Skip to main content
Premium Trial:

Request an Annual Quote

From Bench To Buyout: A Bioinformatics Pioneer Recalls The Building Of A Business


MADISON, Wisc.--Oxford Molecular Group's $20 million purchase of Genetics Computer Group (GCG) here last year not only gave Oxford entrée into the business of providing informatics for drug development, it ensured the longevity of one of the oldest bioinformatics companies around, executives from both firms told BioInform.

Oxford was interested in the 15-year-old GCG, which got its start at the same time as two other bioinformatics pioneers, IntelliGenetics (now also owned by Oxford) and DNAStar, because it provided expertise in an area in which the company had no core strength, according to Mark Cortelyou, Oxford's vice-president for business development in the bioinformatics group. Oxford provides information technology and wet chemistry services to laboratories and companies working on drug discovery.

Founded in the early 1980's, GCG was one of the early players in providing computer services and software for DNA sequence analysis and related work. The company now serves more than 30,000 scientists in 600 institutions worldwide, 70 percent of which are academic labs. The remainder are commercial companies, said John Devereux, GCG president and cofounder along with Maggie Smith, vice-president.

Customers of GCG's Wisconsin Package for sequence analysis receive bimonthly updates of data from GenBank and annual updates from GCG of over 100 interrelated software programs that enable them to search databases, align DNA sequences, locate open-reading frames, identify coding regions, and display restriction sites, among other tasks.

Today GCG's 38 employees occupy a one-story, 10,000-square-foot building here. But the company got its start in a lab on the University of Wisconsin, Madison, campus.

Devereux recalled his company's path from bench to buyout. He said he began in the lab of geneticist Oliver Smithies, who sequenced the genetic code for fetal hemoglobin in 1981. At a length of 11,375 nucleotides, that code made it apparent that computers had to be brought in to deal with all the data, said Devereux. "We had been writing the sequences in pencil on 11-by-14-inch graph paper, then trying to count them. At 11,000 nucleotides, that became impractical," he told BioInform.

The sequence was typed into a computer, where it could be analyzed by doing such things as looking for restriction maps, using software developed by Devereux, whose PhD is in computational genetics. "You could easily tell where the G, A, T, and C keys were--we wore them down," he chuckled.

But the situation in Smithies's lab soon changed dramatically for Devereux, setting him on the path to entrepreneurship. When President Reagan took office in 1981, federal spending reductions became the vogue. Smithies's research, funded by grants from the US National Institutes of Health, faced cuts, and Devereux was told that he was simply too expensive. He was forced to find other funding for his computer-based analytical efforts. That meant selling his services to other faculty members on campus.

"It was really a Reagan cutback that started the Genetics Computer Group," Devereux recounted. The company came to life in March 1982, after he hustled up business from university professors.

"It was a club. You were in it or you weren't. If you were in it you had to pay $2,000 a year. In my first year I think I got eight professors to sign up. That really wasn't enough to run it. We had to look for a wider audience to support this thing," he said.

Although still in Smithies's lab in the university's genetics department--GCG was operating as a cost center in the university--Devereux could not look to the school for support. He recalled meeting with the then-dean of the graduate school, the late Robert Bock, who told Devereux he was on his own. "You can get your users to pay for this or you can drive a cab, I don't care which. I'm not going to pay for it," Devereux recalled Bock saying.

Those unencouraging words were "probably the most valuable thing he could have said," Devereux now believes, because they forced him to shift the focus of his business. He attended a major meeting in Aspen, Colo., in 1982 where he met key players in the burgeoning field of genetic sequence comparisons. That got Devereux invited to a number of labs in the US, where he was able to sell his software, and he also met Greg Hamm, who ran the Nucleotide Sequence Data Library at the European Molecular Biology Laboratory (EMBL) in Heidelberg.

"They didn't have time to do software, but they needed software," Devereux related. "He adopted our software at EMBL, which is a kind of flagship in Europe." Hamm also got Devereux invited to discuss his software at a number of labs in Europe. Within the next 18 months the business was flourishing and "we had more money than we could spend," Devereux said.

By 1990 GCG had outgrown the university and left to go on its own. When it was spun off that year, the company took in an estimated $1 million. In that case, why did its owners agree to be bought out?

According to Devereux, once SmithKline Beecham made its $125 million investment in Human Genome Sciences' DNA sequence data, "our field went from a backwater to a core technology in drug discovery. Just about overnight the entire world had changed. This had become a big business."

"From a risk point of view, Maggie and I were running this little PT boat around a harbor, and there were these steel battleships that could easily sink us," he elaborated. The competition was well capitalized and well organized, he added.

"One of the reasons we liked Oxford was that it was still small and entrepreneurial. That appealed to us," he explained. "Maggie and I had our entire estates in this," he continued, and they expected there would be a shakeout in the field. Neither partner wanted to lose their investment, so they decided it was time to sell.

What Advice Does Devereux Have For Other Bioinformatics Entrepreneurs?

"There are too many bioinformatics companies," he warned. "There's no way to get comfortable in this field. It's going to be a real mud-wrestling show, and most of the players will die," Devereux predicted.

"Most people who have written software and have people pay attention to it have dreams of glory and don't realize how complicated the software industry is," he added. "These little companies are gambling they can make a better mousetrap and the world will come to them, but it's like there's a chain of hardware stores with 30 different kinds of mousetraps. Even if they don't have that particular mousetrap, they can develop it in the margins of their effort. It's going to be very hard for small players."

"The market will ultimately weed itself out," Devereux concluded.

--Harvey Black

Filed under

The Scan

Researchers Compare WGS, Exome Sequencing-Based Mendelian Disease Diagnosis

Investigators find a diagnostic edge for whole-genome sequencing, while highlighting the cost advantages and improving diagnostic rate of exome sequencing in EJHG.

Researchers Retrace Key Mutations in Reassorted H1N1 Swine Flu Virus With Avian-Like Features

Mutations in the acidic polymerase-coding gene boost the pathogenicity and transmissibility of Eurasian avian-like H1N1 swine influenza viruses, a PNAS paper finds.

Genome Sequences Reveal Evolutionary History of South America's Canids

An analysis in PNAS of South American canid species' genomes offers a look at their evolutionary history, as well as their relationships and adaptations.

Lung Cancer Response to Checkpoint Inhibitors Reflected in Circulating Tumor DNA

In non-small cell lung cancer patients, researchers find in JCO Precision Oncology that survival benefits after immune checkpoint blockade coincide with a dip in ctDNA levels.