A gram of lab-made DNA can store roughly as much data as a warehouse of hard drives

Digital files can be translated into sequences built from the four chemical letters of DNA. Because molecular strands pack information at extraordinary density, a tiny amount of synthetic material can theoretically replace racks of magnetic or solid-state storage.

A landmark laboratory system recovered data perfectly at an estimated density of 215 petabytes per gram. The result made the warehouse comparison plausible, but it did not turn DNA into a fast, inexpensive replacement for a data center.

Four chemical bases become a digital alphabet

DNA normally stores biological instructions in sequences of adenine, cytosine, guanine and thymine. A storage system maps binary information into those bases, breaks a file into many short sequences and adds addresses plus error-correcting information.

The sequences are chemically synthesized, stored and later read by a DNA sequencer. Software then identifies the fragments, corrects errors and reconstructs the original file. Unlike a living genome, archival DNA does not need cells; it can be kept as purified molecules in protected containers.

DNA Fountain reached 215 petabytes per gram

Researchers encoded an operating system, a film, a gift card and other files using an approach called DNA Fountain. The code created overlapping packets that allowed recovery even when some synthesized strands were missing.

The 2017 Science paper reported perfect retrieval at 215 petabytes per gram, approaching the information capacity available per nucleotide under its design. A petabyte is one million gigabytes. At that density, one gram corresponds to many thousands of modern high-capacity drives.

Physical density is only one part of a storage system

The quoted gram refers to DNA carrying encoded information, not an entire operational archive. Practical installations also require containers, indexing, automation, sequencers, synthesis equipment and redundant copies. Error correction reduces usable capacity, while access patterns determine how much extra material is needed.

A Columbia University summary of the experiment noted that the method approached 90 percent of the theoretical maximum for its coding model. That achievement addressed density and robustness, not the cost or speed required for daily computing.

Writing remains slow and expensive

Hard drives write and retrieve information electronically in fractions of a second. DNA synthesis builds molecules through chemical steps and sequencing reads enormous numbers of fragments before software reassembles them. Latency makes present systems better suited to cold archives than frequently edited files.

Random access is another challenge. A system needs molecular tags that retrieve a selected file without sequencing the entire pool. Rewriting can require creating new strands rather than flipping bits in place. Automation and enzymatic synthesis may improve both problems, but silicon remains far more practical for active workloads.

Stability gives molecules a different advantage

Properly dried and shielded DNA can retain readable information for long periods without continuous electricity. A Nature Communications review notes both immense information density and the central importance of protecting DNA from heat, moisture, oxidation and other damage.

The format could eventually hold scientific records, cultural archives and compliance data that must survive technology cycles. Its promise is not a microscopic laptop drive. It is a compact, low-power medium for information that may sit untouched for years and still need to be recovered after today’s hardware standards are obsolete.

The gram-sized comparison reveals the power of molecular encoding while hiding the machinery around it. DNA has demonstrated warehouse-scale density in the laboratory; making that density economical, searchable and routine is the remaining engineering project.

Error correction makes imperfect molecules dependable

Synthesis can omit or substitute bases, storage can break strands, and sequencing can misread them. A digital archive must recover files even when some molecules vanish completely. Codes add structured redundancy so missing fragments can be reconstructed from overlapping information.

Addresses embedded in each strand identify its position, but those addresses consume capacity and can themselves contain errors. Designers also avoid sequences that are unusually difficult to synthesize or read, including long runs of the same base and extreme chemical compositions.

DNA Fountain borrowed ideas from fountain codes used in communication systems. It generated enough independent packets that the original data could be recovered from almost any sufficiently large subset. That approach is a major reason the experiment achieved both high density and perfect retrieval.

Preservation needs copies and readable instructions

A molecular archive lasting centuries is useful only if future operators know how to decode it. File formats, base mappings, error-correction rules and indexing schemes must be preserved outside the payload or represented in a self-describing form.

Multiple physical copies protect against fire, contamination and handling mistakes. Encapsulation in silica or other barriers can slow chemical damage, while cool and dry storage extends life. The molecule is durable under the right conditions, not indestructible.

These requirements reduce the dramatic elegance of one gram sitting alone in a vial. They also turn a density demonstration into an archival system. The central tradeoff resembles older media: compact storage is valuable only when integrity, access and institutional memory survive with it.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview