Garry Nolan is an inventor and Stanford professor in the Department of Pathology. With over 350 published research articles and 50 U.S. patents, Nolan has been recognized as one of the top 25 inventors at Stanford University.
He tells EpochTV’s “Bay Area Innovators” about some of his inventions and the secrets behind his success.
“I’ve had plenty of people tell me that the ideas I have will not work. And my response is usually, ‘If you don’t see the future I see, I’m not going to sit here and argue with you all day. I’m just going to go back to the laboratory and prove it,’” Nolan said.
He trained under Leonard Herzenberg for his doctorate and Nobel laureate David Baltimore for postdoctoral work.
One of his inventions is a 293T cell retroviral production system for gene therapy, which delivers genes to make proteins or other RNA.
“A retroviral vector that can carry genes to the target cells. It is simultaneously a way to deactivate the bad aspect of the virus and then reinvent it so that it can be a vehicle for like a car, to carry the cargo of interest to us,” Nolan said.
Nolan said the retroviral vector systems were originally invented by Richard Mulligan, and it took three months to make the retrovirus.
Nolan was working with other researchers in the laboratory of Baltimore, who won a Nobel Prize for retroviruses. There, another postdoc, Warren Pear, was having trouble making a retroviral vector, the way Mulligan did.
“It occurred to me as a gee, if we just were to introduce the DNA quickly into this cell line and get out, because we could transfect deliver 60 percent of the cells to produce the DNA to produce the retrovirus. That would be a quick turnaround. So Warren and I worked together to create the first of the lines,” Nolan said.
However, they realized that the lines were unstable and lost the ability to make the retrovirus.
When Nolan got his position at Stanford and his own laboratory, he created a stable cell line. Together, he and Pear reinvented the cells and called them the Phoenix lines.
“Phoenix means, if you remember the ancient Egyptian story of rising from the ashes. So the idea was that DNA rises from the ashes, becomes a retrovirus again, lives again, and can be used to deliver genes,” Nolan said.
The Phoenix lines became so popular that they gave them out to thousands of laboratories across the world.
“When it was time to start to think about gene delivery for gene therapy in the human realm, to fix people who might have genetic defects, ... it became the system for developing retroviruses for gene delivery in humans,” he said. “People have used it for delivering genes to the brain, delivering genes to the retina, delivering genes all over the place.”
Nolan said that this form of DNA modification just alters the DNA and does not get passed on to the next generation.
His laboratory is now a place where he invents an updated or more efficient way to produce data and answer often hard-to-solve questions.
“I want to give people so much data that they feel like they’re swimming in an opera in a sea of opportunity themselves,” he said. “Because I hate seeing people do things the way that I know is slow, and so why not give them the tool and make it available to them, and provide them, not only the tool, but then the analytic procedures?”
In addition to his research, Nolan has founded several biotechnology companies, two of which are currently trading on NASDAQ.
More recently, he has been involved with artificial intelligence to try to make sense of all that data. He and his team created another company and found that they can feed raw data into large language models. They took structured data to create a step-by-step algorithm to produce data for the next algorithm in the computer program.
“We basically created a scientist in a box. We train the large language model to be a scientist, to think like a scientist, and so we give it the data, and it produces answers. So it’s as good as a sophisticated graduate student or a sophisticated professor. In some cases, you still have to double check. It’s not perfect. It occasionally hallucinates, but we’ve done the best we can to limit that,” Nolan said.
He said the model can figure out what a person wants from a few simple sentences. For example, asking the right question can allow it to generate a hypothesis and explain what experiments should be done.
While some people say this could make students lazy, Nolan said that smart students will find something to do, and “if they don’t, then they shouldn’t be in science.”
“The thing that angers me the most is when I see opportunity lost,” Nolan said. “Usually, what I tell students is, ‘If you know you’re right and you can see the steps to get there clearly, don’t let somebody else dissuade you. Don’t let them demoralize you. Just do it.’”
