of 0.917.ConclusionsThe results indicate that the DNA-QLC encoding scheme guarantees the efficiency and reliability of the DNA storage system and improves the application potential of DNA storage for multimedia information such as images.

Robust multi-read reconstruction from noisy clusters using deep neural network for DNA storage

DNA holds immense potential as an emerging data storage medium. However, the recovery of information in DNA storage systems faces challenges posed by various errors, including IDS errors, strand breaks, and rearrangements, inevitably introduced during synthesis, amplification, sequencing, and storage processes. Sequence reconstruction, crucial for decoding, involves inferring the DNA reference from a cluster of erroneous copies. While most methods assume equal contributions from all reads within a cluster as noisy copies of the same reference, they often overlook the existence of contaminated sequences caused by DNA breaks, rearrangements, or mis-clustering reads. To address this issue, we propose RobuSeqNet, a robust multi-read reconstruction neural network specifically designed to robustly reconstruct multiple reads, accommodating noisy clusters with strand breakage, rearrangements, and mis-clustered strands. Leveraging the attention mechanism and an elaborate network design, RobuSeqNet exhibits resilience to highly-noisy clusters and effectively deals with in-strand IDS errors. The effectiveness and robustness of the proposed method are validated on three representative next-generation sequencing datasets. Results demonstrate that RobuSeqNet maintains high sequence reconstruction success rates of 99.74%, 99.58%, and 96.44% across three datasets, even in the presence of noisy clusters containing up to 20% contaminated sequences, outperforming known sequence reconstruction models. Additionally, in scenarios without contaminated sequences, it exhibits comparable performance to existing models, achieving success rates of 99.88%, 99.82%, and 97.68% across the three datasets.

A dual-rule encoding DNA storage system using chaotic mapping to control GC content.

DNA as a novel storage medium is considered an effective solution to the world's growing demand for information due to its high density and long-lasting reliability. However, early coding schemes ignored the biologically constrained nature of DNA sequences in pursuit of high density, leading to DNA synthesis and sequencing difficulties. This article proposes a novel DNA storage coding scheme. The system encodes half of the binary data using each of the two GC-content complementary encoding rules to obtain a DNA sequence. After simulating the encoding of representative document and image file formats, a DNA sequence strictly conforming to biological constraints was obtained, reaching a coding potential of 1.66 bit/nt. In the decoding process, a mechanism to prevent error propagation was introduced. The simulation results demonstrate that by adding Reed-Solomon code, 90% of the data can still be recovered after introducing a 2% error, proving that the proposed DNA storage scheme has high robustness and reliability. Availability and implementation: The source code for the codec scheme of this paper is available at https://github.com/Mooreniah/DNA-dual-rule-rotary-encoding-storage-system-DRRC.

"Cell Disk" DNA Storage System Capable of Random Reading and Rewriting.

DNA has emerged as an appealing material for information storage due to its great storage density and durability. Random reading and rewriting are essential tasks for practical large-scale data storage. However, they are currently difficult to implement simultaneously in a single DNA-based storage system, strongly limiting their practicability. Here, a "Cell Disk" storage system is presented, achieving high-density in vivo DNA data storage that enables both random reading and rewriting. In this system, each yeast cell is used as a chamber to store information, similar to a "disk block" but with the ability to self-replicate. Specifically, each genome of yeast cell has a customized CRISPR/Cas9-based "lock-and-key" module inserted, which allows selective retrieval, erasure, or rewriting of the targeted cell "block" from a pool of cells ("disk"). Additionally, a codec algorithm with lossless compression ability is developed to improve the information density of each cell "block". As a proof of concept, target-specific reading and rewriting of the compressed data from a mimic cell "disk" comprising up to 105 "blocks" are demonstrated and achieve high specificity and reliability. The "Cell Disk" system described here concurrently supports random reading and rewriting, and it should have great scalability for practical data storage use.

Reconstruction algorithms for DNA-storage systems

Motivated by DNA storage systems, this work presents the DNA reconstruction problem, in which a length-n string, is passing through the DNA-storage channel, which introduces deletion, insertion and substitution errors. This channel generates multiple noisy copies of the transmitted string which are called traces. A DNA reconstruction algorithm is a mapping which receives t traces as an input and produces an estimation of the original string. The goal in the DNA reconstruction problem is to minimize the edit distance between the original string and the algorithm’s estimation. In this work, we present several new algorithms for this problem. Our algorithms look globally on the entire sequence of the traces and use dynamic programming algorithms, which are used for the shortest common supersequence and the longest common subsequence problems, in order to decode the original string. Our algorithms do not require any limitations on the input and the number of traces, and more than that, they perform well even for error probabilities as high as 0.27. The algorithms have been tested on simulated data, on data from previous DNA storage experiments, and on a new synthesized dataset, and are shown to outperform previous algorithms in reconstruction accuracy.

FrameD: Framework for DNA-based Data Storage Design, Verification, and Validation.

DNA-based data storage is a quickly growing field that hopes to harness the massive theoretical information density of DNA molecules to produce a competitive next-generation storage medium suitable for archival data. In recent years, many DNA-based storage system designs have been proposed. Given that no common infrastructure exists for simulating these storage systems, comparing many different designs along with many different error models is increasingly difficult. To address this challenge we introduce FrameD, a simulation infrastructure for DNA storage systems that leverages the underlying modularity of DNA storage system designs to provide a framework to express different designs while being able to reuse common components. We demonstrate the utility of FrameD and the need for a common simulation platform using a case study. Our case study compares designs that utilize strand copies differently, some that align strand copies using Multiple Sequence Alignment (MSA) algorithms and others that do not. We found that the choice to include MSA in the pipeline is dependent on the error rate and the type of errors being injected and is not always beneficial. In addition to supporting a wide range of designs, FrameD provides the user with transparent parallelism to deal with a large number of reads from sequencing and the need for many fault injection iterations. We believe that FrameD fills a void in the tools publicly available to the DNA storage community by providing a modular and extensible framework with support for massive parallelism. As a result, it will help accelerate the design process of future DNA-based storage systems. The source code for FrameD along with the data generated during the demonstration of FrameD is available in a public Github repository at https://github.com/dna-storage/framed (10.5281/zenodo.7757762).

Weakly mutually uncorrelated codes with maximum run length constraint for DNA storage

DNA storage systems have begun to attract considerable attention as next-generation storage technologies due to their high densities and longevity. However, efficient primer design for random-access in synthesized DNA strands is still an issue that needs to be solved. Although previous studies have explored various constraints for primer design in DNA storage systems, there is no attention paid to the combination of weakly mutually uncorrelated codes with the maximum run length constraint. In this paper, we first propose a code design by combining weakly mutually uncorrelated codes with the maximum run length constraint. Moreover, we also explore the weakly mutually uncorrelated codes to satisfy combinations of maximum run length constraint with more constraints such as being almost-balanced and having large Hamming distance, which are also efficient constraints for random-access in DNA storage systems. To guarantee that the proposed codes can be adapted to primer design with variable length, we present modified code construction methods to achieve different lengths of the code. Then, we provide an analysis of the size of the proposed codes, which indicates the capacity to support primer design. Finally, we compare the codes with those of previous works to show that the proposed codes can always guarantee the maximum run length constraint, which is helpful for random-access for DNA storage.

Magnetic DNA random access memory with nanopore readouts and exponentially-scaled combinatorial addressing

The storage of data in DNA typically involves encoding and synthesizing data into short oligonucleotides, followed by reading with a sequencing instrument. Major challenges include the molecular consumption of synthesized DNA, basecalling errors, and limitations with scaling up read operations for individual data elements. Addressing these challenges, we describe a DNA storage system called MDRAM (Magnetic DNA-based Random Access Memory) that enables repetitive and efficient readouts of targeted files with nanopore-based sequencing. By conjugating synthesized DNA to magnetic agarose beads, we enabled repeated data readouts while preserving the original DNA analyte and maintaining data readout quality. MDRAM utilizes an efficient convolutional coding scheme that leverages soft information in raw nanopore sequencing signals to achieve information reading costs comparable to Illumina sequencing despite higher error rates. Finally, we demonstrate a proof-of-concept DNA-based proto-filesystem that enables an exponentially-scalable data address space using only small numbers of targeting primers for assembly and readout.

Content-based filter queries on DNA data storage systems

Recent developments in DNA data storage systems have revealed the great potential to store large amounts of data at a very high density with extremely long persistence and low cost. However, despite recent contributions to robust data encoding, current DNA storage systems offer limited support for random access on DNA storage devices due to restrictive biochemical constraints. Moreover, state-of-the-art approaches do not support content-based filter queries on DNA storage. This paper introduces the first encoding for DNA that enables content-based searches on structured data like relational database tables. We provide the details of the methods for coding and decoding millions of directly accessible data objects on DNA. We evaluate the derived codes on real data sets and verify their robustness.

An image cryptography method by highly error-prone DNA storage channel.

Introduction: Rapid development in synthetic technologies has boosted DNA as a potential medium for large-scale data storage. Meanwhile, how to implement data security in the DNA storage system is still an unsolved problem. Methods: In this article, we propose an image encryption method based on the modulation-based storage architecture. The key idea is to take advantage of the unpredictable modulation signals to encrypt images in highly error-prone DNA storage channels. Results and Discussion: Numerical results have demonstrated that our image encryption method is feasible and effective with excellent security against various attacks (statistical, differential, noise, and data loss). When compared with other methods such as the hybridization reactions of DNA molecules, the proposed method is more reliable and feasible for large-scale applications.

GCNSA: DNA storage encoding with a graph convolutional network and self-attention

DNA Encoding, as a key step in DNA storage, plays an important role in reading and writing accuracy and the storage error rate. However, currently, the encoding efficiency is not high enough and the encoding speed is not fast enough, which limits the performance of DNA storage systems. In this work, a DNA storage encoding system with a graph convolutional network and self-attention (GCNSA) is proposed. The experimental results show that DNA storage code constructed by GCNSA increases by 14.4% on average under the basic constraints, and by 5%-40% under other constraints. The increase of DNA storage codes effectively improves the storage density of 0.7-2.2% in the DNA storage system. The GCNSA predicted more DNA storage codes in less time while ensuring the quality of codes, which lays a foundation for higher read and write efficiency in DNA storage.

RepairNatrix: a Snakemake workflow for processing DNA sequencing data for DNA storage.

There has been rapid progress in the development of error-correcting and constrained codes for DNA storage systems in recent years. However, improving the steps for processing raw sequencing data for DNA storage has a lot of untapped potential for further progress. In particular, constraints can be used as prior information to improve the processing of DNA sequencing data. Furthermore, a workflow tailored to DNA storage codes enables fair comparisons between different approaches while leading to reproducible results. We present RepairNatrix, a read-processing workflow for DNA storage. RepairNatrix supports preprocessing of raw sequencing data for DNA storage applications and can be used to flag and heuristically repair constraint-violating sequences to further increase the recoverability of encoded data in the presence of errors. Compared to a preprocessing strategy without repair functionality, RepairNatrix reduced the number of raw reads required for the successful, error-free decoding of the input files by a factor of 25-35 across different datasets. RepairNatrix is available on Github: https://github.com/umr-ds/repairnatrix.

Generative Adversarial Networks for DNA Storage Channel Simulator

DNA data storage systems have rapidly developed with novel error-correcting techniques, random access algorithms, and query systems. However, designing an algorithm for DNA storage systems is challenging, mainly due to the unpredictable nature of errors and the extremely high price of experiments. Thus, a simulator is of interest that can imitate the error statistics of a DNA storage system and replace the experiments in developing processes. We introduce novel generative adversarial networks that learn DNA storage channel statistics. Our simulator takes oligos (DNA sequences to write) as an input and generates a FASTQ file that includes output DNA reads and quality scores as if the oligos are synthesized and sequenced. We trained the proposed simulator with data from a single experiment consisting of 14,400 input oligo strands and 12,108,573 output reads. The error statistics between the input and the output of the trained generator match the actual error statistics, including the error rate at each position, the number of errors for each nucleotide, and high-order statistics. The code is available at <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><uri>https://github.com/gyfbianhuanyun/DNA_storage_simulator_GAN</uri></monospace> .

RETRACTED: Bionic‐structure thermo‐responsive (best) hydrogels with controllable layer for high‐capacity DNA data storage

AbstractHere, a new‐rising technology (DNA storage) in which the binary digital information/data is converted into a DNA sequence composed of distinct nucleotides, providing a dense, stable, energy‐efficient, and sustainable data storage solution is studied. In principle, this technology offers substantial data density. However, the theoretical limit of DNA storage has not been achieved yet. In this study, a new DNA storage system is proposed based on hydrogels with SiO2 layer (their pore size can be regulated). The highest data density of the hydrogels is obtained to be 6.3 × 109 GB g−1. Further, by coupling them with digital polymerase chain reaction (dPCR), the recovery limit reaches five copies of DNA. Subsequently, three‐dimensional (3D) printing is used to explore the effect of macrostructures on the DNA storage capacity. Finally, bionic‐structure thermally‐responsive robust (Best) hydrogels with four different structures are developed, which achieve a data density of 1.04 × 1010 GB g−1.

Adaptive coding for DNA storage with high storage density and low coverage

The rapid development of information technology has generated substantial data, which urgently requires new storage media and storage methods. DNA, as a storage medium with high density, high durability, and ultra-long storage time characteristics, is promising as a potential solution. However, DNA storage is still in its infancy and suffers from low space utilization of DNA strands, high read coverage, and poor coding coupling. Therefore, in this work, an adaptive coding DNA storage system is proposed to use different coding schemes for different coding region locations, and the method of adaptively generating coding constraint thresholds is used to optimize at the system level to ensure the efficient operation of each link. Images, videos, and PDF files of size 698 KB were stored in DNA using adaptive coding algorithms. The data were sequenced and losslessly decoded into raw data. Compared with previous work, the DNA storage system implemented by adaptive coding proposed in this paper has high storage density and low read coverage, which promotes the development of carbon-based storage systems.

CMOS Capacitive Dry DNA Storage Monitoring: Design, Implementation and Experimental Results

This paper presents the design, and implementation of a new CMOS capacitive sensor and proposes the applicability of the sensor in monitoring the dried DNA mass for the emerging DNA memory technologies. The concept of DNA-based data storage has shown to be a promising solution with an ultra-high data density storage capability of gigabytes per nanogram mass of DNA. There are several drawbacks in achieving a ubiquitous technology and existing research has mostly addressed the translation of digital data to synthesized nucleotides, the physical storage of DNA, and reading out the data via sequencing, as well as computational processes. In this paper, we tackle the challenge of monitoring the physical storage of dried DNA material by developing a CMOS capacitive sensor that detects the dried DNA mass and inherits the highly integrable nature of CMOS chips with other parts to develop a DNA storage system. The successful detection of down to 45 ng <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$/ \boldsymbol {\mu }\text{L}$ </tex-math></inline-formula> mass of dried DNA has been successfully shown with the experimental results that consist of five runs for five different concentrations. Based on the results, the detection range of sub-pico Mol has been achieved which is compatible with the concentrations of DNA used in DNA memory technologies.

A machine learning-based approach to detect threats in bio-cyber DNA storage systems

Data storage is one of the main computing issues of this century. Not only storage devices are converging to strict physical limits, but also the amount of data generated by users is growing at an unbelievable rate. To face these challenges, data centres grew constantly over the past decades. However, this growth comes with a price, particularly from the environmental point of view. Among various promising media, DNA is one of the most fascinating candidate. In our previous work, we have proposed an automated archival architecture which uses bioengineered bacteria to store and retrieve data, previously encoded into DNA. The similarities between biological media and classical ones can be a drawback, as malicious parties might replicate traditional attacks on the former archival system, using biological instruments and techniques. In this paper, first we analyse the main characteristics of our storage system and the different types of attacks that could be executed on it. Then, aiming at identifying on-going attacks, we propose and evaluate detection techniques, which rely on traditional metrics and machine learning algorithms. We identify and adapt two suitable metrics for this purpose, namely generalized entropy and information distance.