Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 79 of the 80 references it cites were found in a registry or an index; 27 of the 49 sources its claims cite were checked for support.

DNA Data Storage.

Open access

Abstract

The demand for data storage is growing at an unprecedented rate, and current methods are not sufficient to accommodate such rapid growth due to their cost, space requirements, and energy consumption. Therefore, there is a need for a new, long-lasting data storage medium with high capacity, high data density, and high durability against extreme conditions. DNA is one of the most promising next-generation data carriers, with a storage density of 10¹⁹ bits of data per cubic centimeter, and its three-dimensional structure makes it about eight orders of magnitude denser than other storage media.

Overview

CoChat Summary

TL;DR

This paper is a brief review of DNA data storage, arguing that DNA is a promising next-generation medium because of its very high density, durability, and low-copying cost. It also surveys the main bottlenecks—encoding, synthesis, preservation, sequencing, and error correction—and notes that these challenges still prevent practical large-scale deployment.

Key findings

  • The paper states that DNA storage can reach a density of 10¹⁹ bits per cubic centimeter, and that 1.7 × 10¹⁹ bits can be stored in 1 g of DNA.
  • It reports that current conventional storage media are limited by cost, space, energy use, and durability, with maximum durability of about 50 years in optimal conditions.
  • For preservation, it states that ordinary solution storage can keep DNA for only about a year, while dehydration and other approaches can extend durability; one cited method achieved stability for 35 days at 65 °C (equivalent to two years at room temperature), and another micro-disc approach reached up to 10¹² bit/mm³ with durability of dehydrated DNA over 100 years.
  • In vivo storage examples include 455 KB of digital files encoded into 2304 kbp synthetic oligonucleotides in *E. coli*, and a 255 kbp yeast artificial chromosome encoding 38 KB of digital data with no mutation after the 100th generation.

Methods

  • The article is a narrative overview/review, not an original experimental study.
  • It synthesizes previously published work on DNA encoding, synthesis, preservation, sequencing, and error correction, citing examples such as Huffman coding, fountain codes, Reed–Solomon codes, Illumina sequencing, and Nanopore sequencing.
  • Its main analysis is descriptive: it compares reported approaches and performance figures from the literature, including storage density, error rates, and preservation durability.

Limitations

  • The paper explicitly states that there is still a need to refine methods for the fast and error-free synthesis of oligonucleotides and, longer term, long DNA chains, and that nucleotide-sequence reading must evolve toward greater credibility.
  • It also notes that current DNA storage methods remain limited by errors during synthesis, storage, and sequencing, and by reduced storage density from protective additives and indexing.
  • Evident from the text: this is a review article rather than a new empirical study, so it does not present a single unified dataset or original experimental results.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/biotech12020044

Cites

79 of 79 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 An example of coding the message “ramy” into an ASCII code. Converting binary data into nucleotide sequences is made by computer algorithms.
Figure 1 An example of coding the message “ramy” into an ASCII code. Converting binary data into nucleotide sequences is made by computer algorithms.CC BY · PubMed Central
Figure 2 The coding scheme implemented by Goldman et al. Digital information (a) is converted to base-3 (b) using a Huffman code and is subsequently is converted to DNA strings (c). Dividing DNA strings as shown generated four-fold redundancy (d).
Figure 2 The coding scheme implemented by Goldman et al. Digital information (a) is converted to base-3 (b) using a Huffman code and is subsequently is converted to DNA strings (c). Dividing DNA strings as shown generated four-fold redundancy (d).CC BY · PubMed Central
Figure 3 An example of coding music in DNA. Fragment of “Mary Had a Little Lamb” encoded using Huffman code. A nucleotide sequence corresponding to the music code is shown in (a) and the encryption part in (b). Adapted from Ailenberg and Rotstein [17].
Figure 3 An example of coding music in DNA. Fragment of “Mary Had a Little Lamb” encoded using Huffman code. A nucleotide sequence corresponding to the music code is shown in (a) and the encryption part in (b). Adapted from Ailenberg and Rotstein [17].CC BY · PubMed Central
Figure 4 Indication of elements of the nucleotide sequence in which a Little Lamb was encoded and an example image presenting a lamb from the “Mary Had a Little Lamb” rhyme encoded by Ailenberg and Rotstein [17]. The sequence of a file type defines it as an image. The geometric shape of the lamb enables the use of only 238 bp of DNA for encoding. Encoding has been performed using a template of signs indicating the type of shape and its spatial coordinates.
Figure 4 Indication of elements of the nucleotide sequence in which a Little Lamb was encoded and an example image presenting a lamb from the “Mary Had a Little Lamb” rhyme encoded by Ailenberg and Rotstein [17]. The sequence of a file type defines it as an image. The geometric shape of the lamb enables the use of only 238 bp of DNA for encoding. Encoding has been performed using a template of signs indicating the type of shape and its spatial coordinates.CC BY · PubMed Central
Figure 5 Smiling emoji and original Citizen Kane poster photograph encoded and decoded by Yazdi et al. [18]. The raw images were encoded and synthesized in the form of DNA strings (a,b). Images received after decoding without homopolymer check codes during processing (c,d). Images received after sequencing DNA strings when homopolymer error correction was made in order to reduce the number of errors that occurred during each encoding and decoding step (e,f). Two errors in the Citizen Kane file were sufficient to make the recovery of the image impossible. One error in the emoji did not influence the image quality.
Figure 5 Smiling emoji and original Citizen Kane poster photograph encoded and decoded by Yazdi et al. [18]. The raw images were encoded and synthesized in the form of DNA strings (a,b). Images received after decoding without homopolymer check codes during processing (c,d). Images received after sequencing DNA strings when homopolymer error correction was made in order to reduce the number of errors that occurred during each encoding and decoding step (e,f). Two errors in the Citizen Kane file were sufficient to make the recovery of the image impossible. One error in the emoji did not influence the image quality.CC BY · PubMed Central
Figure 6 A solid-phase method for the synthesis of oligonucleotides using photolabile compounds. A spacer containing the photolabile group is covalently joined to the surface. Once spots on the surface are exposed to UV light through slits in the physical mask, the photolabile protecting group is removed and the synthesis of oligonucleotide begins. The subsequent appropriate phosphoramidite with the photolabile group is then applied to the entire surface of the plate. It can form covalent bonds only in the absence of the preceding photolabile group. In the subsequent steps, additional spots are exposed to radiation, and another phosphoramidite is applied where necessary. Until the final oligonucleotide is completely synthesized, the chain-extending processes are repeated [29].
Figure 6 A solid-phase method for the synthesis of oligonucleotides using photolabile compounds. A spacer containing the photolabile group is covalently joined to the surface. Once spots on the surface are exposed to UV light through slits in the physical mask, the photolabile protecting group is removed and the synthesis of oligonucleotide begins. The subsequent appropriate phosphoramidite with the photolabile group is then applied to the entire surface of the plate. It can form covalent bonds only in the absence of the preceding photolabile group. In the subsequent steps, additional spots are exposed to radiation, and another phosphoramidite is applied where necessary. Until the final oligonucleotide is completely synthesized, the chain-extending processes are repeated [29].CC BY · PubMed Central
Figure 7 Principle of Reed–Solomon correction: first, the data is divided into parts, and each part is assigned x and y values that determine its location. Based on the coordinates, the points are matched to the polynomial function P(x), which is used to determine the parity symbols. Parity symbols are extra data points that match the original DNA sequence and are stored with the original data. When some of the original data are lost, the remaining data points and parity symbols can be used to recreate the original polynomial function and receive original data.
Figure 7 Principle of Reed–Solomon correction: first, the data is divided into parts, and each part is assigned x and y values that determine its location. Based on the coordinates, the points are matched to the polynomial function P(x), which is used to determine the parity symbols. Parity symbols are extra data points that match the original DNA sequence and are stored with the original data. When some of the original data are lost, the remaining data points and parity symbols can be used to recreate the original polynomial function and receive original data.CC BY · PubMed Central
Figure 8 Depiction of DNA fountain strategy.
Figure 8 Depiction of DNA fountain strategy.CC BY · PubMed Central

Verification

Claims the paper makes

25 Valid · 1 Not valid · 22 Not checked · 6 Own results and interpretations

26 more claims · 10 with their cited sources checked

Create a free account to see all 54 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 9
  • It may state two separate claims as one. 18

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

1 of 1

Citations

Cites

Show all 80 references

Cited by

See all 21

Metadata

Authors
Buko, T., Tuczko, N., & Ishikawa, T.
Affiliations
University of Warsaw (PL)
Corresponding author
Ishikawa, Takao
Venue
BioTech
Publisher
MDPI AG
Published
2023-06-01
Volume
12
Issue
2
Pages
44
ISSN
2673-6284 (ISSN-L)
OpenAlex
W4379232032
Cited by
21
References
80
Citations by year
2026: 11 · 2025: 6 · 2024: 4
FWCI
0.55
Type
journal article
Language
en
Open access
Locations
  • BioTech (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Physical Sciences › Mathematics › Theoretical Computer Science › DNA and Biological Computing
  • Physical Sciences › Computer Science › Computer Networks and Communications › Advanced Data Storage Technologies
  • Physical Sciences › Mathematics › Theoretical Computer Science › Algorithms and Data Compression
Keywords
DNA data storage, data density, storage durability, PCR amplification, oligonucleotide synthesis, long-term data preservation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-09 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: DNA Data Storage. https://app.cochat.ai/papers/10.3390/biotech12020044
The paper
Buko, T., Tuczko, N., & Ishikawa, T. (2023). DNA Data Storage. BioTech. https://doi.org/10.3390/biotech12020044

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.