Google Scientists Unveil the Biggest, Most Detailed Map of the Fly Brain Yet (hhmi.org) 43
An anonymous reader shares a summary from Howard Hughes Medical Institute: In a darkened room in Ashburn, Virginia, rows of scientists sit at computer screens displaying vivid 3-D shapes. With a click of a mouse, they spin each shape to examine it from all sides. The scientists are working inside a concrete building at the Howard Hughes Medical Institute's Janelia Research Campus, just off a street called Helix Drive. But their minds are somewhere else entirely -- inside the brain of a fly. Each shape on the scientists' screens represents part of a fruit fly neuron. These researchers and others at Janelia are tackling a goal that once seemed out of reach: outlining each of the fly brain's roughly 100,000 neurons and pinpointing the millions of places they connect. Such a wiring diagram, or connectome, reveals the complete circuitry of different brain areas and how they're linked. The work could help unlock networks involved in memory formation, for example, or neural pathways that underlie movements.
Gerry Rubin, vice president of HHMI and executive director of Janelia, has championed this project for more than a decade. It's a necessary step in understanding how the brain works, he says. When the project began, Rubin estimated that with available methods, tracing the connections between every fly neuron by hand would take 250 people working for two decades -- what he refers to as "a 5,000 person-year problem." Now, a stream of advances in imaging technology and deep-learning algorithms have yanked the dream of a fly connectome out of the clouds and into the realm of probability. High-powered customized microscopes, a team of dedicated neural proofreaders and data analysts, and a partnership with Google have sped up the process by orders of magnitude. Today, a team of Janelia researchers reports hitting a critical milestone: they've traced the path of every neuron in a portion of the female fruit fly brain they've dubbed the "hemibrain." The map encompasses 25,000 neurons -- roughly a third of the fly brain, by volume -- but its impact is outsized. It includes regions of keen interest to scientists -- those that control functions like learning, memory, smell, and navigation. With more than 20 million neural connections pinpointed so far, it's the biggest and most detailed map of the fly brain ever completed. The scientists have published a pre-print paper describing their work, and have made the data they collected available to view and download.
Gerry Rubin, vice president of HHMI and executive director of Janelia, has championed this project for more than a decade. It's a necessary step in understanding how the brain works, he says. When the project began, Rubin estimated that with available methods, tracing the connections between every fly neuron by hand would take 250 people working for two decades -- what he refers to as "a 5,000 person-year problem." Now, a stream of advances in imaging technology and deep-learning algorithms have yanked the dream of a fly connectome out of the clouds and into the realm of probability. High-powered customized microscopes, a team of dedicated neural proofreaders and data analysts, and a partnership with Google have sped up the process by orders of magnitude. Today, a team of Janelia researchers reports hitting a critical milestone: they've traced the path of every neuron in a portion of the female fruit fly brain they've dubbed the "hemibrain." The map encompasses 25,000 neurons -- roughly a third of the fly brain, by volume -- but its impact is outsized. It includes regions of keen interest to scientists -- those that control functions like learning, memory, smell, and navigation. With more than 20 million neural connections pinpointed so far, it's the biggest and most detailed map of the fly brain ever completed. The scientists have published a pre-print paper describing their work, and have made the data they collected available to view and download.
So they mapped inside BeauHD's cranium? (Score:2, Insightful)
Wishful thinking!
Re: (Score:3, Insightful)
Re: (Score:2)
Give credit where credit is due.
"Google scientists" is an oxymoron. (Score:1, Interesting)
There are no "scientists" at Google, there is a bunch of code monkeys, who try to get answers by stuffing garbage data collected from scraping web pages into an "AI" and returning the garbage output as "science".
Usually failing.
Remember when the "Google scientist" proclaimed "quantum supremacy"?
Remember how it was laughed out of the adult room? https://www.newscientist.com/a... [newscientist.com]
Future AI research might or might not benefit. (Score:4, Interesting)
Emulating natural systems that have evolved over millions of years does not necessarily translate well into binary based algorithms that then can create useful functions in terms of learning. The reasons are that the chemical switching basis of evolution in living organisms is as important as the electrical nervous signalling. The two work together or they often conflict. The conflict between chemical signals and electrical signals is frequently pain stimulus.
Purely digital electrical sensing technology is far too simple for the creation of a system which can learn. Until we can develop analogues of systems with feedback and encoding capabilities even simpler than that which goes on in a brain as simple as one of a fly we will remain only in the beginning stages of creating artificial life.
Re: (Score:3)
"..It's a necessary step in understanding how the brain works.."
Re: Future AI research might or might not benefit. (Score:1)
Re: (Score:1)
Limits of our current abilities are implicitly acknowledged in the following quote:
"..It's a necessary step in understanding how the brain works.."
No question of that assessment. My suggestion is that to create an analogue of a chemical process in terms of a binary algorithm is also a necessary first step that we have not achieved. That is why I used the term emulation in my first statement. To create an analogue of a brain, no matter how simple it is will require software artificial constructs of very advanced chemical processes as well as the switching to control the process. The problem is that the process goes both ways and has feedback loops that
Re: (Score:2)
Emulating natural systems that have evolved over millions of years does not necessarily translate well into binary based algorithms that then can create useful functions in terms of learning. The reasons are that the chemical switching basis of evolution in living organisms is as important as the electrical nervous signalling. The two work together or they often conflict. The conflict between chemical signals and electrical signals is frequently pain stimulus.
Except students pouring over books seem to learn lots of things without any chemical input of note. It's true that the logical part of our brain interact a lot with the reward centers that release chemicals, but that's probably more about our drive and desires than the intelligence to execute on it.
Hmmm.... (Score:1, Offtopic)
Considering all the quantum activity we're finding in biology do we really think the mind is local?
Seems to me this is studying an antenna. We ought be looking for the signal.
Re: (Score:3)
Re: (Score:2)
That is the best shot we have to understand what kind of signal we are supposed to look for, at least for now.
Smarter than Senate (Score:1, Insightful)
Re: (Score:2)
Maybe if they actually had some, that wouldn't be an issue.
Re: (Score:1, Insightful)
If there's no evidence, why is McConnell fighting so hard to stop them presenting it?
A map/connectome is only the beginning (Score:3)
The complete connectome of the nematode C. Elegans was mapped by hand long ago. And it has less than 1000 neurons. But that does not tell you how it thinks.
The fly is much more complex. C. Elegans is a more realistic first step.
The problem is that the synapses that join neurons are complex and different. And the brain is part of the whole organism, and does not operate in a vacuum.
So nice to start mapping the Fly. But do not think that that of itself leads to understanding how it works.
Re: (Score:2)
So nice to start mapping the Fly. But do not think that that of itself leads to understanding how it works.
Don’t worry. Google’s scientists have already announced their next project - documenting the sound of one hand clapping. Their initial abstract states they expect it to be a very faint whooshing noise, like a non-aerodynamic object slowly moving through air.
Re: (Score:2)
It has actually much less. In total 302, 20 of them building a second mini brain.
ooops (Score:1)
They meant to map the human brain, but misinterpreted the order, "let it fly!"
Re: (Score:1)
Finally we can figure out Trump
No that would require a full mapping of the brains in all subspecies of Pulmonata not that of a single member of the Musca genus.
Re: Eurinreeka! (Score:2)
Interviews (Score:2)
I guess they're intending to make their interviews even more invasive than they have been.
Seeing the CURRENT connections ... (Score:3)
Seeing the CURRENT connections probably isn't enough.
One thing that has been noticed - at least in more advanced brains than flies - is that, as the brain is operating, the cell body and dendrites put out little fibers, smaller than the dendrites, that are (usually) reabsorbed shortly thereafter.
One speculation is that these are involved in learning, perhaps by, under the correct circumstances, connecting to surrounding axons and leading them to form new synapses. (e.g. they might do things like detect combinations of nearby nerves that fire shortly before their parent neuron fires and encourage the formation of a "logic gate" that speeds up the computation. There are lots of other possibilites, of course.)
If such selective addition of connections (and, no doubt, other mechanisms to delete others, rather than building a block of nerves that just spasms in an epileptic fit) is a basis of learning, reproducing a snapshot of the current logic gates would miss the core of the functionality.
And now they know.. (Score:2)
The last thing that goes through a Flyâ(TM)s mind right before it hits the windshield of your car...
His ass.
Map? (Score:2)
Without a GPS?
I guess they used the FB AI from the other article on this page.
Surprisingly complex... (Score:1)