It is now easy to edit the genomes of plants, animals and humans
economist.com
economist.com
This is NOT easy. First of all you cannot target any sequence; it must match a certain consensus pattern. Second, while cutting is easy (to delete or knock out a gene), introducing a substitution is hard and requires a lot of chance. Third, the process is extremely noisy. You cannot guarantee your edit will occur. Usually you must verify by resequencing, which is actually worse than shRNA.
Finally we are far away from full body gene therapy this way, which would involve delivering a CRISPR kit to every cell.
In short this is a powerful experimental tool but it is extremely far from freely editing the genome.
My layman's understanding of the situation is that the most likely first use of this in humans will be for personal eugenics of one's offspring. Meaning, removing deleterious genes from, or adding beneficial ones to, your sperm or egg cells in a lab and then implanting an embryo.
My non-layman's prediction is that you are quite incorrect. As others are pointing out, successfully using CRISPER/Cas is hardly as trivial or as easy as the media's hype machine (or university PR offices) would like us to believe. Blatantly obvious ethical issues aside, embryos are hard to tamper with, and the odds of successfully achieving multiple edits to the genome are miniscule.
I think the most likely first application of the CRISPER/Cas system in human medicine will be in an autologous bone marrow transplant to correct a genetic disorder of the blood, likely sickle-cell anemia.
Immunotherapy is likely a clear win for this kind of genome editing technology. Any sort of ex vivo modification and repatriation of cells (like bone marrow transplants) are a good fit.
Personal eugenics is extremely unlikely in the West. It is also very far away because of this regulatory burden. Id just point out that there are cases where you might find this kind of thing permissible. For instance the other day a friend of mine who went through IVF with his wife said "imagine if we had one good embryo from three rounds of IVF, and that one embryo had cystic fibrosis - would you really be comfortable stopping me from repairing that?"
I found it hard to say "yes we should stop that". But I also recognise that if we allow that, its only a small leap to allow anyone to edit anything they want, for whatever reason. Drawing the lines will be tough.
For the deal I mentioned: http://www.xconomy.com/boston/2015/01/07/cart-crispr-1st-of-...
I'm not suggesting tampering with an embryo - rather, editing the sperm or egg cells. I don't think the ethics is going to stop people when there is a chance of giving your offspring a significant advantage. But like I said, this is purely from an "outside" perspective - I don't know much about the field.
Contrast your prediction of elective eugenics for offspring with my own for autologous bone marrow transplants for sickle cell anemia: sickle cell is a single point mutation, there's a lot of tissue that we can sequence and work with, you don't need to transform anywhere near 100% of the tissue to get a good therapeutic effect, and we already do cure sickle cell anemia with bone marrow transplants given the right circumstances. All of that is in part why I think you are completely wrong with your prediction.
What's the current state of the art for using computers to model or simulate how a gene change will impact the rest of an organism?
Separate but related, if you were to collect data on that sort of thing during an experiment, how would you? I assume there's quite a bit of variation depending on experiment goal... Is imaging technology to the point where you can record what happens as video?
Unfortunately, molecular biology and genetics are immensely complex, once you start looking in to them. There very rarely is a 1:1 correlation between one gene and one single function of impact on the entire organism (even for single celled organisms). Moreover, in many cases loss of one gene can be compensated by the function of one or multiple different protein(s).
To be fair, I haven't looked up any literature on computer models of how the change of one a single gene would influence the whole organism. However, that is a very complicated question. The closest analogy I can give you is "are there computer models how changing a resistor/capacitor will impact the quality of the picture of my TV?" Frankly, it's often hard to tell how the (novel) change of a single gene will influence a single cell!
Regarding your second question: the data you want to collect is always related to the question your asking about the gene at hand. "How does loss of gene X impact cognitive capacity" would obviously require a completely different experimental set-up than asking "Which functional role does gene X fulfil in the immune system".
Sorry I can't give a more satisfying answer.
On a slightly related note, this article gives some insight on how biologists think and go about their experimental design and recording what they learn: http://math.arizona.edu/~jwatkins/canabiologistfixaradio.pdf
With regard to your second question, there are many "omics" assays that can now capture e.g. every mRNA, every protein, etc. at a given timepoint but you have to kill the cells. Fluorescence microscopy can work for "movie-like" time resolution on a live cell but you have to label the cellular elements with stains or antibodies and you can only utilize so many of those simultaneously on a single sample (1-5 is typical).
Who? As a 3D software engineer moving into this space, this is extremely relevant to me. :-]
http://www.ncbi.nlm.nih.gov/pubmed/18344982
My active imagination hypothesizes that the people who go to work every day to invent / discover this stuff go to work only with the meager hope of collecting on their 6-figure salaries to provide for their families and pay their children through higher education. Perhaps even a stint at an Ivy League school. Meanwhile empowering large multinational corporations / conglomerates with the enormous wealth of intellectual property that their grandparents would be ashamed of.
The only way out of this trap is to destroy the patents. They add nothing other than concentrating wealth and power. The research will always be done anyway; there will always be people like me.
It's hard for the typical "hard-core capitalist / free-market guy" to understand, but I consistently argue that regardless of where the money flows as long as passionate people are able to sustain a comfortable living studying what they love, there will always be amazing breakthroughs in all fields, regardless of "market forces".
You have ample evidence of what is accomplished without market forces: The Internet, Web, email, Linux, Firefox, Newton, Einstein, almost all science, every political and social idea and movement, everything done by almost any soldier (who give their lives without the pull of market forces), parent, friend, mentor, etc. (EDIT: I forgot NASA and most other explorers!)
The idea that people only act due to market forces and selfishness is obviously absurd, a simplistic ideology used by some to rationalize their own greed.
Total pharma spending on R&D is double that of the NIH. Also, the NIH budget is mostly focused on basic science and much of the work has nothing to do with bringing a new drug to market.
Unless you have a great idea on how to replace $60B+ in annual drug development costs, the current system will stay as it is.
If that is too socialist for you we can merely replace patents with a prize system, where we award people a large sum for developing a drug. This is what pharma largely does anyway (wait for startups then buy them out once they've done the hard stuff) so not much would change. For $1 billion a piece we can get ten drugs a year at huge savings.
I got more ideas, but this suffices to demonstrate how easy this is if you think about it. Patents really are the worst way.
I think there are two challenges with your approaches.
1. The expertise needed to bring a drug to market doesn't currently exist within NIH funded entities. They are trying to create it, but have been having a hell of a time. Discovering new science and bringing a drug to market are two very different things.
2. The prize system has it's own challenges. First off, the prizes would have to be huge. Lipitor sales were over $100B during the 10+ years it was on the market. A $2B prize won't mean much.
3. The biggest challenge with the prize system is how do you judge whether or not someone deserves the prize? For example, if we offer a prize for a new antibiotic for multi-drug resistant bacteria, how do you judge whether or not the new drug met the goals? Evaluating a new drug is not black and white at all. What if it kills the bacteria, but is really toxic? What if it works, but not for everyone? Do they still get the prize?
So, during this window, if you over price it, it isn't better/faster/cheaper. If it is priced competively, people benefit. The only "losers" are those that have to wait 10 years to rip off your research and clone your product and undersell you.
What's the problem here?
Then, start a fund that simply buys the patents and "defends them" because I can get investors a 15% return from legal extortion (as opposed to illegal extortion which is higher risk/higher reward and less scalable.) Those investors will make me rich and give me lots of cool inventor awards and fancy dinners, so it all looks super legit. And make sure to hire lobbyists, because they're pretty good at making sure a good thing doesn't die.
The world is certainly moving more quickly than it used to, and maybe 20 years is too long for patent protection. The alternative to utilization of a patented technology is to keep doing what you have been doing. Many people certainly do. The the patent expires about 10 years (or less) from the time the technology is made available to the market. Is that a problem? Seriously - I don't know. It doesn't seem to be (to me), but maybe the timeline is too long?
It has nothing to do with "protecting the little guy" and in fact protects no one except for non-violent gangsters. You pay them for protection, the courts pay for their protection, and technological progress slows down because engineering around patents becomes a thing.
Look: if you're pro-patents, fine, just realize how the money machine works. In practice, it's about making money and maintaining the power to do so, nothing deeper than that.
What about all the open knowledge and technology, such as most of what's published in scientific journals? Was Einstein's physics research hindered because he didn't patent relativity? Watson and Crick? Norman Borlaug? Think of all the open source computer innovations that have been given away and continue to be (are you excited about Rust?).
This isn't right topic thread for this discussion; I apologize.
If you really think about it, how long do Cell Phones last on the market before they're no longer a threat commercially? If I had to put a number to it I'd say no more than 1 year. After that you better have something new, and hopefully patented to prevent competition.
Now amplify that by 2x, maybe 3x, or 10x. Now perhaps only the patent holder has legal rights to experiment on a given range of problems (given the nature of the patent). I imagine if someone plays their cards right, and hits the "patent lottery", you end up with a "waterfall-effect", or in other words, centuries worth of patents by a single entity with no end in sight.
Perhaps those in the field recognize that biology is just as diverse if not more diverse than most other fields. And my concern could be a red herring. I, personally, have no doubt that the field of biology is enormously diverse, and in most academic settings I'd consider this pretty much a non-issue. I think the over-arching fear is the US has nominated 9 people, none of whom are likely to know a significant amount about genetic engineering, to decide (in essence) the fate of genetic engineering and the "patentability" of specific inventions / discoveries.
What I think the field of genetic engineering may need in its future, is the equivalent of a Linus Torvalds in their field. Someone who will cut through the BS, and donate the overwhelming majority of their work / research to the public domain to prevent or reduce the possibility of some calamitous scenario from unfolding.
I have no doubt that large amounts of biological data will be freely available to the market. And many foundational techniques (such as PCR, and now CRISPR) have quite a long useful lifespan. It isn't as if digital cellular telephony wasn't known before Nokia, just that some vendors (Motorola) misjudged and stuck with analog too long after the cost calculus changed.
You raise an interesting question about waterfall effects. I'll have to ponder this some. In any case, say this does happen - the patent holder gets very rich (fairly or not) and the world gets a constant stream of always better innovative biotech. I don't understand how the world suffers whether the beneficiary of the dollar flow is one entity or a hundred.
I'm certainly not net-negative on my outlook / perspective on human nature. I am, however, one who would promote a "trust but verify" philosophy.
I don't know that I am a "trust but verify" person as much as a "don't solve problems before they occur" person, because: 1) The solution may cause more problems than the problem, 2) We had a different problem than the one expected occur.
If anything, the world seems to be progressing at a pretty good clip and not slowing down. In almost every fundamental field - bio, materials, physics etc. And it does so in the presence of a patent system that may or may not be retarding progress, but if it is, it's not doing a very good job of it.
Things like Uber and Open Source and the like make me inclined towards: "Leave it alone, and see what happens".
I suppose that I am always looking for ulterior motives from the players that you can't see when it comes to massive government/regulatory action. Any large change is sold on a story that sounds about right, but I'm worried that something else is really going on with it. That probably makes me borderline paranoid, but an example would be what's happened with Net Neutrality. We all (mostly) want Net Neutrality - but the regulatory response seemed to be a huge change in regulatory scope. This in response to a problem that really hasn't happened yet, or was being sorted out anyway by the players in the market (Comcast/Netflix). And it leaves the door open for the FCC to weigh in on a lot of things that make me provisionally worried that what we were sold and what we will get aren't necessarily the same thing.
Aaand, now I've gone way (way!) off topic of the thread. Thanks for the discussion.
Didn't most research used to be published openly in scientific journals? Has that changed, or was that a golden age that never existed?
Based on my meager understanding, openly sharing research was one of Bacon's foundations of the Enlightenment and the scientific enterprise.
CRISPR aside, the IP situation in biotech and specifically in synthetic biology (parts-based biology) is not a good one. Currently there exists IP around many small DNA components (you can think of them as genes though this wouldnt be correct in all cases - more like DNA programs).
Currently in biology, if I want to build a composite part, that incorporates many separate parts, each with IP around them, I need to get a license for all of those parts.
The tech industry had a similar problem back in the "golden age" of computer parts development. It is often so expensive and difficult for any innovation to happen to happen under these circumstances, as so many patents and licenses and what not must be obtained, which typically frightens investors from putting cash in small startups. The tech industry has solved this with EDA tools - we need something similar in biotech.
What I'd call a DNA Part or cassette, techs call an "IP" or silicone block IP. They have worked out a system where it appears the big design software houses act as brokers and clearing houses for their users. They aggregate not only the necessary licenses, but also the documentation and design rules for how to use a specific part. Then the design tool compiles all these parts down into something that is sent for synthesis by the foundry. They're really light years ahead of the biotech industry on this.
I recommend this fascinating overview of how intellectual property rights in integrated circuit design have evolved. It would be great to see similar things happen in my industry.
Forgive me if I'm misinformed here, but Monsanto may be a good example of a corporation who has put their GMO patents to good use to gain enormous market share in agriculture.
I'm pretty sure this is why we're only seeing consumer VR now, despite the fact that we had most of this technology in 1995. Patent-wise, it wasn't safe to rip off until now.
This is also (again, I suspect) why the current input situation is so underwhelming: anything exciting is going to reply on technology developed in the last 10 years that the R&D firms are pricing at self-defeatingly high cost.
This assumes the "market" makes a good decision. It's not a God, it's just tool. Like any tool, it works well in some situations and not in others.
One thing it's not good at is allocating resources based on need rather than on wealth. What about the people who need the product but can't afford it? What about the innovations that are delayed, and the innovations based on those innovations, and all the people that would benefit from them?
They can wait 7~10 years, no? Patents are not forever.
Is your theory that billions should be allocated (somehow) to researching a list of things that some people somewhere need with no prospect of recouping that expense? And, if so, of the thousands of things that people think they need, which ones get priority? Bill Gates may be doing some cool things in this direction, but he's not an infinite resource.
For people with many illnesses it means death (literally, "forever"). For others, it means 7-10 years of suffering or debilitating illness.
Regarding your other questions/points, I wonder if I misunderstand them because the answers are very well-known. Sorry if I'm missing something:
> Is your theory that billions should be allocated (somehow) to researching a list of things that some people somewhere need with no prospect of recouping that expense?
The enterprise of science has proceeded for centuries on exactly this basis, and society (via government) has funded it successfully. Much of what you benefit from now was funded by fellow citizens, past and present.
> of the thousands of things that people think they need, which ones get priority?
There is no definitive solution, of course, but this issue has long been addressesed by funding organizations such as the NSF, NIH, etc.
http://www.washingtonpost.com/news/to-your-health/wp/2015/08...
Why? If you ever want this technology to be accessible to the public then someone somewhere is going to have to be making money on it. Drug manufacturing isn't done by colleges or governments.
Also, most biologists aren't making six figures.
1) A ton of contemporary research is funded through the American Cancer Society, Howard Hughes Medical Institute, Herceptin was essentially kicked off through the charitable contributions of the cosmetics company Revlon.
2) Polio was cured through the efforts of a nongovernmental research agency (March of Dimes)
3) For basic research: The Glyn Research Trust was an entity set up to discover the chemiosmotic effect. Few people believed Mitchell, and it is unlikely that the work would have gotten public funding. Not only is the chemiosmotic effect now uncontroversial, but Mitchell won a Nobel Prize for his work.
The entire endowment of Howard Hughes Medical Institute (if the numbers are up-to-date on Wikipedia) is worth ~$19 billion. According to some basic info I've found online it appears March of Dimes averages approximately $200 million per year in donations.
$30+ billion is already alot, but, given the right circumstances, at the drop of a hat a future president might be so bold to ask to redirect even more of the US Federal government's ~$4 trillion annual budget toward those efforts.
>ANIMALS typically have two versions of any given gene stored on two different chomosomes—basically large DNA molecules—and the two versions can have important differences. Offspring normally inherit only one of each pair of chromosomes from each parent, and thus each version of the gene typically gets into only half of them. Technologies like CRISPR make it possible to break this rule with something called a gene drive—a gene that uses gene-editing techniques to copy itself from one chromosome to the other, so that whichever chromosome the offspring inherit they get the same version. The same will then apply to their offspring, too (see diagram).
[0] http://www.economist.com/news/briefing/21661801-giving-bits-...
Edit: and I understand that gene drive could make it easier to introduce biallelic changes, but this isn't really easy because the drive systems themselves are not even 10% efficient at introducing the exact modification which is desired.
In either case a degree of caution is warranted. Which scientists did for genetic engineering, during a period when I was preparing to join them (actually did some E. Coli genetic work in the summer of 1977 in a NSF Summer Science Training Program between my high school sophomore and junior years): https://en.wikipedia.org/wiki/Asilomar_Conference_on_Recombi...
The restrictions resulting from the conference were relaxed over time as appropriate as we gained the relevant knowledge, especially the way in which genes naturally jump around all the time between species. I.e. nature has already tried out a lot of stuff we might want to try, given zillions of organisms and years.
It makes no changes to the delivery mechanism, that is, you still have to make the genomic changes in the cells that matter.
I'm greatly appreciative of CRISPR because it elegantly solves the "specific template problem": by providing a generalized sequence non-specific mechanism, any target sequence can be addressed using DNA synthesis. previously, you would have to engineer a unique protein to locate a specific sequence, and because peoples genomes differ, you'd have to reengineer the protein for individuals.
As a result of this, we must perform time-intensive experiments using model organisms like bacteria, flies, mice, etc.
I'd encourage you to get in touch with us at Desktop Genetics. We are always on the lookout for talented developers with an interest in biology!