Scientists have inserted a GIF of a horse into living bacteria
techcrunch.compreviously: https://news.ycombinator.com/item?id=14755429
Yeah, not to be prude, but this was literally posted yesterday. The actual article, too.
I gloriously await from teenage hackers 10 years from now release CRISPR viruses/worms that insert GIF meme pranks into the human population germ line, eventually triggering the apocalypse.
It seems our ability to mess stuff up is growing faster than our ability to defend against bugs. I shudder to think what happens when sloppy engineering practices or "WannaCry" meet biology. Hey, we've encrypted your germ line, and sterilized you, send Bitcoin to XXXXX to restore your fertility.
Genecoin, cryptocurrency implemented in the human genome. We take a blood sample to check your balance.
edit: putting the bank in blood bank
So vampires would suck your blood.... as a way to rob your bank account?
I think we may have a good (trashy) sci-fi novel here.
>I gloriously await from teenage hackers 10 years from now release CRISPR viruses/worms
You're going to be waiting awhile.
The hard part about biology is that it often doesn't work. To get things working even semi-reliably you need tens of thousands of dollars of sensitive equipment and reagents. Equipment and reagents that would catch the attention of interested authorities if you bought it as an independent.
Computer viruses are the result of a tractable ecosystem and a low cost to participation. Biology is largely untractable, with a tremendous cost of entry.
Yes, CRISPR helps us insert DNA, but it doesn't change all the sensitive steps up to the point of transformation. Even then, CRISPR is limited to the cells being treated.
A custom biological virus is what you are imagining, with the ability to both insert DNA/RNA and with a capsule that protect it during transmission. Unfortunately we are likely centuries away from being able to code a completely custom life form, and even then, the cost and training needed to create such life forms will be cost prohibitive, leaving only corporations and governments the ability to do so.
I don't think you need a completely custom from scratch lifeform. It seems to me you just have to find a flexible existing pathogen that can tolerate a CRISPR edit sequence tagging a long. Perhaps a virus is too small, maybe some common gut bacteria or fungus can be weaponized.
Admittedly, we're a "long way" from that, but for me, I define "long way" as 20 years. Centuries? Seems way past the singularity horizon for me. I think it's really hard to say at this point something is too hard to be solved in 100 years unless it's something that defies the laws of physics, or takes ungodly amounts of mass or energy.
> It seems to me you just have to find a flexible existing pathogen
We can do that now. It's fairly easy for anyone to edit viruses or bacteria using non-crispr methods to make something more virulent.
The problem is there is an evolutionary trade-off. To be more virulent means dedicating resources to it, to the detriment of other kinds of fitness. The result is your virulent pathogen cannot spread, defeating the point of a weapon.
We are many decades away from achieving systemic understanding necessary to improve an organisms fitness using molecular tools, and also account for compensatory changes to fitness.
You can however use selective breeding to do this, but that's millenia old technology.
> I define "long way" as 20 years.
We don't even understand most of the existing proteins in databases yet, and it is going to take decades to meticulously characterize these sequences.
Worse, nobody has come close to a useful denovo protein sequence yet, let alone a genomes worth of proteins and their interactions. I feel safe in saying we are at least 50 years off from these goals, probably more.
Not too many decades ago, computers were also prohibitively expensive and largely only available to corporations and governments.
I don't think we can discard any possibility given the exponential nature of human advancement.
You are assuming growth will continue.
A flash in the pan is also exponential, for a time.
Aren’t you just afraid because you don’t really understand how CRISPR works? It’s not a virus or anything, nor is it airborne.
No, at a high level I understand how it works. I didn't claim CRISPR is a virus. I'm talking about people using other pathogens to piggyback CRISPR edits.
Then again, I know a lot of childfree people who would pay good money to have CRISPR sterilize them.
I'd say "myself included", but my gender transition will take care of that. Hmm... I wonder if a future version of CRISPR could get my body to produce its own estrogen.
Here's the image that was encoded:
https://commons.wikimedia.org/wiki/File:Muybridge_horse_gall...
It's iconic photography!
Can we stop with the slippery slope arguments? No, CRISPR will not be used be teenage hackers in the future to infect humans and keep them hostage. CRISPR can’t be “spread” or “transmitted” and is a local technique to introduce DNA snippets into the main genome of the organism. And to anyone who thinks that CRISPR could be potentially used to infect humans, this technology is nothing new: viruses have already had DNA editing machinery for millennia, search up retroviruses.
Note that when the popular press worries about the destructive potential of "CRISPR", they don't mean CRISPR specifically. They're using the word as a stand-in for the general trend of the exponentially decreasing cost of genome sequencing and editing techniques which probably, eventually, will reach the level where they can be done by teenage hackers. And that is something that requires concern and attention, even if I don't think it merits panic.
I think you are much too sanguine about this, which I've noticed is a tendency among programmers in general. We're used to a world where bugs always have fixes, and can be patched and the fix spread worldwide instantly, for free. None of these things are true in epidemiology, and it causes us to make really bad metaphors that don't apply in this world. For starters, "responsible disclosure is good" and "security-by-obscurity is bad" are obvious truisms in software but not even remotely true in epidemiology.
> viruses have already had DNA editing machinery for millennia, search up retroviruses.
But viruses are subject to evolutionary pressure that puts a ceiling on how bad they can be. Most of the worst human viruses have a "pick two" of 1) airborne 2) highly infectious 3) fatal, because a virus with all 3 burns itself out and can't survive. Engineered organisms do not have this ceiling.
To be fair, we do have examples of some superbugs, such as the Spanish Flu in the 1910s, where an influenza virus managed to wipe out a significant portion of the world. It would be considered by most to be airborne, infectious, and fatal. Maybe my optimism results from the fact that a supervirus has never been engineered (which I agree is fallacious), but I don’t think CRISPR, a research technique, even begins to encroach on the topic of genetically engineered superbugs.
> No, CRISPR will not be used be teenage hackers in the future to infect humans and keep them hostage. CRISPR can’t be “spread” or “transmitted” and is a local technique to introduce DNA snippets into the main genome of the organism
I think most people, if not everyone, already got this.
> viruses have already had DNA editing machinery for millennia
But humans never had it. That's probably what creeps everyone out.
I'm not sure I understand your argument. The fact that "this technology is nothing new" isn't precisely what proves that it can be potentially used to infect humans?
They should encode some music on it and then watch the RIAA go apeshit trying to track down and kill all the copyright-infringing bacteria.
If someone finds some protected data as controversial as AACS encryption key[2] and creates a "fork" of HIV which contains that information, I'm optimistic that we will get a cure for it in a few years.
On a more serious note, I'm not sure, with this kind of power, if we are quickly approaching a Great Filter[4] or not.
[4]: https://en.wikipedia.org/wiki/Great_Filter
[2]: https://en.wikipedia.org/wiki/AACS_encryption_key_controvers...
> I'm not sure, with this kind of power, if we are quickly approaching a Great Filter or not
We're approaching several.
We -- or rather, some members of our species -- are clever enough to build amazingly powerful technologies (nuclear weapons, biotechnology, maybe strong AI), but as a species we seem to often use them unwisely, either because political leaders are power-obsessed (in the case of nuclear weapons), or because some technologies might become able to be harnessed by individuals/groups without much resources (DIY biotechnology).
It's like toddlers playing with matches on a petrol-soaked carpet.
so storing data on multiple (billions of bacteria). I assume that you would have populations of duplicates and would piece together your entire original dataset (e.g. a tar) using start and end encoding tags? How do you re-sequence multiple populations of bacteria that you just grab from a sample sitting in a -80c freezer? You'd also have to prevent them from modifying their own genome and messing up your data.
You could store it redundantly, in a genetic RAID
I read the article, but I'm still quite confused. How exactly did they "insert" a computer image into living bacteria?? Do they mean they have stored binary code in the bacteria?? Or did they just physically create a fine image somewhere in the bacteria???
They inserted the binary code in the dna using CRISPR and CAS-9
they store the data in the dna, they encode pixels using ATGC, and re-sequence. But they only have one population of bacteria that contain all of the same data.
I don't know much about how DNA works, but by inserting ATGC sequences, does that not mess up the bacteria's replication process or something? How would the cell 'read' this encoded gif in the DNA? Would it try to build a protein because of it, or would it ignore the sequence? Do cells usually have a checksum-type of functionality to detect if their DNA is valid/unmutated? The article itself didn't make any mention, but could there possibly be negative impacts to the cell hosting this data by having its DNA altered?
In order for proteins to be produced, there is usually a very specific start/stop sequence that has to be present in order for the mechanism to take action.
Most likely this is just "junk" DNA that has no physical manifestation.
As for reading the picture, it is only seen when they extract the DNA and use methods of sequencing to produce the image, meaning it will only show the little gif when analyzed the way it was designed for.
DNA of many organisms has "dead" sequences that currently don't do anything, but still get copied/replicated. (or are only activated under certain circumstances, which even can be programmed in some sense by adding specific other sequences around them)
Right I'm no biologist, but my understanding is that there's vast amount of dna that doesn't really do anything (doesn't encode any proteins). I don't remember why... maybe it used to do stuff a long time ago.
You are thinking of introns. Their function is still not very well-known. https://en.wikipedia.org/wiki/Intron#Biological_functions_an...
So, eventually we can look forward to getting rickrolled by a bout of food poisoning. And maybe get sued for piracy afterwards, for unauthorized duplication and distribution. Unless consideration of fair use is updated to include involuntarily crapping out copies.
How likely is it than storing arbitrary data in a bacteria's DNA would allow it to create something dangerous for humans?
Approximately like the chance that DNA sequence of a virus, when executed as shell code, will infect your machine.
Buy your subscription to Norton Anti-virus 2030 today!
If not for yourself, do it for the children.
The reason I ask is that it seems subsequences of DNA are enough to generate proteins, or other molecules. (am I right? if not anyone who knows biology please correct me)
It doesn't have to be the whole code that's dangerous, what about a section of it?
That still makes it very improbable, but orders of magnitude more than the whole sequence.
The encoding used by these sorts of projects avoids 'ATG' (start site), and possibly the Shine-Dalgarno Sequence (where ribosomes attach), if they do it right. The bacterial DNA can mutate, but the likelihood of that mutation creating a viable protein, let alone a dangerous one, is very low. Honestly, since these stored data aren't coding, the chances of them surviving for long, even in a population of bacteria, is low. Bacteria tend to favor smaller genomes, with higher percentage of coding DNA.
Inspecting for exfil just got a lot harder.
Seriously, now border & customs agents will have to require unlocking your DNA as well as your phone.
What are the odds this is being used for exfiltrating sensitive data? (Reminds me of a Culture novel, I think Excession, where one character tries to sneak information out of a ship by encoding it into the DNA of bacteria on, essentially, packages.)
Zero right now, but the future where borders agents require you to unlock your DNA as well as your phone makes for a great sci-fi dystopia.
DNA of which organs should the border agents examine in that future?
What does this have to do with the GIF image format? Wouldn't using something other than LZW compression give them a much faster converging restoration rate?
Anyone have any info on the amount of data that was encoded?
It looks like it was a low resolution image with 5 frames, so I expect it to be about 1 Kb, max?
Nature paper say 2.5 kilobytes of data, and I guessimate a few Kilobases of sequence.
If I hadn't seen this tweeted around earlier, I would've guessed this title/story generated by the HNSimulator markov bot.
When did techcrunch get bought out by Buzzfeed? 'did your head just explode' is such a link-bait-douchy title
The next big leap: putting Pepe memes in a frog
Okay great, but how did the scientists pronounce it?
Look at my horse. My horse is amazing.
Pay-walled science don't you just love it? Why do they bother - it isn't science until you publish IMHO. This is just showing their rich friends.
I suspect that falsely denouncing paywalled research as "not science" isn't going to work in bringing about the outcome that you want. Probably a strategy without so many scotsmen would work better.