Will we ever… simulate the human brain?

A billion dollar project claims it will recreate the most complex organ in the human body in just 10 years. But detractors say it is impossible. Who is right?

For years, Henry Markram has claimed that he can simulate the human brain in a computer within a decade. On 23 January 2013, the European Commission told him to prove it. His ambitious Human Brain Project (HBP) won one of two ceiling-shattering grants from the EC to the tune of a billion euros, ending a two-year contest against several other grandiose projects. Can he now deliver? Is it even possible to build a computer simulation of the most powerful computer in the world – the 1.4-kg (3 lb) cluster of 86 billion neurons that sits inside our skulls?

The very idea has many neuroscientists in an uproar, and the HBP’s substantial budget, awarded at a tumultuous time for research funding, is not helping. The common refrain is that the brain is just too complicated to simulate, and our understanding of it is at too primordial a stage.

Then, there’s Markram’s strategy. Neuroscientists have built computer simulations of neurons since the 1950s, but the vast majority treat these cells as single abstract points. Markram says he wants to build the cells as they are – gloriously detailed branching networks, full of active genes and electrical activity. He wants to simulate them down to their ion channels – the molecular gates that allow neurons to build up a voltage by shuttling charged particles in and out of their membrane borders. He wants to represent the genes that switch on and off inside them. He wants to simulate the 3,000 or so synapses that allow neurons to communicate with their neighbours.

Erin McKiernan, who builds computer models of single neurons, is a fan of this bottom-up approach. “Really understanding what’s happening at a fundamental level and building up – I generally agree with that,” she says. “But I tend to disagree with the time frame. [Markram] said that in 10 years, we could have a fully simulated brain, but I don’t think that’ll happen.”

Even building McKiernan’s single-neuron models is a fiendishly complicated task. “For many neurons, we don’t understand well the complement of ion channels within them, how they work together to produce electrical activity, how they change over development or injury,” she says. “At the next level, we have even less knowledge about how these cells connect, or how they’re constantly reaching out, retracting or changing their strength.” It’s ignorance all the way down.

“For sure, what we have is a tiny, tiny fraction of what we need,” says Markram. Worse still, experimentally mapping out every molecule, cell and connection is completely unfeasible in terms of cost, technical requirements and motivation. But he argues that building a unified model is the only way to unite our knowledge, and to start filling in the gaps in a focused way. By putting it all together, we can use what we know to predict what we don’t, and to refine everything on the fly as new insights come in.

Continue reading

Why we’re building a €1 billion model of a human brain

We want to reach a unified understanding of the brain and the simulation on a supercomputer is the tool. Today you have neuroscientists working on a genetic, behavioural or cognitive level, and then you have informaticians, chemists and mathematicians. They all have their own understanding of how the brain functions and is structured. How do you get them all around the same table? We think of the project as like a CERN for the brain. The model is our way of bringing everyone, and our understanding, together.

Putting Our Heads Together: Canines May Hold Clues to Human Skull Development

Man’s best friend may touch our hearts with their empathy, companionship, playfulness and loyalty, and they may also lead us to a deeper understanding of our heads.

In the article, “The Genetics of Canine Skull Shape Variation,” in the February issue of the Genetics Society of America’s journal, GENETICS, Jeffrey J. Schoenebeck, PhD, and Elaine A. Ostrander, PhD, researchers at the National Human Genome Research Institute (NHGRI), the National Institutes of Health (NIH), review progress in defining the genes and pathways that determine canine skull shape and development that have been made in the eight years since the dog genome was mapped.

The implications of this research extend beyond the interests of dog fanciers and breeders. “Dogs can serve as a model for skull growth and shape determination because the genetic conservation between dogs and humans makes it highly likely that craniofacial development is regulated similarly between both species,” Dr. Schoenebeck said. “These discoveries are important for human health and biology, especially for children born with craniofacial deformities,” Dr. Ostrander, added. In humans these deformities include Apert, Crouzon and Pfeiffer syndromes, where skull bones fuse prematurely causing facial malformations, such as wide-set bulging eyes and broad foreheads, resulting in dental, eye and other physiological problems.

Skull shape is a complex trait, involving multiple genes and their interactions. Thanks to standardized canine breeding, which documents more than 400 breeds worldwide, and their distinct morphological features, researchers can disentangle traits such as skull shape, which in many breeds is a breed-defining variation.

For example, researchers are beginning to identify which genes cause a Bulldog or a Pug to have short pushed-in faces, or brachycephaly, and those that cause Saluki’s or collies to have narrow, elongated snouts, or dolichocephaly. Between these two distinct canine cranium shapes are many variations that are also breed specific but can’t be neatly categorized as brachycephalic or dolichocephalic, such as the rounded skull of the Chihuahua or the downward pointing snout of the Bull terrier. Researchers now use genome-wide association studies (GWAS) to identify loci of interest that may be associated with these kinds of subtle differences.

The use of GWAS in determining genetic variation in dogs is in its infancy. What’s exciting said Dr. Schoenebeck is that with these studies and the tools researchers now have to map these variations “we may find new roles for genes, never before implicated in cranium development” and because similar genes and genetic pathways operate in humans, unexplained craniofacial developmental defects may become better understood.

Identifying the causative genetic mechanisms of these variations in canines offer researchers who study human cranial abnormalities “a way to figure out what sort of genetic variation matters and what doesn’t,” said Dr. Ostrander.

Drs. Schoenbeck and Ostrander clearly show there’s a lot more research to do on craniofacial development in dogs. It is also clear that the connection between us and our canine friends is in our heads as well as our hearts.

(Image: Villemarette)

Peering into living cells — without dye nor fluophore

In the world of microscopy, this advance is almost comparable to the leap from photography to live television. Two young EPFL researchers, Yann Cotte and Fatih Toy, have designed a device that combines holographic microscopy and computational image processing to observe living biological tissues at the nanoscale. Their research is being done under the supervision of Christian Depeursinge, head of the Microvision and Microdiagnostics Group in EPFL’s School of Engineering.

Using their setup, three-dimensional images of living cells can be obtained in just a few minutes – instantaneous operation is still in the works – at an incredibly precise resolution of less than 100 nanometers, 1000 times smaller than the diameter of a human hair. And because they’re able to do this without using contrast dyes or fluorescents, the experimental results don’t run the risk of being distorted by the presence of foreign substances.

Being able to capture a living cell from every angle like this lays the groundwork for a whole new field of investigation. “We can observe in real time the reaction of a cell that is subjected to any kind of stimulus,” explains Cotte. “This opens up all kinds of new opportunities, such as studying the effects of pharmaceutical substances at the scale of the individual cell, for example.”

Watching a neuron grow

This month in Nature Photonics the researchers demonstrate the potential of their method by developing, image by image, the film of a growing neuron and the birth of a synapse, caught over the course of an hour at a rate of one image per minute. This work, which was carried out in collaboration with the Neuroenergetics and cellular dynamics laboratory in EPFL’s Brain Mind Institute, directed by Pierre Magistretti, earned them an editorial in the prestigious journal. “Because we used a low-intensity laser, the influence of the light or heat on the cell is minimal,” continues Cotte. “Our technique thus allows us to observe a cell while still keeping it alive for a long period of time.”

As the laser scans the sample, numerous images extracted by holography are captured by a digital camera, assembled by a computer and “deconvoluted” in order to eliminate noise. To develop their algorithm, the young scientists designed and built a “calibration” system in the school’s clean rooms (CMI) using a thin layer of aluminum that they pierced with 70nm-diameter “nanoholes” spaced 70nm apart.

Finally, the assembled three-dimensional image of the cell, that looks as focused as a drawing in an encyclopedia, can be virtually “sliced” to expose its internal elements, such as the nucleus, genetic material and organelles.

Toy and Cotte, who have already obtained an EPFL Innogrant, have no intention of calling a halt to their research after such a promising beginning. In a company that’s in the process of being created and in collaboration with the startup Lyncée SA, they hope to develop a system that could deliver these kinds of observations in vivo, without the need for removing tissue, using portable devices. In parallel, they will continue to design laboratory material based on these principles. Even before its official launch, the start-up they’re creating has plenty of work to do - and plenty of ambition, as well.

My gray matter might be waning. Then again, it might not be. But I swear that I can feel memories — as I’m making them — slide off a neuron and into a tangle of plaque. I steel myself for those moments to come when I won’t remember what just went into my head.

I’m not losing track of my car keys, which is pretty standard in aging minds. Nor have I ever forgotten to turn off the oven after use, common in menopausal women. I can always find my car in the parking lot, although lots of “normal” folk can’t.

Rather, I suddenly can’t remember the name of someone with whom I’ve worked for years. I cover by saying “sir” or “madam” like the Southerner I am, even though I live in Vermont and grown people here don’t use such terms. Better to think I’m quirky than losing my faculties. Sometimes I’ll send myself an e-mail to-do reminder and then, seconds later, find myself thrilled to see a new entry pop into my inbox. Oops, it’s from me. Worse yet, a massage therapist kicked me out of her practice for missing three appointments. I didn’t recall making any of them. There must another Nancy.

Am I losing track of me?

Waiting for the Forgetting to Begin by Nancy Stearns Bercaw

This Is Your Brain On Movies: Neuroscientists Weigh In On The Brain Science of Cinema

In movies, we explore landscapes far removed from our day-to-day lives. Whether experiencing the fantastical adventures of Star Wars or the dramatic throes of The English Patient, movies demand that our brains engage in a complex firing of neurons and cognitive processes. We enter into manipulated worlds where musical scores enhance feeling; where cinematography clues us into details we’d normally gloss over; where, like omniscient beings, we voyeuristically peek into others’ lives and minds; and where we can travel from Marrakech to Mars without ever having left our seat. Movies reflect reality, yet are anything but.

“Movies are highly complex, multidimensional stimuli,” said Uri Hasson, a neuroscientist and psychologist at Princeton University. “Some areas of the brain analyze sound bites, some analyze word context, some the sentence content, music, emotional aspect, color or motion.” Just as many people must come together to work on different elements of a movie’s script, score, visuals or costumes, he explained, so many areas of the brain must also be engaged in processing those disparate elements.

The relatively new field of neurocinematic studies seeks to untangle our neurological experience of film and, in doing so, learn not only the mechanisms behind movie watching but also how movies might teach us more about ourselves.

Continue reading

Sensing the light, but not to see: Study offers insight on the evolution of photosensitive cells

In a primitive marine organism, MBL scientists find photosensitive cells that may be ancestral to the “circadian receptors” in the mammalian retina.

Among the animals that are appealing “cover models” for scientific journals, lancelets don’t spring readily to mind. Slender, limbless, primitive blobs that look pretty much the same end to end, lancelets “are extremely boring. I wouldn’t recommend them for a home aquarium,” says Enrico Nasi, adjunct senior scientist at the Marine Biological Laboratory (MBL). Yet Nasi and his collaborators managed to land a lancelet on the cover of the Journal of Neuroscience last December. These simple chordates, they discovered, offer insight into our own biological clocks.

Nasi and his wife, MBL adjunct scientist Maria del Pilar Gomez, are interested in phototransduction, the conversion of light by light-sensitive cells into electrical signals that are sent to the brain. The lancelet, also called amphioxus, doesn’t have eyes or a true brain. But what it does have in surprising abundance is melanopsin, a photopigment that is also produced by the third class of light-sensitive cells in the mammalian retina, besides the rods and cones. This third class of cells, called “intrinsically photosensitive retinal ganglion cells” (ipRGCs), were discovered in 2002 by Brown University’s David Berson and colleagues. Now sometimes called “circadian receptors,” they are involved in non-visual, light-dependent functions, such as adjustment of the animal’s circadian rhythms.

"It seemed like colossal overkill that amphioxus have melanopsin-producing cells," Nasi says. "These animals do nothing. If you switch on a light, they dance and float to the top of the tank, and then they drop back down to the bottom. That’s it for the day." But that mystery aside, Gomez and Nasi realized that studying amphioxus could help reveal the evolutionary history of the circadian receptors.

If you give a bioengineer a cookie…

“When you grab a cookie and want to break off a piece with a chocolate chip,” says Maurice Smith, balancing a crumbly bit between two of his fingers, “your brain must represent that action plan extrinsically, as it is an activity based in the world.”

The cookies are on hand to celebrate the bioengineer’s birthday in his lab at 60 Oxford Street, a white squat building located on the northernmost edge of the Harvard campus. A half moon of chocolate cake with a line of colored candles still intact also sits nearby.

Gesticulating with the cookie, Smith, Associate Professor of Bioengineering at the Harvard School of Engineering and Applied Sciences (SEAS), further teases out the intricacies of motor memory.

“An intrinsic representation is one that’s body-based and procedural. It relates to the complex series of muscle and joint movements your body has to make to complete a task,” Smith says. 

“When I first had the thought to grab the cookie and rip off a chunk with a chocolate chip, my body responded appropriately,” he notes.

Understanding the way the brain represents extrinsic and intrinsic actions, and the relationship between the two, has been of great interest to researchers who seek to understand motor control and motor learning—or, put simply, how we learn to move. 

Just a few months ago, Smith and his colleagues in the Neuromotor Control Lab laid out a generalizable theory about how the brain encodes such motor memories. Writing in the Journal of Neuroscience, they showed that units of motor memory are not so binary after all, but instead a mixture of both the intrinsic and the extrinsic.

“There’s no question that our actions are inherently spatial, but the nature of the coordinate frame used in motor memory to represent space for action planning has been hotly debated,” explains Smith. “The predominant idea had been that in memory we maintain separate intrinsic and extrinsic representations of action and translate between the two when necessary. But our work shows that memory representations are combinatorial rather than separate.” 

Individual neurons in several different motor areas of the brain encode multiplicative combinations of intrinsic and extrinsic representations, a property that neurophysiologists have called gain-field encoding. This much was known before, but it was thought that gain-field encoding simply provided a way to translate between intrinsic and extrinsic representations.

“We found that this gain-field encoding, which leads to a combinatorial representation of space, is not simply an intermediary in the transformation between representations, but is in fact the encoding on which motor memories are based,” says Smith. “This suggests that the neurons which display gain-field encoding are the same ones that store the motor memories associated with the actions we learn.”

Secrets of lasting love are hidden inside the brain