Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Reelya]

Some pretty dangerous shit could be made with a setup that small.

I don't really understand how this would be any more dangerous than any other compact storage medium, especially when it comes to illegal drugs. We already have small enough media to fit hundreds of gigabytes in the palm of one's hand, and it's not like any customs office asks you to pull out the SD card from your camera to show it doesn't have anything illegal on it.

I think wierd's point was that you could easily make yeast that synthesizes drugs, making it harder for the DEA to enforce drug laws, as well as attaching dangerous stuff to microbes that can replicate quickly. The DNa-data-storage technique was the way you get the arbitrary sequence in there to say, make THC yeast or something.

[Trekkin]

In this particular case, he appears to be suggesting that someone create yeast that can synthesize things like tetrahydrocannabinol, which is...doable but not particularly useful in terms of yield, particularly in a garage setup.

Some drugs are more potent than others.

That's not really the issue here, since the more potent the compound is (and thus the less efficient the synthesis needs to be) the smaller the quantity that logically needs to be shipped/smuggled/synthesized chemically. In the case of something like botulinum toxin, which seems to be the most common starting point for my-drug-is-deadlier-than-yours pissing contests, you can certainly get yeast to express it, but you can also get C. botulinum to express it or just go get Botox via whatever dastardly means, both of which have a genetic engineering cost of zero.

See, people forget that the bacteria and yeast we work with in lab are deliberately designed to accept foreign genetic material and express as much as possible. That's great for our purposes, but in making them be so accommodating we've inactivated most of their immune systems; they're trivially easy to kill and anything from the wild will outgrow them even before you start making weird demands of their metabolism by inserting useless gene products. While it is hypothetically possible to grow them in a garage environment, it's certainly more expensive than just growing the organism that's already capable of making whatever compound you want, particularly if it's a small molecule rather than a peptide you're after. For all the silly ways that people try to cheaply sterilize media, dirt is still going to be cheaper per unit output, and ultimately that's the metric dictating this hypothetical misbehavior.

[wierd]

RE: Reelya

Bingo.  The intent is to remove the produced synthetic DNA molecule the device produces, from the machine, then mass replicate it with amplification, and then do your nefarious deeds with it.

That it lets you encode arbitrary data, and produces consistent codon arrangements based on that arbitrary input, means you can program the machine to produce arbitrary USEFUL codon sequences to do many "interesting" things, where what is of interest to some of the people this kind of tech could one day enable are drug cartels, drug cooks, and tin-pot dictators.

Re: Trekkin

Not all people that could be enabled by the advancement and proliferation of this kind of technology are going to be garage tinkerers.  As mentioned, some are drug cartels (which can afford buildings and equipment, but are likely to suffer a lack of qualified workers) or tinpot dictators (which could get access to both of those, but may have reduced access to equipment and raw materials due to sanctions and logistical supply chain problems caused by strife, open rebellion, systemic corruption, etc.)  Either of those can realistically produce a properly contained facility in which to grow the premadonna microbes produced. The danger is in reducing the needs of skilled staff, and of lowering the bars on costs of equipment and development through proliferation.

[Trekkin]

The danger is in reducing the needs of skilled staff, and of lowering the bars on costs of equipment and development through proliferation.

Oligonucleotide synthesis was never the limiting factor, though; you can get liquid-handling robots, random semiliterate schmucks, or even undergraduates to make whatever oligos you want on the bench in tubes, and nucleotide phosphoramidites aren't prohibitively expensive. The limiting factor is still yield per unit effort, and, like I said, if as an example you want THC, pot plants are good at making it and pot farmers are cheaper than synthetic biologists.

I'm also not convinced the device you posted can in fact make arbitrary sequences; it's entirely possible they're using a biologically incompatible barcoding scheme.

[wierd]

I think there is a miscommunication here.

I am concerned about this device, and consider it dangerous, because of what it represents, not necessarily it being intrinsically dangerous.

EG, it represents the impetus of disparate market forces converging on very inexpensive synthetic biology synthesis.  (Microsoft is a software company, and the notion that you could integrate GAN based AI into the mix, along with improvements in protein fold prediction and function, gives rise to a potential (and highly desirable by industry) future in which the need for specialist synthetic biochemists becomes lower and lower, and the barrier to entry becomes more political than logistical or economical. 

What I find scary, is that a device for encoding DNA without human oversight is being skunkworked, at a "reasonably" affordable one-off price, (10k) and people are uncaring or unconcerned about the implications of these market trajectories.

Similar kind of dangerous outcome that has already come to pass:

"Why would you be concerned about VoIP tech? Why would you be concerned about falling prices to make phone calls?"

Oh, No reason.

(As the cost barrier goes down, the incentive to become a bad actor increases, as there is no cost, and significant areas for gain. The same would be true of vastly expanded synthetic bio tech, with computer assisted synthesis and integration. The costs associated would cease being significant barriers against malicious actors, as their perceived gains would only grow. Eventually, like with Robocalls, the notion of NOT doing the bad things would become laughably naive.  The time to make a stink about it is BEFORE the markets can reach unstoppable momentum-- EG, right now.)

[Trekkin]

I think there is a miscommunication here.

I am concerned about this device, and consider it dangerous, because of what it represents, not necessarily it being intrinsically dangerous.

Indeed there was:

this plus some infamous techniques like CRISPR and some yeast, could make the DEA's job literally impossible to enforce.
[...]
Some pretty dangerous shit could be made with a setup that small.

That said, my point was that what it represents already exists and is already available. You can posit additional advances dropping out of the ether that would make this more dangerous, but if you want to spend $10k on a way to make arbitrary gene sequences, the public has been able to do exactly that for about a decade now with paper-based microfluidics. You can similarly posit bad actors with arbitrarily ample resources to handwave away the costs of actually making anything from the DNA machines like this give you, but in that case the cost of one of the cheaper parts of the pipeline is even less relevant.

The technology driving this machine is well over thirty years old at this point, and it's not doing anything new or even more cheaply; what we have here is an expensive machine competing with an inexpensive public-facing service. Solid-phase oligonucleotide synthesis has gotten orders of magnitude cheaper to do en masse, but bad actors don't need to operate on those scales; they don't want to make hundreds of thousands of variants of some bad thing and if they did the cost of characterization would vastly exceed the cost of synthesis anyway.

All this device needs to do, and all it apparently does, is encode information in low-complexity oligonucleotides; that's not gene synthesis per se and doesn't necessarily lead to it, since synthesis-inhibiting secondary structure is much easier to avoid if you can intercalate your data-storing base pairs with compensatory junk regions, even on an ad hoc basis. We already do that without human intervention in scaffolding complex regions for sequencing, as well as in barcoding for next-gen sequencing. It offers no real advantage in solving the harder problem of exactly synthesizing arbitrary protein-coding DNA over any other automatic fluid handler, and those are already publicly available along with the algorithms to optimize reverse translation.

Your disparate market forces already built us better synthesizers for the pharmaceutical industry well over a decade ago, they've been in the lab for years before that, and bad actors can already use them more cheaply than they can this device for the sorts of things you posit. This is just a simpler version for solving a simpler problem.

[Kagus]

Does this go here? I see there's already a DNA conversation happening, but I can never tell which science thread to post in...

Anyways, Scottish Grandma Wolverine feels no pain and apparently heals without developing scar tissue.

Scientists: "This is pretty neat!"

Journalists: "Scientists say that this discovery could definitely be used to help surgery patients and other pain cases!"

Also Journalists: *Neglect to mention anything about her healing without scar tissue, which while definitely interesting makes me think it must take a lot longer for her wounds to properly close. Also neglect to really mention anything about how terrifying zero-pain can be, and that gene editing generally isn't considered a practical alternative to anaesthetics*

[wierd]

A suitable antagonist for the FAAH gene modeled after the protein produced by FAAH-out could be administered a few hours before surgery, resulting in temporary suppression of the pain reception pathway. (but I doubt journalists realize that would be at least 20+ years away, because it would have to first be developed, than proceed through animal model trials, then through phase 1 and 2 clinical trials, and then finally through the FDA approval process after the phase 1 and 2 populations were tracked for long therm effects.... Yeah. That's even assuming a stalwart set of researchers WANT to do that kind of thing..)

See also, the experimental HIV treatments modeled after CCR5 delta 32.  They dont give normal people the delta 32 mutation, they just act to render that receptor useless, like the mutation causes.  They only work while they are in your system.

More interesting is how she does not develop scar tissue.  They identified the novel gene that controls FAAH, but did not mention the novel pathway to suppress the formation of scar tissue.  A related mechanism in older people is the degeneration of collagen scaffolding in body tissues, the pathway for which is also involved in scar tissue formation. (at least in mice anyway. There was an interesting trial with (SDF1) knockout mice that ended up having flawless skin as a side effect. IIRC, they were investigating scar tissue from cardiac events.) I am curious how much collagen and subcutaneous fat this woman retains in her old age.  The loss of collagen in the skin is the leading reason for paper thin skin and the risk for skin tears. 

[Kagus]

https://www.sciencenews.org/article/3d-printed-shrimp-claw-make-plasma?tgt=more

Scientists think snapping shrimp is a pretty cool dude, then they try to snap just like that cool cat.

Cavitation bubbles are nifty.

[Kagus]

Double posting, with whatever the fucking fuck this fucklical fuckery is supposed to be.

Matter states are just going off the rails these days.

[smjjames]

Double posting, with whatever the fucking fuck this fucklical fuckery is supposed to be.

Matter states are just going off the rails these days.

Sounds like aka a noncoherent gel, or like, a gel that is really runny or something.

[Starver]

It made me think it was a form of material on the edge of a state boundary between a solid phase associated with one form and a liquid phase arising from another (like Ice IV being present in other Ice'N' areas of the pressure/temperature diagram due to insufficient ability to enact the change in packing necessary to conform) such that the matrix of resiliently solid material stands like a gothic crypt ceiling as the counter-matrix of 'meltable' material packed within it is more like a store of gravel shoveled way up to the voussoirs.  (It couldn't leak out, as described, because that would imply the pressure is relieved on the 'solid' as well as the 'liquid' forms.)

Or it's an artefact of the simulation used, somehow.

Personally, I like the idea of the transparent sodium. With the 'minor' issue of needing to maintain the temperatures and pressures required, I've decided that this sounds like much more fun than the aluminium version!

[da_nang]

Firefox is on fire today.

Add-on certification isn't working, so the next time Firefox checks the certification, the add-on will be disabled and possibly removed, since the certification has expired.

Not a good weekend for Mozilla devs.

[methylatedspirit]

From the Mozilla Discourse thread about said issue:

12:50 p.m. UTC / 03:50 a.m. PDT: We rolled-out a fix for release, beta and nightly users on Desktop. The fix will be automatically applied in the background within the next few hours, you don’t need to take active steps.

In order to be able to provide this fix on short notice, we are using the Studies system. You can check if you have studies enabled by going to Firefox Preferences -> Privacy & Security -> Allow Firefox to install and run studies.

You can disable studies again after your add-ons have been re-enabled.

We are working on a general fix that doesn’t need to rely on this and will keep you updated.

[Max™]

I hate everything about the internet right now, fuck everything, goodnight.