Paul Allen’s Big Bet to ‘Uncover the Essence of What Makes Us Human’

3/22/12Follow @xconomy

Following his second brush with cancer a couple of years ago, billionaire Paul Allen had some time to think about the legacy he’ll leave, beyond being the co-founder of Microsoft. Yesterday, he made clear that he wants to be the guy who helped spark new understanding of the human brain.

“I’ve always been fascinated by the workings of the human brain, and awed by its enormous complexity,” Allen said at a press conference announcing his new $300 million commitment to brain science. “Our brains are many magnitudes more advanced, in the way they work, than any computer software.” He added: “Our dream is to uncover the essence of what makes us human.”

There was a lot more ambitious rhetoric going around yesterday at this press conference, in which Allen announced his plan to supercharge the efforts of the Seattle-based nonprofit Allen Institute for Brain Science. The four-year, $300 million commitment from Allen brings his total support for the nonprofit center to more than $500 million since its founding in 2003. This massive new push for brain science drew press attention from the big guns at the Wall Street Journal, New York Times, Forbes, Bloomberg News, and Time, as well as a litany of local biotech power brokers. Lee Hood of the Institute for Systems Biology, Larry Corey of the Fred Hutchinson Cancer Research Center, Ken Stuart of Seattle Biomedical Research Institute, and Chris Rivera of the WBBA were among those who turned out, and generally beamed about Allen’s big new contribution.

“It’s enormously valuable, exactly the right thing to do,” Hood said afterward, noting that the Allen Institute has adopted the Institute for Systems Biology’s multi-disciplinary “big science” approach. Corey, who personally buttonholed Allen after the press conference, said “It’s wonderful for Seattle.”

Paul Allen, chatting with Larry Corey, the president of the Fred Hutchinson Cancer Research Center, after the press conference.

For those who missed the story yesterday, here’s the gist. The Allen Institute will roughly double in size, from 185 people today to 350 over the next four years, to carry out a new mission that charges way beyond what it has done in its first nine years. So far, the institute has created unique, functional gene expression maps of the mouse brain, the human brain, and spinal cords. All of this is put out in the public domain, and is freely accessible to researchers at universities, biotech companies, and pharmaceutical companies. CEO Allan Jones said the institute hopes to complete its sixth human brain map by the end of this year.

The three new initiatives are hugely complex, and to my ear, sounded breathtaking in their scope and ambition. “The job gets a lot tougher now. We’re going to tackle some of the biggest challenges in science today,” Allen said. “My commitment doesn’t just continue the work of the institute—it greatly expands scope of institute. We hope to fuel more discoveries in neuroscience.”

So what does the institute plan to do with that money? Here’s a basic rundown of what the institute’s three new initiatives are about, based on the very speedy descriptions of chief scientist Christof Koch (thank goodness for digital recorders, otherwise I would never have been able to get everything he said.)

—First, the institute will seek to better understand how the brain stores, encodes, and processes information in networks. The institute plans to build what it calls “brain observatories” of the cerebral cortex in mammals. These observatories will be made up of a range of sophisticated instruments and computers to “exhaustively identify, catalog, record, intervene in the cognitive networks underlying visual perception, visual behavior, and visual consciousness in the mouse,” Koch said.

—Second, the institute will seek to create a comprehensive catalog of the types of neurons in the human and mouse cerebral cortex. “We want to create an exhaustive and comprehensive taxonomy,” Koch said. “Expanding on our past Atlas work, we will use the brain observatories to ascertain the behavior, shape, form, the connections and biochemistry of these cells.” When complete, it will be the first full picture of the brain’s cellular building blocks. “We’ll count and classify every neuron in the mouse visual system, and visualize their activity throughout the brain, from eyes to muscles in mice,” Koch said. Unlike functional MRI imaging tests of today, which look at regions of the brain that get activated during certain activities, the Allen Institute looks to drill much deeper. “Our brain observatories will be able to listen to individual neurons,” Koch said. “That’s critical because it’s the neurons that are the atoms of perception, of thought, of memory, of consciousness.”

With the emerging technology of what he called “optogenetics”—sort of like “deep-brain stimulation on steroids”—Koch said “we’ll be able to turn groups of specified neurons on or off at will, and to observe the effect of this manipulation on the behavior of the animal.” Computer models will then be used to help explain and predict what is going on in the brain, he said.

—Third, the institute will … Next Page »

Single Page Currently on Page: 1 2

By posting a comment, you agree to our terms and conditions.

  • http://www.xconomy.com/author/ltimmerman/ Luke Timmerman

    Eric Schadt, the director of Mt. Sinai School of Medicine’s Institute for Genomics and Multiscale Biology, offered a long, detailed and technical response when I asked him for comment on on the new Allen Institute initiatives. Readers should note that Schadt is quite familiar with what’s going on, as he is an advisor to the Allen Institute. Here’s what he had to say:

    Eric Schadt: “I think this is a really big deal. Diseases such as autism and schizophrenia are very complex, involving a strong genetic variance component but also environmental component. The genetic variance component is very complex though, with a predicted 500-1000 genetic loci, for example, underlying the genetic variance component of autism. Further, because these diseases involve some form of neurological dysfunction, one cannot study it at the molecular level as easy as cancer where tumor biopsies are readily available or diseases of peripheral organs in which again biopsies are more readily doable.

    One of the revolutions of the last decade has been reprogramming of cells like fibroblasts and then reprogramming them into pluripotent stem cells and then differentiating those into cell types or tissues that are relevant to the study of disease. Whether differentiating stem cells into adipocytes and beta cells to study molecular and cellular phenotypes associated with diabetes or neurons to study diseases like schizophrenia and Alzheimer’s, iPSCs generated from patients with these diseases offers a novel and exciting way to elucidate the complexity of complex human diseases. The naturally occurring genetic variation that exists within patient populations that underlie diseases like autism perturb molecular networks that underlie the pathophysiology of the disease. Therefore the isolation of cells from patients that harbor those genetic perturbations and reprogramming and differentiating those cells into disease relevant contexts, provides the opportunity to directly characterize the molecular and cellular changes that lead to pathophysiological states.

    This type of resource will be incredibly valuable to the scientific community at large because it will enable construction of the molecular networks of disease, thereby enabling an understanding of mechanisms, provide the appropriate reagents (disease relevant cell types) for drug screening, enable detection of disease biomarkers and drug response biomarkers and so on. So big enabling potential here.

    This goes well beyond what could be done in a PI driven environment because this is big science. Running at the scale of hundreds of patients and controls to isolate fibroblasts or other cell types that can be reprogrammed to iPSCs, carrying out the reprogramming, the differentiation in a reproducible way, generating all of the omics and cellular phenotype data off of that, involves a number of technical challenges that require big technology to solve, focused teams with expertises in many different areas (from robotics, to molecular biology, to informatics and computation biology), and of course considering resources (money). In addition, targeting specific genes or constellations of genes in cells to validate findings, perturb networks in relevant ways, etc., in ways that allow modulation of the gene’s activity in different contexts, again has many technical challenges in running at scale (so beyond handfuls of genes that might classically be studied in an individual lab).

    Okay, Luke, sorry to drag on, but I think this is a truly exciting project that can help transform our understanding of some of the most difficult, but prevalent, neuropsychiatric diseases, and they have recruited Ricardo [Dolmetsch], one of the more capable researchers on the planet in this area to lead this effort with the expert team assembled and growing at the Allen Institute.”

  • Lance Stewart

    Luke thank you for covering this important story.

    Collectively, we suffer from a lack of coordination and sharing of neuroinformatics data that could contribute to the greatest healthcare improvements of our lifetime. Paul G. Allen’s gift to promote open neuroscience is truly remarkable.

    There is tremendous opportunity to harness the combined investments of industry, academia, government, and patient advocacy in a focused game-changing approach to neuroscience.

    My hope is that over the next 10 years we will see a social networked approach to discovery of new preventions, diagnoses, and treatments for brain disease. Specifically, I can see a day when Pharma/biotech companies donate their chemical and device resources to academic/gov/research institution labs for full interrogation of potential utility in an open science mode of shared data. In turn, the new insights will provide enough data for smart computer systems to help guide the engineering of a new intellectual property that enables a paradigm shift in neuro-medicine. The stakes are too high for this not to happen.