DNA sequencer

The automated DNA sequencer emerged because molecular biology was generating more data than humans could process. Frederick Sanger's dideoxy method worked beautifully—but it required radioactive labeling, manual gel handling, and technicians reading autoradiographs by eye. A skilled operator might produce a few hundred base pairs per day. The Human Genome Project was already being discussed; three billion base pairs at that rate would take millennia.

Leroy Hood at Caltech saw that the bottleneck wasn't chemistry but detection. In 1985, he and his colleagues—Lloyd Smith, Mike Hunkapiller, and Tim Hunkapiller—proposed replacing radioactive labels with four different colored fluorescent dyes, one for each nucleotide. The fragments could be run in a single gel lane instead of four, and a laser detector could read the sequence automatically. No more squinting at X-ray films.

The adjacent possible had aligned. Argon ion lasers could excite fluorescent dyes efficiently. Hewlett-Packard computers could record and analyze the data. Applied Biosystems, where Mike Hunkapiller worked, had the manufacturing capability to commercialize the instrument. Hood's academic insight met industrial execution.

The ABI Model 370A launched in 1986, the first commercial automated DNA sequencer. It used slab gel electrophoresis, could run sixteen samples simultaneously, and output sequence data directly to computer. Early adopters included J. Craig Venter at NIH, who would later lead the private effort to sequence the human genome. Within a year, the Model 370A had transformed genetic research from artisanal craft to industrialized data production.

The cascade of improvements accelerated exponentially. Capillary electrophoresis replaced slab gels in the 1990s, increasing throughput and eliminating gel preparation. The ABI 377 and 3700 series became workhorses of the Human Genome Project. By 2000, the project's sequencing centers operated hundreds of machines running 24/7. What had taken Sanger years to sequence manually—bacteriophage λ's 48,502 base pairs—could be completed in an afternoon.

Next-generation sequencing in the 2000s would eventually displace Sanger-based automation, using massively parallel approaches that read millions of fragments simultaneously. But the conceptual leap from radioactive manual sequencing to fluorescent automated detection—the transformation Hood pioneered—made large-scale genomics possible. Without the Model 370A, there would have been no Human Genome Project.

By 2026, sequencing has become routine. Clinical laboratories sequence patients' tumors to guide treatment. Consumer companies sequence millions of customers for ancestry insights. The $1,000 genome—once a moonshot goal—became reality and then obsolete. The instrument Hood and the Hunkapillers conceived in a Caltech lab transformed biology from a hypothesis-driven science to a data-driven one.