Storing Digital Data in DNA: Advances Toward Practical Implementation

Abstract

The exponential growth of global digital data, projected to exceed 175 zettabytes by 2025, has exposed fundamental limitations in conventional storage infrastructure, including finite media lifespan and unsustainable energy consumption. This review examines synthetic deoxyribonucleic acid (DNA) as a next-generation archival storage medium, systematically analyzing the encoding architectures, synthesis methodologies, error correction strategies, and retrieval mechanisms that define the DNA data storage pipeline. DNA offers volumetric storage densities of approximately 500 GB/mm³ and molecular stability, enabling data preservation across centuries. The four-nucleotide alphabet theoretically permits 2 bits per nucleotide, though biochemical constraints reduce practical efficiency to 1.5–1.98 bits per nucleotide. Error correction has advanced substantially, with fountain code implementations achieving near-theoretical efficiency at sequencing coverage depths below 11x. However, significant barriers impede deployment: synthesis costs of $0.05–$0.10 per base, write throughput limited to kilobytes per second, and absent universal standards. Recent advances, including CRISPR-based overwriting, microfluidic platforms achieving 97.67% synthesis yields, and enzymatic methods producing oligonucleotides exceeding 1,000 nucleotides, offer promising pathways toward economic viability. The review concludes that DNA storage is approaching feasibility for hybrid archival hierarchies targeting rarely accessed data, with initial commercial deployment anticipated within 10–20 years.