Reading the Blueprint: How the Discovery of DNA’s Structure Launched an Industry
Every tool you use in a modern biology lab traces back to a single decade. PCR, restriction cloning, recombinant protein expression, DNA sequencing, the entire conceptual framework of genetic engineering: all of it was built between roughly 1970 and 1980 by a small number of people working across a handful of labs. Before that decade, biology was largely observational. Scientists could describe living systems in extraordinary detail but could not meaningfully intervene at the molecular level. After it, biology became engineering. The entire modern biotech industry (Moderna, Genentech, CRISPR Therapeutics, the mRNA vaccine platform that ended a pandemic) runs on intellectual infrastructure assembled by fewer than a dozen people in about ten years. Understanding how that happened is not just history. It is a template. The same pattern is unfolding right now in computational biology, in spatial genomics, in acoustic reporter genes. The people building the tools rarely know they are founding an industry. The people in Boyer’s lab in 1973 did not know either.
Four Groups, One Race
The story begins not in the 1970s but in the early 1950s, with a race that most people know only by its outcome.
By 1952, the scientific community broadly understood that DNA carried genetic information, but its three-dimensional structure was unknown, and without the structure the mechanism of heredity remained obscure. Four groups were working on the problem simultaneously. Linus Pauling at Caltech, already the world’s most famous chemist and the man who had solved the alpha-helix structure of proteins, was building molecular models and had recently proposed (incorrectly) a triple-helix structure for DNA. At King’s College London, Rosalind Franklin and her graduate student Raymond Gosling were using X-ray crystallography to produce the highest-resolution diffraction images of DNA fibers ever obtained. Also at King’s, Maurice Wilkins was pursuing a parallel crystallography effort largely separate from Franklin’s. At Cambridge, James Watson and Francis Crick were not running experiments at all. They were building physical models, synthesizing data from the literature and from other labs, trying to find a structural solution that fit everything known.
Franklin’s contribution to the resolution of this race is frequently understated and was, at the time, not acknowledged at all. Photo 51, the X-ray diffraction image she and Gosling produced in 1952, was the sharpest image of B-form DNA ever taken. It contained the critical measurements: the helical periodicity, the rise per base pair, the diameter of the helix, and the position of the phosphate backbone on the outside of the structure. These numbers were shown to Watson without Franklin’s knowledge by Wilkins, and Watson and Crick also had access to an unpublished Medical Research Council report containing Franklin’s unit cell measurements. When Watson saw Photo 51, he later wrote that his pulse began to race.
The 1953 Nature papers appeared as three back-to-back articles: Watson and Crick’s structural model, Franklin and Gosling’s X-ray data, and Wilkins’ fiber diffraction work. Franklin’s paper was presented as experimental support for a model that had been built substantially on her unpublished measurements. She was not consulted about this arrangement. She died of ovarian cancer in 1958, four years before Watson, Crick, and Wilkins received the Nobel Prize. The Nobel is not awarded posthumously.
The structure Watson and Crick proposed was correct. The double helix, with its two antiparallel strands held together by specific base pairs (adenine with thymine, guanine with cytosine), immediately suggested how genetic information was stored and copied. Their paper contained one of the most understated sentences in the history of science: “It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material.” Base complementarity was not just a structural feature. It was the principle of inheritance.
The Gap Between Knowing and Doing
The double helix was published in 1953. The first biotech company was founded in 1976. That twenty-three year gap is not a story of slow uptake. It is a story of necessary tool-building.
Knowing the structure of DNA did not immediately make DNA manipulable. You could understand how the molecule worked without being able to cut it at a specific site, copy a particular gene, insert a sequence into a new host, or read the order of bases along a strand. Each of those capabilities required a separate discovery, and they arrived in sequence across the 1960s and 1970s.
The genetic code was cracked between 1961 and 1966, establishing the correspondence between DNA triplets and amino acids. DNA polymerases and RNA polymerases were characterized, revealing the molecular machinery of replication and transcription. Reverse transcriptase was discovered in 1970 by Howard Temin and David Baltimore, showing that genetic information could flow from RNA back to DNA, a finding that upended the central dogma and opened the door to working with RNA as a molecular tool.
The two discoveries that made genetic engineering directly possible arrived in rapid succession. In 1970 and 1971, Werner Arber, Hamilton Smith, and Daniel Nathans characterized restriction endonucleases: bacterial enzymes that cut double-stranded DNA at specific sequence motifs. Different enzymes recognized different sequences. A tool for cutting DNA at a chosen address now existed. Arber, Smith, and Nathans received the Nobel Prize in 1978. In 1973, Stanley Cohen at Stanford and Herbert Boyer at UCSF combined restriction enzyme cutting with DNA ligation to insert a frog gene into a bacterial plasmid, replicate it in E. coli, and demonstrate that the foreign gene was expressed. This was the first recombinant DNA organism. The founding experiment of genetic engineering had been done, and Boyer knew immediately what it meant.
The Asilomar Pause and the Policy That Made an Industry
The power of recombinant DNA technology was obvious enough that it frightened the people who invented it. In 1974, a group of leading molecular biologists including Paul Berg, Boyer, and others published a letter in Science calling for a voluntary moratorium on certain recombinant DNA experiments until the safety implications could be assessed. In February 1975, 140 scientists gathered at the Asilomar Conference Center in Pacific Grove, California, to debate the risks. The result was a framework of biosafety guidelines that allowed the research to continue under defined containment conditions. Asilomar is still cited as one of the few examples in history of a scientific community proactively regulating itself before harm occurred.
Research resumed. In 1977, Frederick Sanger published the chain-termination method for reading DNA sequences, making it possible to determine the order of bases along any DNA strand. You could now make recombinant DNA and verify exactly what you had made.
In 1976, Robert Swanson, a venture capitalist with no scientific background, cold-called Herbert Boyer and asked for fifteen minutes. The meeting lasted three hours. They co-founded Genentech on a $500 investment each. In 1982, Genentech and Eli Lilly brought recombinant human insulin (Humulin) to market: the first bioengineered therapeutic protein approved for human use, produced by inserting the human insulin gene into E. coli and letting bacteria do the manufacturing. It was also, as described in the first post in this series, the proof of concept that made the entire GLP-1 drug development arc possible.
The Policy That Opened the Floodgates
One more piece was needed to turn academic discovery into industry at scale. In 1980, the Bayh-Dole Act passed the US Congress, allowing universities to patent inventions made with federal research funding and to license those patents to private companies. Before Bayh-Dole, discoveries made with NIH or NSF money were considered public property. After it, a university lab that discovered a new gene, enzyme, or method could own that discovery and collect royalties on its commercial use. The incentive structure for academic-to-industry technology transfer changed overnight.
Amgen was founded in 1980. Biogen in 1978. The second wave of biotech companies each built on the recombinant DNA platform that Boyer and Cohen had demonstrated and that Sanger’s sequencing method allowed researchers to characterize. Each new tool enabled the next: restriction enzymes made recombinant DNA possible, recombinant DNA made engineered proteins possible, sequencing made verification possible, and PCR (invented by Kary Mullis in 1983) made copying and amplifying any DNA fragment possible from vanishingly small starting material.
The decade from 1970 to 1980 was not a period of one breakthrough. It was a period of tool accumulation, each tool enabling the next, none of them individually sufficient, all of them together transformative. The researchers who built them were mostly thinking about the next experiment, not the industry their work would generate. Boyer famously kept the Genentech founding a secret from his colleagues at UCSF for months, uncertain how academia would receive the idea of a professor starting a company.
The pattern has repeated. The researchers who developed CRISPR in 2012 were thinking about bacterial immune systems. The researchers building acoustic reporter genes are thinking about ultrasound contrast. The researchers doing spatial transcriptomics are thinking about tissue architecture. The industries those tools generate will be named later, by other people, once the tools have had time to propagate. That is how it has always worked.