5 ms·

Neuroscientists decode the brain activity of the worm

100 points by henriquemaia 11y ago

Animats 11y ago
"They used worms genetically engineered with a fluorescent protein that caused the worm's neurons to flash when they were active." Now that's impressive. So can OpenWorm, the C. Elegans simulator, get the same patterns?
- etrautmann 11y ago
  I'm currently working on extending this technique to higher order primates. It's unbelievably cool that these protein sensors work. If you're curious about the details, check out GCaMP and calcium imaging, or these papers (paywall) https://en.wikipedia.org/wiki/Calcium_imaging https://en.wikipedia.org/wiki/Calcium_imaging http://www.nature.com/nature/journal/v499/n7458/abs/nature12354.html http://www.nature.com/nature/journal/v499/n7458/abs/nature12... http://www.jneurosci.org/content/32/40/13819.full http://www.jneurosci.org/content/32/40/13819.full
tux1968 11y ago
It sounded as if they have mapped all of the connections between neurons in previous work. Which suggests these connections are genetic rather than learned, and the same in every individual worm. While that makes studying them much easier, I find it surprising. I suppose that's how innate abilities are conferred on the next generation even in humans where many connections will indeed be individually learned/constructed. Very interesting.
- spin 11y ago
  Yes. The worm c.elegans has 302 neurons, connected in a specific way. From the article: "For almost 30 years, the list of connections between individual neurons has been known. Despite the low number of neurons, its neuronal networks possesse a high degree of complexity and sophisticated behavioral output; the worm thus represents an animal of choice to study brain function."
  - skybrian 11y ago
    Apparently they just found two more in males: http://www.nature.com/news/surprise-mystery-neurons-found-in-male-worms-1.18558 http://www.nature.com/news/surprise-mystery-neurons-found-in...
- dekhn 11y ago
  Yeah, C. elegans is eutelic (every instance has identical set of cells in same location) and after https://en.wikipedia.org/wiki/John_Sulston https://en.wikipedia.org/wiki/John_Sulston determined the lineage fate, scientists subsequently determined the full connectome. C. elegans is the model organism of model organisms- a simple animal with many of the properties of humans but unbelievable easy to work with in the lab and it has a long list of properties that make it really awesome for mechanistic studies.
iandanforth 11y ago
Full paper: http://www.cell.com/cell/pdfExtended/S0092-8674(15)01196-4 http://www.cell.com/cell/pdfExtended/S0092-8674(15)01196-4
dekhn 11y ago
Interestingly, C. Elegans also autofluoresces- without the need for any special genes to be cloned in. When it dies, a bright wave of light passes through its body.
- etrautmann 11y ago
  Lots of proteins have natural fluorescence. You'll notice this if you shine blacklight in ocean water - it'll glow brightly from all of the little microorganisms that aren't intending to fluoresce but do so incidentally. There are also bioluminescent organisms but that's different.
  - dekhn 11y ago
    Sure, IIRC from my college chemistry classes it's mostly Trp or Try fluorescence. In this case, it's still considered fairly puzzling why the flash is so intense or what its "actual function" or whether it's protein fluorescence at all: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735983/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735983/
ChuckMcM 11y ago
Ok, that is pretty cool. Next up, inducing neural activity in worms to drive them around. If you can go both ways (observe and predict, and induce and demonstrate) you really do have the worm's brain all figured out. Then you build a mining robot with mineral sensors that can detect the ore your interested in. Now you strap in your worm brain and you have your local computer translate motion signals into robot motion, and mineral detection into food detection. And voila you have a robotic miner which forages around for ore to process.
- xuhu 11y ago
  Sure Brain, but how much budget will there be for spare parts when these robot worms start producing mating signals in those neurons ? Because they will.
- dekhn 11y ago
  Driving mice is already possible. https://med.stanford.edu/news/all-news/2010/09/light-workout-scientists-use-optogenetics-to-effectively-stimulate-muscle-movement-in-mice.html https://med.stanford.edu/news/all-news/2010/09/light-workout... I believe this was also done in insects, although I can't find the article. With C. Elegans, the difference is that we will be able to understand it in far more detail. As for mining robots, biology already has you beat. The entire earth in enervated with slow-growing mining bacteria. They are probably a better place to start for industrial purposes.
Aardwolf 11y ago
Is it possible to access the data of how the 302 cells are connected somewhere? Would be awesome to emulate in software.
- girvo 11y ago
  Do you mean like this: http://www.openworm.org http://www.openworm.org ?
- shpx 11y ago
  This might be what you are looking for http://www.wormatlas.org/neuronalwiring.html http://www.wormatlas.org/neuronalwiring.html specifically this file http://www.wormatlas.org/images/NeuronConnect.xls http://www.wormatlas.org/images/NeuronConnect.xls that you can read in pandas

100 points by henriquemaia 11y ago

"They used worms genetically engineered with a fluorescent protein that caused the worm's neurons to flash when they were active." Now that's impressive. So can OpenWorm, the C. Elegans simulator, get the same patterns?

etrautmann 11y ago
I'm currently working on extending this technique to higher order primates. It's unbelievably cool that these protein sensors work. If you're curious about the details, check out GCaMP and calcium imaging, or these papers (paywall) https://en.wikipedia.org/wiki/Calcium_imaging https://en.wikipedia.org/wiki/Calcium_imaging http://www.nature.com/nature/journal/v499/n7458/abs/nature12354.html http://www.nature.com/nature/journal/v499/n7458/abs/nature12... http://www.jneurosci.org/content/32/40/13819.full http://www.jneurosci.org/content/32/40/13819.full

tux1968 11y ago

It sounded as if they have mapped all of the connections between neurons in previous work. Which suggests these connections are genetic rather than learned, and the same in every individual worm. While that makes studying them much easier, I find it surprising. I suppose that's how innate abilities are conferred on the next generation even in humans where many connections will indeed be individually learned/constructed. Very interesting.

spin 11y ago
Yes. The worm c.elegans has 302 neurons, connected in a specific way. From the article: "For almost 30 years, the list of connections between individual neurons has been known. Despite the low number of neurons, its neuronal networks possesse a high degree of complexity and sophisticated behavioral output; the worm thus represents an animal of choice to study brain function."
- skybrian 11y ago
  Apparently they just found two more in males: http://www.nature.com/news/surprise-mystery-neurons-found-in-male-worms-1.18558 http://www.nature.com/news/surprise-mystery-neurons-found-in...
dekhn 11y ago
Yeah, C. elegans is eutelic (every instance has identical set of cells in same location) and after https://en.wikipedia.org/wiki/John_Sulston https://en.wikipedia.org/wiki/John_Sulston determined the lineage fate, scientists subsequently determined the full connectome. C. elegans is the model organism of model organisms- a simple animal with many of the properties of humans but unbelievable easy to work with in the lab and it has a long list of properties that make it really awesome for mechanistic studies.

iandanforth 11y ago

Full paper: http://www.cell.com/cell/pdfExtended/S0092-8674(15)01196-4 http://www.cell.com/cell/pdfExtended/S0092-8674(15)01196-4

dekhn 11y ago

Interestingly, C. Elegans also autofluoresces- without the need for any special genes to be cloned in. When it dies, a bright wave of light passes through its body.

etrautmann 11y ago
Lots of proteins have natural fluorescence. You'll notice this if you shine blacklight in ocean water - it'll glow brightly from all of the little microorganisms that aren't intending to fluoresce but do so incidentally. There are also bioluminescent organisms but that's different.
- dekhn 11y ago
  Sure, IIRC from my college chemistry classes it's mostly Trp or Try fluorescence. In this case, it's still considered fairly puzzling why the flash is so intense or what its "actual function" or whether it's protein fluorescence at all: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735983/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735983/

ChuckMcM 11y ago

Ok, that is pretty cool. Next up, inducing neural activity in worms to drive them around. If you can go both ways (observe and predict, and induce and demonstrate) you really do have the worm's brain all figured out. Then you build a mining robot with mineral sensors that can detect the ore your interested in. Now you strap in your worm brain and you have your local computer translate motion signals into robot motion, and mineral detection into food detection. And voila you have a robotic miner which forages around for ore to process.

xuhu 11y ago
Sure Brain, but how much budget will there be for spare parts when these robot worms start producing mating signals in those neurons ? Because they will.
dekhn 11y ago
Driving mice is already possible. https://med.stanford.edu/news/all-news/2010/09/light-workout-scientists-use-optogenetics-to-effectively-stimulate-muscle-movement-in-mice.html https://med.stanford.edu/news/all-news/2010/09/light-workout... I believe this was also done in insects, although I can't find the article. With C. Elegans, the difference is that we will be able to understand it in far more detail. As for mining robots, biology already has you beat. The entire earth in enervated with slow-growing mining bacteria. They are probably a better place to start for industrial purposes.

Aardwolf 11y ago

Is it possible to access the data of how the 302 cells are connected somewhere? Would be awesome to emulate in software.

girvo 11y ago
Do you mean like this: http://www.openworm.org http://www.openworm.org ?
shpx 11y ago
This might be what you are looking for http://www.wormatlas.org/neuronalwiring.html http://www.wormatlas.org/neuronalwiring.html specifically this file http://www.wormatlas.org/images/NeuronConnect.xls http://www.wormatlas.org/images/NeuronConnect.xls that you can read in pandas