December 1, 2014

Not Very Uplifting

What responsibility do we have for the things we make?

At its root, this is a fairly straightforward science story. Neuroscience researchers at the University of Rochester and the University of Copenhagen successfully transplanted human glial progenitor cells (hGPCs) into a newborn mouse (here's the technical article in The Journal of Neuroscience, and the lay-friendly version in New Scientist). While glial cells are generally considered a support cell in the brain, positioning, feeding, insulating, and protecting neurons, they also help neurons make synaptic connections. The hGPCs out-competed the mouse glial cells, basically taking over that function in the mouse brain, and -- as had been found in similar research (with adult glial cells) -- the mice demonstrated greater intelligence than their unaltered fellows.

So, mice with grafted human brain support cells are smarter than regular mice. The next phase is testing with rats, which start out even smarter. The researchers insist that there's nothing especially human about these altered mice:

"This does not provide the animals with additional capabilities that could in any way be ascribed or perceived as specifically human," he says. "Rather, the human cells are simply improving the efficiency of the mouse's own neural networks. It's still a mouse."

However, the team decided not to try putting human cells into monkeys. "We briefly considered it but decided not to because of all the potential ethical issues," Goldman says.

(...A statement that somewhat undermines his whole "it's still a mouse" argument -- after all, wouldn't it still be a monkey?)

As always, I'm mostly interested in the "what happens next?" question. It's likely that rats with hGPC will show increased intelligence; same with dogs. And just because this set of researchers won't add the hGPC special sauce to monkeys doesn't mean that somebody else won't do it. And maybe even throw in a few neuron precursors for flavor.

But even sticking with hGPCs, the fact remains: we're making these non-human animals demonstrably smarter. We are, in a very limited fashion, uplifting them (to use David Brin's terminology). They will be able to understand the world a bit (or even a lot) better than others of their kind. And at some point, we may well even end up with test subjects significantly smarter than typical and able to demonstrate behaviors unsettlingly close to our own.

What rights should any of these types of uplifted animals have? Do we need to spell out a greater set of rights for the human chimera mice in the news report? Or as we create increasingly more-intelligent-than-typical animals, will there a point at which they could no longer be limited to the rights given to all scientific research animals? At what point would it become a crime to kill them, no matter how humanely or in accordance with ethical standards? It would be easy to draw the line if the uplifted animals exhibit human-like behavior -- complex communication, for example, or the creation of art -- but what about intelligence-boosted animals that exhibit forms of higher intelligence that don't readily map to human-specific behavior but are clearly beyond what a typical animal of that species could do? When do we give them a say in their own lives?

This connects in fairly obvious ways to the ongoing efforts to provide more expansive rights to the Great Apes or Cetaceans, but it's equally an issue for the Magna Cortica project. What it's not is a science fiction question for our distant descendants. This is happening now, and these issues need to be addressed now.

May 22, 2010

Synthetic Genome: TL;DR

If you already understand what's happening with the Venter Institute synthetic genome announcement, and just want to see my response, here's the money quote from the end of the previous post:

One suggestion that we know is possible, because a variation appeared in the Venter announcement: all synthetic genomes should be signed. According to Wired:

“They rebuilt a natural sequence and they put in some poetry,” said University of California at San Francisco synthetic biologist Chris Voigt. “They recreated some quotes in the genome sequence as watermarks.”

What Voigt refers to as a "watermark" should instead be thought of as a "DNA signature." We should require that all synthetic genomes include something like this, unique sequences following a designated pattern, identifying the organization behind the genome, the lab responsible, the date, and any other useful bits of information. Multiple copies should appear throughout the synthetic genome, so it doesn't get mutated away.

That way, if something unexpected happens, we know whom to talk to.

Give My Creation... Life!

The Venter Institute announcement that it had successfully crafted the first self-replicating synthetic organism caused quite a stir, even among people who are otherwise pretty jaded about emerging tech.

It's useful to understand exactly what is -- and what isn't -- going on here.

Where we are:

  • Synthetic genome copied from natural genome and transplanted into existing cell structure.

    This is a moderately big deal, but only that; it's a stepping-stone to a real big deal down the road. What the Venter Institute has done is synthesize a genome that reproduces the genome of an existing organism, then insert that genome into the body of an existing cell, replacing its own DNA. That cell was then able to self-replicate, indicating that the synthetic DNA copy was sufficiently complete.

    "Synthetic" here doesn't mean artificial, by the way. The DNA of the synthetic genome comprises the same base pairs and nucleotides as a natural genome, but was synthesized in the lab rather than replicated from an earlier cell. The best analogy I can think of is if, rather than copying the MP3 of your favorite song, you pulled together a really sophisticated music creation application and reproduced the song yourself, exact in every detail. It's the same, but a synthetic version.

    If that sounds like a lot of work to get something that is essentially the same as the natural/original version, you're right. But this step was never the real goal -- it's just preparation. The real goal is to create an entirely novel life form, comprising both entirely new DNA and an entirely new cell. That's still to come.

    Where we aren't:

  • Transgenic synthetic genome (natural genome copy with genetic code from other kinds of organisms).

    The synthetic genome created by the Venter Institute is a streamlined version of the original Mycoplasma mycoides bacteria, containing enough of the original code to replicate and function as M. mycoides. Adding transgenic features -- that is, genetic material copied from non-M. mycoides species -- should be fairly straightforward, as it's essentially doing standard bioengineering.

    In principle, this should actually be somewhat safer than current transgenic biotech, as they'll have much more precise control over the engineered genomes.

  • Novogenic synthetic genome (entirely constructed novel genome).

    The ultimate goal would be to create an entirely new bacterial species by creating genes that do new things, or by combining diverse known DNA sequences to create a functional, replicating bacteria that doesn't mimic any existing species. This will be hard, but clearly not impossible.

    The bonus goal:

  • De novo creation of cell structure.

    The cell in which the synthetic DNA is housed already existed, but with different DNA (it was the cell of a related species of Mycoplasma). One likely future step will be to create an entirely synthetic cell by throwing together the right set of proteins in just the right way. Like the latest breakthrough, that will undoubtedly start out by simply reproducing an existing cell structure. Ultimately, they'll want to create cellular bodies that have novel features, such as (conjecture here) additional mitochondria for added power.

    Where we go:

    So what does this all mean?

    The idea is to turn bacteria into microscopic machines, carrying out designated tasks in massively-parallel operations. Given the extreme range of things that bacteria can do in nature, the extent to which bacterial machines might be used is pretty staggering, particularly concerning environmental response. This would be a perfect platform for methanotrophic remediation of melting permafrost, for example; the Venter folks are already talking about building synthetic bacteria to do carbon capture. Biofuels are also high on the agenda.

    The big concern about synthetic biology is the potential for the creation of hazardous materials -- aggressive, infectious bacteria, for example. We should also consider, at the same time, its biomedical potential. Are there ways of delivering drugs via synthetic bacteria?

    One advantage of the big splash this relatively modest development has made is that it opens up the possibility of laying out the parameters of what ethical, responsible management of this technology would look like before have to confront its fully-developed form.

    Should we require a "shut-off" gene in any novogenic organism, one that kills the cell if certain conditions are (or aren't) met? A reproduction-limiting set of genes that only permits replication in the presence of a rare chemical? Public registration of all novogenic genomes?

    One suggestion that we know is possible, because a variation appeared in the Venter announcement: all synthetic genomes should be signed. According to Wired:

    “They rebuilt a natural sequence and they put in some poetry,” said University of California at San Francisco synthetic biologist Chris Voigt. “They recreated some quotes in the genome sequence as watermarks.”

    What Voigt refers to as a "watermark" should instead be thought of as a "DNA signature." We should require that all synthetic genomes include something like this, unique sequences following a designated pattern, identifying the organization behind the genome, the lab responsible, the date, and any other useful bits of information. Multiple copies should appear throughout the synthetic genome, so it doesn't get mutated away.

    That way, if something unexpected happens, we know whom to talk to.

  • June 24, 2009

    Your Brain Hates You

    The "blue" and the "green" are identical colors (RGB 0/255/150). Discuss.


    Yes, they are. Grab this image and test the colors yourself.

    Here are more.

    From Phil Plait, the Bad Astronomy blogger, who says:

    The orange stripes go through the "green" spiral but not the "blue" one. So without us even knowing it, our brains compare that spiral to the orange stripes, forcing it to think the spiral is green. The magenta stripes make the other part of the spiral look blue, even though they are exactly the same color. [...]

    This is why I tell people over and over again: you cannot trust what you see even with your own eyes. Your eyes are not cameras faithfully taking pictures of absolute truth of all that surrounds you. They have filters, and your brain has to interpret the jangled mess it gets fed.

    In other words, your brain hates you.

    May 2, 2009

    Open Source Flu (updated)

    In case you were curious:

            1 atgaaggcaa tactagtagt tctgctatat acatttgcaa ccgcaaatgc agacacatta
           61 tgtataggtt atcatgcgaa caattcaaca gacactgtag acacagtact agaaaagaat
          121 gtaacagtaa cacactctgt taaccttcta gaagacaagc ataacgggaa actatgcaaa
          181 ctaagagggg tagccccatt gcatttgggt aaatgtaaca ttgctggctg gatcctggga
          241 aatccagagt gtgaatcact ctccacagca agctcatggt cctacattgt ggaaacatct
          301 agttcagaca atggaacgtg ttacccagga gatttcatcg attatgagga gctaagagag
          361 caattgagct cagtgtcatc atttgaaagg tttgagatat tccccaagac aagttcatgg
          421 cccaatcatg actcgaacaa aggtgtaacg gcagcatgtc ctcatgctgg agcaaaaagc
          481 ttctacaaaa atttaatatg gctagttaaa aaaggaaatt catacccaaa gctcagcaaa
          541 tcctacatta atgataaagg gaaagaagtc ctcgtgctat ggggcattca ccatccatct
          601 actagtgctg accaacaaag tctctatcag aatgcagatg catatgtttt tgtggggtca
          661 tcaagataca gcaagaagtt caagccggaa atagcaataa gacccaaagt gagggatcaa
          721 gaagggagaa tgaactatta ctggacacta gtagagccgg gagacaaaat aacattcgaa
          781 gcaactggaa atctagtggt accgagatat gcattcgcaa tggaaagaaa tgctggatct
          841 ggtattatca tttcagatac accagtccac gattgcaata caacttgtca gacacccaag
          901 ggtgctataa acaccagcct cccatttcag aatatacatc cgatcacaat tggaaaatgt
          961 ccaaaatatg taaaaagcac aaaattgaga ctggccacag gattgaggaa tgtcccgtct
         1021 attcaatcta gaggcctatt tggggccatt gccggtttca ttgaaggggg gtggacaggg
         1081 atggtagatg gatggtacgg ttatcaccat caaaatgagc aggggtcagg atatgcagcc
         1141 gacctgaaga gcacacagaa tgccattgac gaaattacta acaaagtaaa ttctgttatt
         1201 gaaaagatga atacacagtt cacagcagta ggtaaagagt tcaaccacct ggaaaaaaga
         1261 atagagaatt taaataaaaa agttgatgat ggtttcctgg acatttggac ttacaatgcc
         1321 gaactgttgg ttctattgga aaatgaaaga actttggact accacgattc aaatgtgaag
         1381 aacttatatg aaaaggtaag aagccagcta aaaaacaatg ccaaggaaat tggaaacggc
         1441 tgctttgaat tttaccacaa atgcgataac acgtgcatgg aaagtgtcaa aaatgggact
         1501 tatgactacc caaaatactc agaggaagca aaattaaaca gagaagaaat agatggggta
         1561 aaactggaat caacaaggat ttaccagatt ttggcgatct attcaactgt cgccagttca
         1621 ttggtactgg tagtctccct gggggcaatc agtttctgga tgtgctctaa tgggtctcta
         1681 cagtgtagaa tatgtattta a

    That's (Update:) one gene from the Influenza A virus (A/Texas/04/2009(H1N1)) to you and me. Follow the link for sequences from other key nucleotides and proteins from the virus. Collect them all!

    (via Glyn Moody/Open...)

    (Thanks for the correction, debcha!)

    January 15, 2009

    Life on Mars? Why It Matters


    News today from NASA that they've confirmed the presence of methane in the Martian atmosphere, concentrated in three areas (one of the major sources, Nili Fossae, is shown here). For a variety of reasons, this offers the strongest evidence yet that Mars may have an active biology under the surface.

    While both geology and biology can produce methane on Earth, inorganic production of methane is generally associated with volcanic and tectonic activity, none of which has been witnessed on Mars (it's clear that Mars was once geologically active, but there's little or no evidence of current vulcanism). In addition, the three source areas each have very different geologies, further complicating the argument that the methane comes from geological activity. Finally, the "serpentinization" process on Earth tends to plug up sources of methane. NASA's Lisa Pratt, one of the scientists delivering the press conference today, argues that while this isn't positive proof that the methane comes from biological activity, it does make the geological argument harder to sustain and makes the biological argument "more plausible."

    An additional bit of complexity is that the methane seems to be leaving the Martian atmosphere faster than the chemical composition of the Martian environment would suggest. A biological process -- where the methane was being consumed by microbial life -- would fit the evidence. Follow-up research, unfortunately not possible with the current satellites and robots working Mars now, should be able to find more definitive (positive or negative) evidence.

    So what would we have if we determined that there were microbes making methane, and other microbes consuming methane? An ecosystem -- the first ecosystem found someplace other than Earth.

    This would be amazingly important for a variety of reasons, not the least of which being that we'd finally have a chance to do comparative ecology.

    Everything we know about how ecosystems work, how biology works, comes from a single data point: Earth. And while there's quite a bit of diversity within the Earth's ecology, it's all based on more-or-less the same basic biological stuff. What would Martian microbes have as the equivalent of DNA? Genes? Would there be elements of their biochemistry that would be unusually surprising?

    Then there's the possibility that said Martian microbes would have a biology essentially identical to that found on Earth. The most plausible explanation for that would be that Earth life actually started on Mars (which cooled faster than Earth, so would have started its biology sooner) and was exported via Martian rocks ejected from massive impacts and hitting Earth as meteorites. We've discovered Mars-origin meteorites on Earth, so we know this is plausible.

    So many questions. Hopefully, NASA will get the funding it needs to look for the answers.

    August 1, 2008

    Solar Hydrogen (Update: Not So Much the Solar)

    [Updated, changes made throughout.] A possible breakthrough at MIT in energy storage: store the generated energy as hydrogen, using a new, incredibly cheap and easy process that functions akin to photosynthesis. This could be big, and it could give a new boost to the fuel cell field.

    For a few years now, I've been in the "hydrogen is a dead-end" camp (the most prominent member probably being Joe Romm, author of The Hype About Hydrogen). The compromises required to get a hydrogen infrastructure up and running -- not the least of which being abandoning the clean path by reforming hydrocarbons rather than cracking water -- coupled with the clear advances in with hybrid and full-electric vehicle technologies have really put hydrogen out of the running as a technology path worth pursuing, in my view. Ultracapacitors and nano-enabled batteries seem like the winners, and given how low-profile the fuel cell world has been in the last couple of years, it seemed like my view wasn't all too uncommon.

    But along comes MIT's Daniel Nocera, with a new method -- similar to the way that plants derive energy from sunlight -- that he claims will turn regular ph-neutral water into oxygen and hydrogen using low-cost, easily-obtained materials. (Science abstract here.)

    Nocera argues that this will make solar the dominant energy-producing technology, not simply through direct electricity generation, but through the production of hydrogen for fuel cells, which can be used in vehicles, for overnight power, and so forth. I'm unclear as to why Nocera is emphasizing solar here -- if this is as much of a breakthrough as he claims, it would be applicable to any kind of electricity generation.

    Fuel cells actually make a great deal more sense as a building power system than for cars, in my view. Issues around weight and density of the storage of hydrogen are far less problematic when all the fuel cell power systems have to do is sit on the ground. Similarly, public concerns about the safety of hydrogen (the Hindenburg will haunt us all for decades more) can be more readily alleviated when the fuel cell has a near-zero likelihood of being in a collision.

    I'm still inclined to lean towards battery/ultracapacitor electrics over fuel cells for transportation power, but I'm happy to see revived competition from the hydrogen sector.

    May 26, 2008

    Update: Mars Gets the Robots it Needs


    May 24, 2008

    Mars Need Robots!

    NASA's latest Mars lander, Phoenix, is scheduled to land at the north pole of Mars Sunday at around 4:30pm PDT. Unlike the Mars rovers, Phoenix is a static lander (and, as the above video simulation shows, even goes all old-school with a rocket landing system rather than the giant airbag cushions).

    Plenty of links for the Areophiles out there: the Planetary Society's blog will carry updates; NASA TV will show live footage from mission control; and believe it or not, the Mars Phoenix mission has a Twitter stream.

    November 11, 2007

    I Spy With My Orbital Eye...

    flaring.jpgOne of my favorite early pieces for WorldChanging was the essay Greens In Space, arguing that space exploration, particularly robotic exploration, is ultimately in support of the Bright Green future. Of particular importance are the satellite systems used to observe changes on the Earth's surface. Two articles this past week nicely underscore that point.

    The first (via James Hughes) comes from a report at the American Society of Tropical Medicine and Hygiene conference in Philadelphia: NASA satellites help health policy experts around the world watch for and respond to disease outbreaks, and can potentially help head off a pandemic.

    The use of remote sensing technology aids specialists in predicting the outbreak of some of the most common and deadly infectious diseases today such as Ebola, West Nile virus and Rift Valley Fever. The ability of infectious diseases to thrive depends on changes in the Earth’s environment such as the climate, precipitation and vegetation of an area. [...] Remote sensing technology not only helps monitor infectious disease outbreaks in highly affected areas, but also provides information about possible plague-carrying vectors -- such as insects or rodents -- globally and within the U.S.

    It's a simple story, but a useful reminder: we have the tools to fight these crises.

    The second (via Ethan Zuckerman) is even more directly Bright Green: the use of satellite imagery to detect natural gas "flaring," in order to track its impact on the environment.

    Natural gas often comes along with oil drilling, and -- amazingly -- some companies find it cheaper to burn off the gas onsite ("flaring") than to capture and sell it. This, in turn, appears connected to acid rain and lung disease, and simply dumps more carbon into the atmosphere without even generating useful work out of it. In 2002, Norway and the World Bank initiated the Global Gas Flaring Reduction partnership, trying to reduce the frequency of the practice. In order to enforce the agreement, and get a better handle on just how much gas flaring is underway, the GGFR brought in experts in satellite analysis to begin poring over data stretching back nearly 20 years.

    The results found with this new tool are surprising. Conventional wisdom says that gas flaring is decreasing - the study found that it’s actually been pretty constant from 1995 to 2006. It’s been accepted that Nigeria is the biggest offender in gas flaring, conducting 20% of worldwide gas flaring. But the Nigerian government - in part driven by activism and violence in the Niger Delta, as well as concerns about health and environment - has been attempting to reduce gas flaring. [...]

    According to the analysis by NGDC, the real bad boys of gas flaring are the Russians, who flare twice as much gas as the Nigerians. Russia, unfortunately, is not a member of the Global Gas Flaring Reduction consortium - having data that shows that they’re the largest offender might help bring them within the fold.

    Ethan's notes on the details of the gas flaring analysis are well-worth reading, and I encourage you to follow the link.

    Back at WorldChanging, I used to post quite frequently about the variety of satellite-based projects underway to make information about the environment, human rights, agriculture, disease, etc. etc. more transparent and available. I haven't done that much over here at OtF -- in a way, the eyes in orbit have just been an assumed baseline. But with massive cuts to NASA's environmental satellites division, such an assumption is no longer warranted. We should take note of, and celebrate, the remaining satellite-based efforts while we can.

    June 8, 2007

    Surface Water on Mars? (Update: Probably Not)

    blue_mars.jpgWoah. If this is confirmed, it's big.

    New analysis of Mars Rover images taken a couple of years ago in the "Endurance" crater seem to show standing pools of water on the Martian surface.

    Along with fellow Lockheed engineer Daniel Lyddy, [physicist Ron] Levin used images from the Jet Propulsion Laboratory's website. The resulting stereoscopic reconstructions, made from paired images from the Opportunity rover's twin cameras, show bluish features that look perfectly flat. The surfaces are so smooth that the computer could not find any surface details within those areas to match up between the two images.

    The imaging shows that the areas occupy the lowest parts of the terrain. They also appear transparent: some features, which Levin says may be submerged rocks or pebbles, can be seen below the plane of the smooth surface.

    This would greatly boost the likelihood of finding near-surface Martian life; in fact, the father of one of the authors, Gilbert Levin, laid out the evidence for Martian life (PDF) in a paper delivered at the Carnegie Institution Geophysical Laboratory last month, relying in part on the water discovery. Gilbert Levin was principal investigator on the Viking lander experiment that appeared to show signs of life.

    The main argument against this idea is that the density of the Martian atmosphere is so close to vacuum that water coming to the surface should just sublime away instantly. But some areologists have proposed that water may be able to exist for longer periods on the surface if certain conditions are met -- conditions that are most likely to occur in deep craters like Endurance.

    Update: It looks like Ron Levin didn't do his homework on this. Follow the link Peter Erwin provides in the comments. Short version: the image that Levin processed for his research turns out, when examined in context, to be part of a tilted cliff face, not a horizontal surface -- not a good spot for still water.

    May 24, 2007

    Here Be Dragons

    blackholeofmars.jpgThis is a picture of a mystery -- and a tantalizing possibility.

    Click on it for the original. It's a picture of a "subterranean void" on Mars, taken by the HiRISE ultra-high-resolution camera on the Mars Reconnaissance Orbiter. The resolution on that photo is 25 centimeters per pixel; the void shown is about 100 meters across.

    This is one of seven holes in Mars; all seven are along the flank of Arsia Mons, the southernmost of the Tharsis volcanos. Presumably they're cave entrances, but -- so far -- even the HiRISE camera can't see anything in there. Mars has a dusty atmosphere; if these were shallow depressions or cave openings, scattered light would be visible in enhanced images. But absolutely nothing is visible. At the very least, that means they're really, really deep.

    What's particularly exciting about these caves is that they may be the best places to find extant life on Mars. According to USGS scientists (PDF):

    Subterranean void spaces may be the only natural structures on Mars capable of protecting life from a range of significant environmental hazards. With an atmospheric density less than 1% of the Earth’s and practically no magnetic field, the Martian surface is essentially unprotected from micro-meteoroid bombardment, solar flares, UV radiation and high-energy particles from space.

    Thermal imaging of the voids show that they maintain a relatively constant temperature, remaining relatively warm in the cold Martian night.

    Who's up for a bit of spelunking?

    April 25, 2007

    Don't Pack Your Bags Just Yet

    reddwarfplanet_sm.jpgYou can't swing a dead cat-5 cable on the Interwebs today without running across a link to the "new Earth" discovered around a red dwarf star called Gliese 58, about 20 light years away from Earth (not just in our back yard, but -- relatively speaking -- right behind us, reading over our shoulder, breathing stale dorito breath in our face). Let's guess the order of the blog storm: first, someone will say "it's a habitable planet!"; then, someone will say "we should move there!"; then, someone will say "you just want to trash the Earth and leave it like a cheap rental!"; finally, someone will say "hold on, folks, all they found was a planet that's likely to be "rocky" (instead of gaseous like Jupiter) in an orbit that would allow water to remain water. That's it. No signs of water, no actual proof of habitability, certainly no signs of life. Calm down."

    So, jumping to the conclusion: calm down.

    This is cool news, to be sure, but really only from the perspective that it supports the argument that the preconditions for Life As We Know It -- i.e., water and a stable orbit around a reasonably long-lived star -- appear to be, in fact, about as commonplace as had been conjectured.

    But don't worry about Earth-haters moving there. By the time we have the technology that would make a 20 light year trip even remotely plausible (the fastest space craft yet made would still take thousands of years to get there), we probably won't be all that interested in living in a watery gravity hole anyway. Nope -- give us some nice, massive gas giants to convert to computronium!

    April 24, 2007

    The Early Signs of the Long Tomorrow

    pbrain_sq.jpg(Or "I, for one, welcome our new cyber-mouse overlords!")

    Ahoy, BoingBoing readers! I was going to update this anyway, but with the BB link, it's extra-important: this is a simulation of a cortical network with the size, link complexity and signal activity of a mouse brain, but without the structure -- so, arguably, it isn't a really a simulated mouse brain, but a functional platform upon which a mouse brain sim could run. Depending upon your perspective, this is a minor quibble or makes all the difference.

    It's hard to see this as anything but a distant early warning of some pretty remarkable changes on the near horizon. IBM researchers James Frye, Rajagopal Ananthanarayanan, and Dharmendra S. Modha assembled a simulated mouse cortical hemisphere (that is, a functional half of a mouse brain) on one of the smaller BlueGene/L supercomputers. They then ran the simulation -- at ten seconds of computer processing equal to one second of brain function.

    In other words: they ran a simulated mouse brain at 1/10 time.

    Neurobiologically realistic, large-scale cortical and sub-cortical simulations are bound to play a key role in computational neuroscience and its applications to cognitive computing. One hemisphere of the mouse cortex has roughly 8,000,000 neurons and 8,000 synapses per neuron. Modeling at this scale imposes tremendous constraints on computation, communication, and memory capacity of any computing platform.

    We have designed and implemented a massively parallel cortical simulator with (a) phenomenological spiking neuron models; (b) spike-timing dependent plasticity; and (c) axonal delays.

    We deployed the simulator on a 4096-processor BlueGene/L supercomputer with 256 MB per CPU. We were able to represent 8,000,000 neurons (80% excitatory) and 6,300 synapses per neuron in the 1 TB main memory of the system. Using a synthetic pattern of neuronal interconnections, at a 1 ms resolution and an average firing rate of 1 Hz, we were able to run 1s of model time in 10s of real time!

    The team published the write-up in the February 5, 2007, edition of Computer Science; a PDF is available of the one-page research report, providing a few technical details.

    The human brain has some 100 billion neurons, so this mouse brain simulation is still about 1/12,500 of a simulated human brain. That may sound like a daunting challenge, until a glance at computer history makes clear that such computational capabilities will likely be possible on within 20 years, easily, if not even sooner.

    But well before that point, we'll be able to run simulations of animal brains at accelerated speeds, raising a provocative test of just how important raw cognitive speed is to the emergence of artificial intelligence. Would an accelerated mouse brain simulation simply be a fast-calculating mouse, or will it have other properties and capabilities deriving from the sheer speed? Which would be smarter -- a 6,000X faster mouse brain sim, or a 1/2-speed human brain sim?

    Some of that is going to depend upon how much of the simulation models actual brain structure, rather than simply the number of connections. That's likely to be crucial. The brain isn't simply a haphazard mass of neural junctions, and a functional structure simulation may well prove to be a far greater challenge than simply getting the neural connection sim working. Still, this is not an unsolvable problem, by any extent.

    But this raises the question of whatt kinds of programming will be possible with these simulated brains. The IBM simulation simply showed that a functional simulation was possible; evidently, they didn't try to do anything with the cyber-mouse. It's not entirely clear what could be done with it. We're now on the brink of facing a question that had, in the past, been essentially the province of science fiction:

    How does one program a simulated mind?

    (Thanks to Miron for the tip!)

    March 20, 2007

    Crescent Moon from Earth's Orbit

    Today's Astronomy Picture of the Day picture.

    February 26, 2007

    Rosetta and the Craters of Mars

    I love pictures from space probes, and particularly get a kick out of the shots that include some evidence of the probe in question -- tracks in the sand, antenna booms, and such. These pictures offer a much greater sense of "being there" than do the traditional panorama scenes (lovely though those may be). The vast majority of these shots come from landers, so it's always a delight to see a picture from a robotic spacecraft that includes a bit of itself.

    The picture is from the Rosetta probe, heading to its 2014 meeting with the a comet in the far reaches of the solar system. On February 25, Rosetta made its closest approach to Mars for the slingshot speed boost, and snapped this picture as it did so. (Technically, Rosetta's lander, called the Philae, took the picture, but still.)

    Link to the European Space Agency report on the shot; click the image for a larger version.

    February 12, 2007

    Happy Birthday, Mr. Darwin!


    Charles Darwin, born 198 years ago today...

    November 20, 2006

    A Life or A Person?

    Well, that didn't take long.

    Back in September, I pointed to news that the sleeping drug zolpidem (sold in the US as Ambien) could awaken patients in persistent vegetative states about 2/3s of the time; that post led to ensuing discussions of the ethical and legal issues that could emerge from this discovery (see here and here). The first of what may end up being many legal battles over the use or non-use of this treatment has now taken place in the UK. Perhaps surprisingly, the position taken by the family of the patient was to reject the use of zolpidem, and to allow their relative -- who had suffered serious brain trauma -- to die in peace. The doctors, conversely, wanted to try the drug, and the high court agreed with the doctors.

    The family clearly realized something I mulled in the last of the three pieces from September: that just because zolpidem may "awaken" the PVS patient, the severe physiological trauma that produced the state remains, and the quality of life of the newly awakened patient may be torturous for both the patient and family.

    We already live in a world in which medical science can keep a body alive despite horrific damage; we're approaching a world in which the same will hold true for the brain. But what does "alive" mean in this context? The more we understand life as a purely mechanical function -- the muscular action and the chemical flows and whatnot -- the more it becomes necessary for our culture to separate being "alive" from being "a person." Our norms and values around human life emerged from a time when the "breath of life" and the "beating heart" were truly indicative of a living person; now we can instill both in a body, almost regardless of any other trauma. We now declare death to be the lack of brain function... but what happens when we can stimulate some level of brain function, almost regardless of any other trauma?

    For some of us, these are academic questions, provocations to be discussed around a lecture hall or coffee table; for others, they are moral questions, with a clear, immutable answer. But, as this case should remind us, for some people, these are deeply personal questions, about a loved one who once laughed, and felt, and thought, and truly lived. Let's not lose sight of that.

    May 25, 2006

    Synthetic Biology

    sblogo-small.jpgThe Synthetic Biology 2.0 conference just ended, and Rob Carlson (of open biology fame) and Oliver Morton (author of the terrific and under-appreciated Mapping Mars) attended and blogged the event. Carlson is working on his book on open biology (Learning to Fly: The past, present, and future of Biological Technology), and used his comments about the event to offer up a sample of his book-in-progress. Morton's notes are more extensive (unsurprising, given his day job as a writer/blogger for Nature), and look in some detail at the question of just how the synthetic biology tools would be used.

    So what is synthetic biology? I wrote about it a few times at WorldChanging, and the following description still works:

    [Synthetic biology] is the application of mathematically-driven engineering principles to the construction of novel genetic structures; in contrast, genetic engineering is often a trial-and-error process, with numerous opportunities for and examples of unanticipated results. Many of the reasonable concerns about GMO foods and animals come from this hit or miss aspect of biotech. Biological Engineers have a more systematic approach, and use an increasingly deep understanding of how DNA works to then make microorganisms perform narrowly specified tasks.

    The engineering model underlying synthetic biology goes so far as to include the use of "bio-bricks" as construction elements.

    Synthetic biology specialists (it seems a bit off to call them "synthetic biologists") have managed to create both re-engineered versions of existing single-cell organisms and entirely novel "vesicle bioreactors," objects which display most of the characteristics of life.

    As Carlson notes in his blog, the difference between the Synthetic Biology 1 conference and the Synthetic Biology 2 conference was that the first was all about the science, and the second was increasingly about the money. Synthetic biology is getting awfully close to commercial and potentially practical applications; this means it's getting awfully close to needing some kinds of regulation and scrupulous oversight.

    It could, however, eventually become the organic equivalent of Lego, a way to build bio-objects quickly and safely, for experimentation, education, and occasional practical use. The use of pre-designed modules would go a long way towards keeping the whole process relatively safe. If these bio-Lego came with some kind of Creative Commons or GPL-style license allowing for the distribution of products, you could even imagine a kind of open source synthetic bio movement.

    What's particularly interesting to me about Synthetic Biology, however, is that it's a first draft of what we'll see at the advent of molecular nanotechnology: a simpler, less-capable, model, perhaps, but offering many of the same regulatory and access questions that will emerge when nanofabbers become possible. If we can work out reasonable rules, we're almost certain to apply them to similar future technologies; if we can't, that foreshadows even more difficulty for complex future technologies. How the scientific, engineering, marketing and policy-making communities work together to figure out how to manage the commercial use of synthetic bio will likely have a great deal of influence over how molecular nanotech is regulated. CRN and other interested parties, take note.