Bringing a Virtual Brain to Life
By TIM REQUARTH
For months, Henry Markram and his team had been feeding data into a supercomputer, four vending-machine-size black boxes whirring quietly in the basement of the Swiss Federal Institute of Technology in Lausanne.
The Blue Brain computer has 10,000 virtual neurons. The colors represent the neurons’ electric voltage at a specific moment.
The boxes housed thousands of microchips, each programmed to act like a brain cell. Cables carried signals from microchip to microchip, just as cells do in a real brain.
In 2006, Dr. Markram flipped the switch. Blue Brain, a tangled web of nearly 10,000 virtual neurons, crackled to life. As millions of signals raced along the cables, electrical activity resembling real brain waves emerged.
“That was an incredible moment,” he said, comparing the simulation to what goes on in real brain tissue. “It didn’t match perfectly, but it was pretty good. As a biologist, I was amazed.”
Deciding then that simulating the entire brain on a supercomputer would be possible within his lifetime, Dr. Markram, now 50, set out to prove it.
That is no small feat. The brain contains nearly 100 billion neurons organized into networks with 100 trillion total connections, all firing split-second spikes of voltage in a broth of complex biological molecules in constant flux.
In 2009, Dr. Markram conceived of the Human Brain Project, a sprawling and controversial initiative of more than 150 institutions around the world that he hopes will bring scientists together to realize his dream.
In January, the European Union raised the stakes by awarding the project a 10-year grant of up to $1.3 billion — an unheard-of sum in neuroscience.
“A meticulous virtual copy of the human brain,” Dr. Markram wrote in Scientific American, “would enable basic research on brain cells and circuits or computer-based drug trials.”
An equally ambitious “big brain” idea is in the works in the United States: The Obama administration is expected to propose its own project, with up to $3 billion allocated over a decade to develop technologies to track the electrical activity of every neuron in the brain.
But just as many obstacles stand in the way of the American project, a number of scientists have expressed serious reservations about Dr. Markram’s project.
Some say we don’t know enough about the brain to simulate it on a supercomputer. And even if we did, these critics ask, what would be the value of building such a complicated “virtual brain”?
Henry Markram traces his fascination with the brain to a school assignment in his native South Africa. He was 14, and as he sat in the library reading about depression, he was astonished to discover there might be “molecular explanations to mental illness” that could be treated with drugs.
That set him on a path to medical school, where he planned to become a psychiatrist. But as a medical student, he realized that we know next to nothing about what prescription drugs really do to the brain.
To understand mental illness, he reasoned, we need to understand the brain first. “So I dropped out of medical school,” he said, “and got on a plane to do some real neuroscience.”
He went to the Weizmann Institute of Science in Israel to earn a Ph.D., followed by a stint at the National Institutes of Health in the United States on a Fulbright scholarship. That work led to a position with the Nobel Prize-winning neurophysiologist Bert Sakmann at the Max Planck Institute in Germany.
At Dr. Sakmann’s lab, Dr. Markram made his most famous discovery.
He was pondering how the brain learns cause and effect. He set up an experiment to record the electrical activity from two connected neurons in a slice from a rat’s brain, and discovered that the neurons required a precise sequence of voltage spikes to change the strength of their connections. He speculated that the mechanism might be at the root of our notion of causality.
That work has now been cited thousands of times. Yet as Dr. Markram’s reputation grew, so did his impatience.
Neurons are organized into interconnected circuits that can number in the millions. Dr. Markram realized that to make real progress linking neurons to behavior, experimenting on two neurons at a time “just wasn’t enough.”
In his first faculty position, at the Weizmann Institute, he set up a wildly ambitious new experimental rig that could record data not just from 2 neurons in a rat’s brain but also from 12.
“His rig made NASA look tame,” recalled Dr. Markram’s postdoctoral adviser at the N.I.H., Elise F. Stanley, who visited him at the Weizmann in 1995. “There was so much equipment that you couldn’t even see the brain tissue.”
Soon Dr. Markram would learn that his son, Kai, had autism. “You know how powerless you feel,” he said. “You have this child with autism, and you, even as a neuroscientist, really don’t know what to do.”
He began to question the impact of his work. “I realized that I could write a high-profile research paper every year, but then what?” he said. “I die, and there’s going to be a column on my grave with a list of beautiful papers.”
Dr. Markram decided he needed to change his approach. Experiments, he realized, were not enough.
After hearing of a new I.B.M. supercomputer, he asked himself, What if each microchip of the supercomputer represented a neuron in the brain? You could run simulations to perform virtual experiments and, unlike in real experiments, watch every single “neuron” in action. “If I build in enough biological detail,” he reasoned, “it would behave like a real brain.”
Dr. Markram moved his lab to the Swiss Federal Institute of Technology, which agreed to buy the $10 million supercomputer. Armed with data from 20,000 experiments, Dr. Markram began to build Blue Brain.
By 2008, he said, his team had created a “digital facsimile” of a cylindrical piece of tissue in the rat cortex. In 2011, the team announced it had simulated a “virtual slice” of brain tissue with one million neurons.
He proposed the Human Brain Project, which would scale up Blue Brain to simulate the human brain. Dr. Markram would not be able to do it alone, so he appealed to the broader scientific community for support.
But many scientists are highly skeptical of Blue Brain’s accomplishments.
While the team may have achieved a computer simulation of something, critics say, it was not a brain slice.
“It was completely meaningless, just random activity,” said Alexandre Pouget, a neuroscientist at the University of Geneva, referring to the stunning visualizations that Dr. Markram’s group presents at conferences. “The claim that he simulated a rat’s cortex is completely ridiculous.”
And in a time of increasing competition for research grants, some scientists worry that the Human Brain Project will make funds even scarcer. “There could be indirect effects,” acknowledged Andrew Houghton, a deputy at the European Commission.
But concerns run even deeper.
Some researchers say it is premature to invest money in a simulation while important principles of brain function remain to be discovered.
“We’re probably in the time of Galileo in biology,” said Christof Koch, of the Allen Institute for Brain Science in Seattle. “Darwin, Crick and Watson have given us the equivalent of Galileo and Newton, but there isn’t any equivalent to Einstein’s theory of relativity.”
Other critics say the project is too open-ended — that it makes little sense without clearly defined criteria for success.
“It’s not like the Human Genome Project, where you just have to read out a few billion base pairs and you’re done,” said Peter Dayan, a neuroscientist at University College London. “For the human brain, what would you need to know to build a simulation? That’s a huge research question, and it has to do with what’s important to know about the brain.”
And Haim Sompolinsky, a neuroscientist at the Hebrew University of Jerusalem, said: “The rhetoric is that in a decade they will be able to reverse-engineer the human brain in computers. This is fantasy. Nothing will come close to it in a decade.”
Some say the controversy surrounding Dr. Markram’s work distracts from the real issue: How should neuroscience harness its resources to achieve true understanding of the brain?
“Some 10,000 laboratories worldwide are pursuing distinct questions about the brain,” Dr. Koch of the Allen Institute wrote in the journal Nature. “Neuroscience is a splintered field.”
Dr. Markram agreed. The Human Brain Project, he said, will provide a “unifying principle” for scientists to rally around.
For the first time, data from laboratories around the world will be in one place, he said, adding that trying to build a simulation will drive advances in fields like computing and robotics. An entire division of the project is devoted to creating a new breed of intelligent robots with “neuromorphic” microchips designed like neurons in the human brain.
“The biggest success for me,” Dr. Markram said, “would be if after 10 years we have a new model for neuroscience, where everyone works together. It’s about a new foundation.
“Putting the problem on the horizon is very important,” he continued. “When people say, ‘Well, the brain is so complicated that our grandchildren will solve it,’ we put it over the horizon.”