Jim Ostell’s journey to NCBI?
“It was a long and winding and strange road,” he said with a laugh, looking back on life before he joined the National Center for Biotechnology Information (NCBI).
But when the road ended, Ostell found what he wanted and needed—and then some.
For the man who once gave away computer programs for free just when people were wondering how they might one day make money, he found an openness and freedom at NCBI.
As a person who enjoyed hanging out with but didn’t quite fit in with the do-it-yourselfers in his Vermont community (lute making anyone?), NCBI offered an intellectual home.
And as someone always up for a challenge, Ostell found many at NCBI—and he just took on the biggest of them all.
In September, Ostell, an NIH distinguished investigator, became the director of NCBI. A division of the National Library of Medicine, NCBI supports and maintains some of the world’s most widely used biomedical databases, including PubMed, GenBank, BLAST, Entrez, RefSeq, dbSNP, PubMed Central, and dbGaP. It also provides researchers with access to analysis and computing tools to better understand genes and their role in health and disease.
Classes that seemed inconsequential
Ostell’s journey to NCBI began innocently enough.
As a zoology major at the University of Massachusetts (UMass) in Amherst, Ostell took two classes that foreshadowed his future. He dismissed one as uninteresting and the other as impractical.
Biochemistry, which involved memorizing the metabolic pathway of the Krebs cycle (also known as the citric acid cycle) held little interest for Ostell, who was studying large animals. And then there was the course he took in programming, which focused on the latest developments in the field—punch cards and Fortran. “It was incredibly tedious; it was faster to work with a paper and pencil,” he said, “I thought, there’s no future in computers.”
“Of course, biochemistry and computer technology are the two things that I do now,” Ostell said with a grin.
This wasn’t science
Fresh out of college in 1972, Ostell landed a job collecting live materials—bullfrog tadpoles, insects, plants, mollusks—for a biological supply house.
“I thought it was the perfect job. I was outside all the time, working with natural organisms,” he said.
But then it became clear to him: “This wasn’t science, and it wasn’t good for the environment.”
After leaving the supply house, Ostell admits, “I was sort of unemployed for a while.”
But that was okay.
“This was the early 70s; it was counter-culture days,” he explained.
Ostell said he looked and acted the part. “I was long haired, shaggy, and wearing bell bottom jeans,” he explained.
He did some carpentry, picked apples, and freelanced as a handyman—until he realized he needed a bit of money, which meant a “real job.”
Getting hired as a medical technologist at a state hospital was easy for Ostell.
“They were desperate to hire someone, and I didn’t need certification. I had lots of chemistry so I could do the job,” he said. “I’d go up on the floors, draw blood, and do chemistries.”
He stayed working in hospital laboratories a few years, moving up to larger hospitals and better positions.
When he left the job, it wasn’t because of ethics or money. It was the lack of a challenge.
Back to class
Late in August of 1976, Ostell returned to UMass to talk to one of his favorite professors. Uncertain about his future, he confessed, “I’m thinking of going back to grad school, and I’m just looking for advice.”
The professor saw an opportunity for the zoology department and for Ostell. Two weeks later, Ostell was teaching cell physiology as a grad student and taking classes.
“I had no career plan in mind,” he explained. “It was more like if they’ll pay me to learn, I’ll do that.”
Having already been on his own and working, Ostell admitted he also had a bit of an attitude. He said, “I had been out in the world a little, so I wasn’t intimidated by the senior staff. I was just going to follow my nose and do what’s interesting.”
What interested him surprised the young man who had been so captivated by vertebrates.
One of Ostell’s first professors in grad school was an invertebrate zoologist and an electron microscopist. That’s how Ostell found out: “Invertebrates are incredible,” he said. “They can do everything imaginable.”
He cites the tunicates, a marine filter feeder that can concentrate the element vanadium, which is rarely found in nature, in their lumen. “Very bizarre,” said Ostell. Then there are tardigrades, creatures so resilient said Ostell that “they can dry out and stay desiccated for 100 years until you add water and they come back to life.”
It seemed inevitable: he would do an electron microscopic study of one of these invertebrates for his thesis.
He found just the right creature in the common cricket.
The body part he chose to study was the male accessory gland, which produces the capsule that contains the sperm. There wasn’t much literature available on the subject.
He was on his own.
Combining “classical, old school techniques,” including light microscopy and different histochemical techniques, along with protein electrophoresis, a new technique at the time, Ostell reconstructed the gland in three dimensions with classical anatomical drawing techniques (no digital 3D images at this time), and mapped the dozens of secretion products to the parts of the gland where they were produced.
For a graduate student who could wax poetic about the cricket, Ostell had the perfect name for his thesis: The Accessory Gland of the Male cricket, Acheta Domesticus: The Cadillac of Copulation.
He won first prize at the American Zoological Association graduate school conference for his work.
Yet a catchy title and the award couldn’t help Ostell publish his thesis.
That proved to be impossible.
After all, his thesis was more than 100 pages long, more of a classical anatomical monograph than a typical scientific paper of today.
But the thesis was also a way of saying “I’m back in academia.”
From crickets to Cambridge
But what would he study and where would he go?
“I started out studying anatomy. And then I asked, how does anatomy happen? That’s embryology. That’s development. What causes this development? Well it’s proteins, differential protein expression. And so now I was kind of up to the next stage which was what makes the proteins. Well, that’s DNA.”
His advisor suggested, “Maybe you should find a place that’s doing development in invertebrates, applying these modern techniques of DNA.” The advisor also suggested that Ostell not pick a school but pick a person.
Ostell knew who was who in the field. He mailed his thesis and application letters to specific professors at Caltech, UC-Berkeley, and Harvard. The first call came from Caltech, asking if he could fly out to campus.
Ostell was ready, with a caveat: “Are you paying?”
“I was living in a shabby apartment over a bar in Northampton, Massachusetts,” he said. “I didn’t have any money and had never been anywhere beyond New England.”
He was able to interview at Caltech and Berkeley in one trip. “My mother went out and bought me a suit so I would look somewhat respectable,” said Ostell. “I still had long hair and was wearing raggedy jeans.”
Arriving in California was a little surreal—blue skies, warm weather, and palm trees swaying in the breeze impressed the man from Massachusetts.
His first stop was Caltech.
There was only one problem: He was late.
Fortunately, Caltech had heard that his plane was delayed. All Ostell had to do was change into his suit, he explained to the department secretary when he arrived. But she just smiled and said that wouldn’t be necessary.
Ostell walked down the hall to find Eric Davidson, a long-haired professor wearing a fringed vest and cowboy boots with his feet up on the desk. “We had a great conversation,” recalled Ostell, “and then I spent a few days hanging out with the postdocs. It was a different world for me, because I’d gone to a college where I was always this weird, geeky science guy. And at Caltech, that was everyone. The whole trip was wonderful.”
His visit to Berkeley also went well, although Caltech would be a better fit.
But he hadn’t heard from Harvard.
Fresh from his trip out west, Ostell said, “I got a little cocky.” He called the chair of Harvard’s department of cellular and developmental biology, Fotis Kafatos, PhD, and asked if he had read his application letter and looked at his thesis.
Kafatos had received Ostell’s thesis but hadn’t read it. Ostell explained that he already had an offer from Caltech and that Dr. Kafatos “should at least do me the courtesy of looking at it.”
Kafatos did and called Ostell in for an interview.
Minus the weather and palm trees, the intellectual energy that Ostell loved at Caltech was also at Harvard.
Even though Ostell said that “Harvard initially had blown me off,” he liked Harvard, too.
Once again, he sought his advisor’s counsel. The molecular embryology department at Harvard could offer more flexibility because it was larger.
The long and winding road would continue at Harvard.
Next up: sequencing
And what an exciting time it was.
“Kafatos’ lab was the first in the world to develop a full-length cDNA,” he said. “It was all very heady. We were just getting into the genome and the structure of the DNA and stuff.”
Ostell learned to sequence DNA—manually, of course.
He says, “It was a big deal to sequence 100 kilobases of DNA then. It took months of painstaking work. Today we sequence megabases in a single sequencer run.”
But the task of translating DNA into protein was tedious, manual, and error-prone.
There had to be a better way.
Ostell started asking around: “What’s next on the journey after this? I’ve got the DNA. How do I understand this? And someone said, ‘I understand they’re starting to use computers to understand DNA.’”
A computer in an unlikely place
The technology had moved beyond punch cards, so Ostell was willing to try using computers.
Using a computer in the Computer Science Department at Harvard seemed like a logical choice, except they were only available for students in that department or students taking a computer science course. Ostell found access to a big center called MOLEGEN, an NIH-funded center at Stanford that was supposed to be able to use artificial intelligence for analyzing molecular biology. “It was incredibly difficult to use, and it didn’t tell me much,” said Ostell.
Undeterred, Ostell searched for other options.
He discovered a computer that might work in an unlikely place: the secretary’s desk in the Kafatos lab. It was also the only computer he had access to.
“It was a huge box with a big electric printer connected to a little monitor,” Ostell recalled. “It had 48 kilobytes of memory and used something called CPM, an operating system before PCs.”
Fortunately, there were some rudimentary instructions. “There was a photocopied Fortran manual by some little startup on the West Coast called Microsoft,” said Ostell. “Nobody’d ever heard of them.”
Using the computer came with a catch: He could only use it in the evening when the secretary wasn’t at work.
Night after night, Ostell worked at the computer. One evening he encountered so much difficulty he had to take apart the entire computer and solder a few parts back together. He finished rebuilding it moments before the secretary arrived.
Eventually Ostell started “producing a little programming.”
“Some people in the corridor heard I was doing this and asked, ‘Can we use your programs?’ I said, sure come on down, but it’s gotta be at night,” said Ostell. “So, I started to have some users.”
And these users had requests.
He agreed to help, despite an uncooperative computer.
“When printing, the computer would back up, miss lines, and screw up. I was sending too much data,” he said. “I realized there had to be a way to have the printer tell the computer when it’s got too much data and pause, but this operating system was so primitive that it didn’t support that, so I decided if I’m going to use this, I’m going to have fix this, so I learned assembly language and ended up programming the input/output chips on the computer and linking my custom drivers into the operating system.”
Maybe what Ostell was learning could help more scientists.
“I wrote a summary of how I wrote these programs and how they work for molecular biology for the first special issue of Nucleic Acid Research, which was featuring software and databases,” said Ostell.
Ostell did more than get his summary published.
He offered to send programs to anyone who was interested—for free.
The result? You might say it presages his work at NCBI, equipping the world’s scientists with the tools they need.
“I was swamped with requests,” said Ostell, who fielded requests from the United States, Europe, and the Eastern Bloc. “There wasn’t a whole collection of utilities that allowed molecular biologists to do molecular biology. And it was also the only one or one of the few that would run on the microcomputer.”
To accommodate customers with different microcomputers, Ostell learned how to do things in different disc formats and even onto tape. The program package became so portable it was eventually running on microcomputers, minicomputers, and mainframes all over the world.
Ostell was invited to teach a class called “Computer Anatomy and Physiology” for the biology department. He set up the secretary’s computer, monitor, and printer in front of the class.
“The first night, I walked out into the auditorium and the place was packed, and the whole front row was faculty,” he recalled.
He took a deep breath.
He looked up at his students and said, “Today, we’re going to learn how computers work.” He walked over to the computer, pressed one key on the monitor, and the printer printed one letter. “I said, ‘If at the end of the course, you can tell me what just happened, that’s the whole course.’”
Then he challenged the students: “How do we figure out what happened?”
His students were perplexed, but he explained what to do in terms they understood.
He told them, “We’re biologists. What’s the first thing we do? We kill it. So, I unplugged it. Then we cut it up. I opened it up, pulled out the boards and named them.”
Over the semester he covered binary logic, integrated circuits, hardware, assembly language, higher level languages, and operating systems. The class was a success.
About that time, he even received the okay to do a bit of commercial work.
But everything was far from hunky-dory.
To be continued on Tuesday, November 14.