Slashdot Mirror


A Nanoscale Look At a Complete Fly Brain (cemag.us)

An anonymous reader quotes a report from Controlled Environments Magazine: Two high-speed electron microscopes. 7,062 brain slices. 21 million images. For a team of scientists at the Howard Hughes Medical Institute's Janelia Research Campus in Ashburn, Virginia, these numbers add up to a technical first: a high-resolution digital snapshot of the adult fruit fly brain. Researchers can now trace the path of any one neuron to any other neuron throughout the whole brain, says neuroscientist Davi Bock, a group leader at Janelia who reported the work along with his colleagues on July 19 in the journal Cell.

The fruit fly brain, roughly the size of a poppy seed, contains about 100,000 neurons (humans have 100 billion). Each neuron branches into a starburst of fine wires that touch the wires of other neurons. Neurons talk to one another through these touchpoints, or synapses, forming a dense mesh of communication circuits. Scientists can view these wires and synapses with an imaging technique called serial section transmission electron microscopy. First, they infuse the fly's brain with a cocktail of heavy metals. These metals pack into cell membranes and synapses, ultimately marking the outlines of each neuron and its connections. Then the researchers hit slices of the brain with a beam of electrons, which passes through everything except the metal-loaded parts. "It's the same way that x-rays go through your body except where they hit bone," Bock explains. The resulting images expose the brain's once-hidden nooks and crannies.

14 of 73 comments (clear)

  1. 100k neurons only by hcs_$reboot · · Score: 4, Insightful

    And yet the fly adapts to so many different situations, flies,"eats", copulates etc.... that emphasizes how powerful the 3D brain structure is, and how our current AI is not.

    --
    Slashdot, fix the reply notifications... You won't get away with it...
    1. Re:100k neurons only by Anonymous Coward · · Score: 4, Funny

      As smart as they are, they can't seem to adapt to the flypaper in my house ;)

    2. Re:100k neurons only by hcs_$reboot · · Score: 2, Insightful

      This is more of natural selection... genetics takes time!

      --
      Slashdot, fix the reply notifications... You won't get away with it...
    3. Re:100k neurons only by ShanghaiBill · · Score: 4, Insightful

      This is more of natural selection... genetics takes time!

      If an organism relies on genetics to solve every novel problem, then it is stupid.

    4. Re:100k neurons only by ShanghaiBill · · Score: 2

      Do you think Darwin Award winners are smart?

      Evolution is just brute force trial and error with 99% collateral damage. There is nothing "intelligent" about it. The individuals unable to adapt to their environment without being Darwined out of existence are not intelligent either.

  2. Tiny worm C. Elegans is still a mystery by aberglas · · Score: 5, Insightful

    It has about 300 neurons, 1000 cells all up (so about a third are neurons). Its connectome (wiring diagram) has been known for decades, and unlike human brains is identical in each worm. But how it actually thinks remains a mystery.

    So good work to understand a fruit fly, and no doubt useful. But do not mistake it for understanding.

    While understanding neurons might be helpful for building an AI, I think it is unlikely that an AI will be any direct mapping. Aeroplanes are not built out of feathers.

    1. Re:Tiny worm C. Elegans is still a mystery by ShanghaiBill · · Score: 4, Interesting

      Much progress has been made at understanding C. elegans. The OpenWorm project has simulated the neuron to muscle pathway, and they estimate they are about 30% of the way to a complete simulation.

      Simulating a fly brain will be a big step beyond that.

    2. Re:Tiny worm C. Elegans is still a mystery by raftpeople · · Score: 4, Informative

      Unless the OpenWorm project is also simulating the astrocytes that modulate and manage synapse activity, it's probably not going to be successful. For further info, google astrocyte tripartite synapse and you will find that neuroscientists now consider the astrocyte to be managing/controlling the synapse, (responding to neurotransmitters and signalling neurons on either side of the synapse as well as other glial cells).

    3. Re:Tiny worm C. Elegans is still a mystery by ShanghaiBill · · Score: 2, Interesting

      Unless the OpenWorm project is also simulating the astrocytes ...

      They are. OpenWorm is a project to digitally simulate the entire organism at the biochemical level. All the cells. Every chemical pathway.

      Worms first. Then flies. Then humans.

    4. Re:Tiny worm C. Elegans is still a mystery by sheramil · · Score: 2

      But how it actually thinks remains a mystery.

      How much time do you think C. elegans spends contemplating "does existence precede essence?"

    5. Re:Tiny worm C. Elegans is still a mystery by raftpeople · · Score: 3, Insightful

      I'm very aware of the OpenWorm project and I think it's a great idea to start with the smallest brain rather than others that think they can jump right into simulating a rat brain or a human brain.

      Having said that, they will first need to figure out how the astrocytes performs their computations and neuroscientists are absolutely nowhere near that level of information. In addition, even in the neuron, DNA is involved in computations (synapse strength is altered by turning on/off genes via epigenetic mechanisms and those alternations are what sustain the synapse at the current strength). So, OpenWorm idea is good but they are a looooong way off from having enough info to simulate the computations that happen.

    6. Re:Tiny worm C. Elegans is still a mystery by rahvin112 · · Score: 2

      We won't be successful with a real AI until we return to analog computing. The brain in biological organisms is analog and operates roughly equivalent to a pokey 600mhz. The difference is the 100 billion connections with upwards of dozens of routing paths per neuron and analog communication. We are a long way from being able to create something similar. Although we wouldn't need the full 100 billion because much of the brains computing power is devoted not to inteligence but simply managing living processes it still going to be a very long time before we will ever get close to this.

  3. I don't understand, where's the tea? by Thud457 · · Score: 2

    You'd be programmed to not notice the difference.

    --

    the preceding comment is my own and in no way reflects the opinion of the Joint Chiefs of Staff

  4. Re:the Crazy one by CODiNE · · Score: 3, Funny

    *Looks at label on the brain jar*
    "Abby Normal... what a nice name for a fruit fly. I'll use this brain."

    --
    Cwm, fjord-bank glyphs vext quiz