Biotechnology & Synthetic Biology

A Silicon Chip That Writes DNA in Water, One Sequence at a Time

The same industry that gave us microprocessors has built a chip that prints genetic code: 64 distinct DNA sequences, written in parallel using nothing but electricity, enzymes, and water.

• The Harvard-led chip synthesized 64 different DNA strands simultaneously, each up to 39 nucleotides long, on a single silicon surface.
• Electric current lowers the pH at selected spots on the chip to trigger an enzymatic reaction, while leaving neighboring sites untouched — a kind of electronic stencil.
• Water replaces the harsh organic solvents that have dominated synthetic DNA production since the phosphoramidite chemistry of the 1980s.

Custom-made DNA is the raw material of modern biotechnology: diagnostics, gene editing, gene therapies, and lab-grown proteins all depend on being able to manufacture exact genetic sequences quickly and cleanly. Today, almost all of that DNA is produced by phosphoramidite chemistry, a 40-year-old process that stitches nucleotides onto a support strand using reactive phosphorus compounds and strong organic solvents. It works well, but it is wasteful, and the solvents make scale-up messy and expensive.

The new approach uses enzymes — the same kind of molecular machines that living cells use to copy and build DNA — which run in water. The tricky part is control. To write many sequences in parallel, the chip must selectively activate the enzyme only where a particular letter is being added, without spilling over into the strand next door. The chip solves this with its own silicon circuitry: by applying a precise voltage, it locally drops the pH exactly where synthesis is needed, firing the enzymatic step on command and keeping everything else quiet. Each synthesis cycle places one nucleotide across all 64 sites, and repeating the cycle builds the full strands.

The practical payoff is a cleaner, more scalable way to manufacture synthetic DNA. Enzymatic synthesis in water produces less hazardous waste and sidesteps the solvent disposal that limits how cheaply custom DNA can be made at scale. If the technology matures, it could lower the cost of gene therapies and diagnostics, and let biologists iterate designs faster — closer to the way software engineers spin up new code than the way chemical engineers refine a batch.

It is worth being clear about what this is and is not. The chip produces short sequences up to 39 nucleotides; stitching these into the long gene-length strands that therapies require still needs additional assembly steps. It is a better printing head, not yet a finished printer. Still, swapping organic chemistry for enzymatic chemistry on a semiconductor platform is a structural change in how biological materials are manufactured, and the silicon-dna intersection is exactly where the next generation of biomanufacturing is headed.

Knowledge takeaway: Enzymes that copy DNA can run in water instead of toxic solvents; a silicon chip uses electrically controlled pH to write many sequences in parallel; the process is cleaner and points toward cheaper synthetic biology at scale.