Patents are destroying the soul of academic science

The soul of academic science is being destroyed, one patent at a time.

Nowhere is this more evident than in the acrimonious battle between the University of California and The Broad Institute of Harvard and MIT over who owns the rights to commercialize gene and genome editing systems based on the CRISPR  immune system of bacteria. There are a dizzying number of patents involved in this dispute, and many more players staking claims to what has the potential to be billions of dollars in royalties down the road. But the heart of the matter is rather simple.

UC claims it should own broad rights to CRISPR-based gene editing because UC Berkeley’s Jennifer Doudna and colleagues were the first to show how a protein (Cas9) from the bacterium Streptococcus pyogenes could be weaponized to permit the easy editing of DNA. (Full disclosure: I am a professor in Doudna’s department). The Broad counters that they should get the rights to the application of CRISPR-based gene editing in humans and other eukaryotes (which include all animals, fungi and plants – i.e. most of the organisms where there is money to be made) because, they assert, The Broad’s Feng Zhang was the first to demonstrate the use of Cas9 in eukaryotic cells.

Last week a panel of judges of the U.S. Patent Trial and Appeal Board sided with The Broad, finding that the application of CRISPR-based gene editing to humans and other eukaryotes was not an obvious extension of demonstrating the basic utility of the system, and hence constitutes a separate patentable invention.

I encourage you to read the judges’ decision. Far from being a descent into an arcane warren of patent law, as most people seem to expect, this case is very straightforward, resting on the simple question of whether the extension of CRISPR from bacteria and a test tube to human cells would have been expected to work by someone or ordinary skill and experience in the field. I don’t agree with the ruling, but the judges offer a lucid and very accessible account of what was presented to them and how they arrived at their decision.

While on the surface this case seems like a fairly mundane “I invented it first! No I did!” dispute, albeit with unusually large financial stakes, to me it represents something far more important: a battle for the very soul of academic science and the principles upon which it is based.

When I first heard, in 2012, that scientists in the Doudna lab had discovered that the Cas9 protein cuts DNA at a specific point based on instructions in a small piece of RNA, and that they had invented a way to simplify its application, I didn’t give a moment’s thought to patents. Instead I marveled that evolution, through the never ending fight between organisms and the viruses that plague them, had created a protein whose key properties were just what was needed to allow molecular biologist to easily edit the DNA of their favorite organism.

If academic science worked like it should we all would have spent the subsequent five years focused only on figuring out all the cool things we could do with this new toy – and there are a lot of cool things. But where we should have seen nothing but scientific opportunity, many saw dollar signs, and the flurry of CRISPR activity beginning in 2012 has become as much a patent gold rush as a journal of discovery.

The academic quest for patents is no longer the side story. Where once technology licensing staff rushed to secure intellectual property before scientists blab about their work, patents now, in many quarters, dominate the game. Experiments are done to stake out claims, new discoveries are held in secrecy and talks and publication are delayed so as not to interfere with patent claims. This is bad enough. But the most worrying trend has been the willingness of some researchers and research institutions to distort history, demean their colleagues and misrepresent the scientific process to support these efforts.

And while all of academia is complicit, The Broad Institute has taken the game to a new level. In 2015, as the patent fight was heating up, The Broad published a “CRISPR timeline” which defined history as ending with Feng Zhang’s demonstration of CRISPR gene editing in human cells. It also demoted the work from Berkeley to playing second fiddle to the work of Virginijus Siksnys’s group who, conveniently, did not have a competing patent claim.

The Broad set up a website describing their patent claims, which includes the following statement:

In April, 2014, the USPTO granted Patent No. 8,697,359 to Broad Institute, MIT and Dr. Feng Zhang. This Patent (which draws priority from a provisional patent application filed in December 2012) contained successful experiments.

Implying, absurdly, that the Berkeley application did not.

Then there was The Broad Institute Director Eric Lander’s widely derided “Heroes of CRISPR” essay in Cell which further rewrote history. Written under the conceit of giving credit to forgotten scientists, Lander wove a sweeping story of scientists toiling in obscurity until The Broad stepped in an made their work important. Doudna and her close collaborator Emmanuelle Charpentier were once again reduced to bit players in this narrative.

This was all clearly done as part of a public relations strategy to support their patent case, in which the assault on reality continued.

We can agree or disagree whether or not it was obvious that Cas9 would ultimately work in eukaryotic cells. The expert testimony shows that cogent arguments can be made on either side. But The Broad chose not to rely on cogent arguments. They had a trick up their sleeves. They scoured Jennifer Doudna’s public statements about the process of getting CRISPR to work in human cells, and found some where she appeared to be making The Broad’s case for them, which they submitted as evidence for their case and trumpet on their website.

For example, The Broad highlighted Doudna saying she experienced “many frustrations” in getting CRISPR to work in eukaryotic cells. But one can believe that it was obvious that CRISPR would work in eukaryotic cells, and still not expect that it would work the first time someone tried it or that the process would be free or frustration. Because that’s how science works! It is often difficult and frustrating – indeed it almost always is – even when you’re working on something that is obvious. Lander knows this. He was once a scientist. And yet he and The Broad are perfectly happy to misrepresent the scientific process to bolster their legal case.

A second quote featured by The Broad has Doudna saying “it was not known whether such a bacterial system would function in eukaryotic cells.” But this is an absolutely true statement that any good scientist would say even if they believed CRISPR would work in eukaryotic cells. In science something is not known until you demonstrate it. This is what any good scientist should say when they have yet to prove that something is true. By pretending that this is a statement about the patentability of CRISPR in eukaryotic cells, The Broad is once again misrepresenting the scientific process and condemning Doudna for little more than being a careful scientist and speaking honestly about it.

Is this the lesson we really want to learn from CRISPR? That scientists working in fields with commercial potential should never speak honestly about their work and the scientific process? That if they do they will get screwed over by someone unscrupulous who prioritizes winning patents and trains their scientists to behave like clandestine operatives rather than the public servants they really are?

By making the scientific process itself party to their legal case, Lander and The Broad are doing more than just securing victory in court; they are willfully undermining science for personal and institutional gain. If academics – including one of the most prominent academic scientists in the world – are willing to lie to promote their own and their institution’s financial interests, why should anyone believe anything they say? If there’s one thing that’s more dangerous than fake news, it’s fake science.

And it’s not just The Broad. While in this case UC’s defense of their CRISPR intellectual property could rely on a truthful account of its discovery, I have no doubt that they would be willing to resort to unsavory tactics and falsehood to secure victory (see their history of trying to coverup cases of sexual misconduct).

Sadly, there will always be venal and weak people in science – it is, after all, a human endeavor. But we do not need to feed them. As we decry The Broad’s behavior, we have to recognize its source – the transformation of academic science from an engine of discovery into a source of institutional and personal riches. And there is a simple way to reverse this trend: deny academic institutions intellectual property in their research and inventions.

Academic science is, after all, largely funded by the public. By all rights discoveries made on with public funds should belong to the public. And not too long ago they did. But legislation passed in 1980 – the Bayh-Dole Act – gave universities the right to claim patents on inventions made by their researchers on the public dime. Prior to 1980 these patents belonged to the federal government and many languished unused. The logic of Bayh-Dole was that, if they owned patents in their work, universities and other grantees would be incentivized to have their inventions turned into products, thereby benefiting the public.

But this is not how things worked out. Encouraged by a small number of patents that made huge sums, universities developed massive infrastructure to profit from their researchers. Not only do they spend millions on patents, they’ve turned every interaction scientists have with each other into an intellectual property transaction. Everything I get from or send to a colleague at another academic institution involves a complex legal agreement whose purpose is not to promote science but to protect the university’s ability to profit from hypothetical inventions that might arise from scientists doing what we’re supposed to do – share our work with each other.

And the idea that this system promotes the transformation of inventions made with public funding into products is laughable. CRISPR is a perfect case in point. The patent battle between UC and The Broad is likely to last for years. Meanwhile companies interested in actually developing CRISPR into new products are stymied by a combination of a lack of clarity about with whom to negotiate, and universities being difficult negotiating partners.

It would be so much easier if the US government simply placed all work arising from federal dollars into the public domain. We have a robust science and technology industry ready to exploit new ideas, and entrepreneurs and venture capitalists eager to fill in where existing companies are uninterested. Taxpayers would benefit by allowing the market, and not university licensing offices, to decide whose ideas and products make the best use of publicly funded inventions.

And most importantly we all would benefit returning academic science to its roots in basic discovery oriented research. We see with CRISPR the toxic effects of turning academic institutions into money hungry hawkers of intellectual property. Pursuit of patent riches has transformed The Broad Institute, which houses some of the most talented scientists working today, into a prominent purveyor of calumny.

We have to fix this problem now or there will be countless other Jennifer Doudnas slimed by colleagues, their contributions to science attacked not for their validity or importance, but for their impact on some other institutions patent portfolio. The soul of academic science is at stake.

This entry was posted in Berkeley, CRISPR, intellectual property. Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. Jeffrey Boore
    Posted February 21, 2017 at 9:28 am | Permalink

    I doubt that denying patent rights to universities would be sufficient for that desired effect. Wouldn’t there be simple ways to work around that, like the university setting up a for-profit spin-off company with itself as the major shareholder that would, then, file the patent application? Of course, there may be ways to legislate against that, but I’d bet that would create a legal war of varying corporate structures to try to evade the laws and protect the intellectual property that would go on for many years.

    • Posted February 22, 2017 at 4:05 am | Permalink

      One problem is that researchers are “double-dipping” — first taking public money to perform work, then privatizing the profit. This could be limited to some extend — for instance, grant applications state what experiments would be performed, and that could be considered evidence of public funding for the invention. Similarly, grant progress reports include publications funded by the grant, and there would also be records of which researchers had their salaries paid by the grant.
      As for the “ownership” issue, the most straightforward circumventions (like a unviresity-owned company) would be pretty obvious. In the end, the university may be so far removed from the control and profits that they could not profitably apply any sort of pressure or incentive on the researcher to restrict publication of results.

  2. Posted February 21, 2017 at 9:45 am | Permalink

    You are half way there. Don’t just eliminate patents for Academics. Eliminate them for all.

    • Posted February 22, 2017 at 4:25 am | Permalink

      Monopoly protections corrupt everything they touch — universities, the government, and even commercial corporations.

  3. Fair Miles
    Posted February 21, 2017 at 10:24 am | Permalink

    As seen elsewhere, “So wrong. Tremendous harm to science. Sad!”

    (sorry, just trying humor against depression; I would rather go for pills, but they are so patent encumbered…)

  4. Will Robertson
    Posted February 21, 2017 at 12:16 pm | Permalink

    I think it’s time to call out the behaviour of universities and institutes. Such fine mission statements on using science to benefit humanity but behaviour that is focused on enrichment and entrenches the inequitable access to it’s benefits. It straight out bs and they should be held to account for the discordance.
    Scientists should be inspired by Salk rather than Shkreli. Now is a good time to shift the paradigm.

  5. Posted February 21, 2017 at 12:52 pm | Permalink

    I agree with much of what you have written, but I don’t think either of these groups deserve this patent. It would be analogous to awarding the Nobel Prize to the first person to use RNAi to inhibit gene expression in a mouse, though the principles were all worked out in worms and plants (actually I don’t know who has the patents on RNAi; maybe the mouse guys have it). But at least in that case, whether mammals had the machinery to use RNAi had to be discovered. This is not the case for Cas; we already knew the machinery was there.

    It was well-established in the early 1980s (Rod Rothstein and Jack Szostak’s labs) that a site-specific double-strand break would direct homologous recombination in a eukaryote (budding yeast). Following the demonstration that HO endonuclease promoted homologous recombination by a site-specific cleavage of DNA (Kostriken et al. 1983) and that the enzyme could be induced from a conditional promoter (Jensen and Herskowitz, 1984), my lab showed that we could initiate efficient homologous recombination between bacterial LacZ sequences cloned into yeast (Rudin et al, 1989). The first use of the site-specific I-SceI endonuclease to promote homologous recombination by cleaving a chromosomal site in a eukaryotic nucleus was done by Bernard Dujon’s lab (and myself) in 1992. By 1994 Dujon’s lab and Maria Jasin had used I-SceI to create both gene modifications using a template sequence and insertion/deletion mutations. So the idea that a site-specific cut in a chromosome could be used in eukaryotes for efficient gene editing was known long before Feng Zhang showed this to be the case for Cas9.

    But of course, the I-SceI enzyme cut only one or a very limited number of sequences. To have an easily programmable system to make cuts at any chosen location was devoutly to be wished. First were zinc-finger nucleases (ZFNs) in 2001 (Bibikova et al). TAL effector nucleases (TALENs) emerged as the next big thing (Christian et al, 2010). Though a bit cumbersome, these proteins combined a nuclease “head” with a modular protein body that could be assembled to bind to specific DNA sequences. So the idea that one could devise a programmable targeting system that would allow both indels and gene editing in eukaryotes also wasn’t new.

    Enter CRISPR/Cas. What made all the difference was that CRISPR/Cas had “the potential to specify RNA-guided, sequence-specific DNA cleavage via simple Watson–Crick pairing rules, rather than through complex protein–DNA interactions,” (Sontheimer and Marraffini, 2010). Seems to me that, by this time, anyone who had been following the field knew from this statement and from the demonstrations by Marraffini and Sontheimer (2010) and from Moreau’s group of collaborators (Garneau et al. 2010) that these RNA-protein complexes indeed cleaved DNA site-specifically, that the applications to gene targeting in eukaryotes were obvious. There were other important papers, but my point is that by 2010 anyone who was following this story and working in DNA repair would have concluded that this approach was going to work (apparently the judges didn’t agree).

    In 2012, both Siksnys and Charpentier/Doudna published key papers, the latter simplifying the system to a two-component complex that has been used by probably thousands of labs. By then it seemed even more obvious: “Our study reveals a family of endonucleases that use dual-RNAs for site-specific DNA cleavage and highlights the potential to exploit the system for RNA-programmable genome editing” (Charpentier/Doudna). Moreover, Siksnys noted that their “Experimental demonstration that RuvC and HNH active-site mutants of Cas9 are functional as strand-specific nicking enzymes opens the possibility of generating programmed DNA single-strand breaks de novo. Taken together, these findings pave the way for the development of unique molecular tools for RNA-directed DNA surgery.” Doudna and Chapentier’s paper provided similar evidence for nicking. Without question, Cas9 has revolutionized gene editing in eukaryotes.

    But I still don’t know why the earlier publications showing CRISPR/Cas site-specific DNA cleavage in vivo (but not in eukaryotes) did not satisfy the requirement for a demonstration of prior art, that someone with ordinary skill and experience would have expected such experiments to work. All one would need would be the eukaryotic-specific promoters to express the protein and RNA(s) and a standard amount of hard work to optimize the system. And that’s what happened.

    That being ranted, I fully agree with your broader concerns of how patent-fixation distorts the scientific enterprise and poisons relationships.

    • Steve Elledge
      Posted February 21, 2017 at 5:47 pm | Permalink

      Great description of the history of HR in eukaryotes. I might add there was also a restriction enzyme mediated mutagenesis method developed by Adam Kuspa in 1992.
      They used EcoR1 but any enzyme would have works and no doubt also resulted in NHEJ mutations.
      Kuspa, A. and Loomis, W. F. (1992) Tagging developmental genes in Dictyostelium by restriction enzyme-mediated integration of plasmid DNA. Proc. Natl. Acad. Sci. USA 89, 8803-8807.

  6. Dana Carroll
    Posted February 21, 2017 at 7:04 pm | Permalink

    Jim’s penultimate paragraph summarizes the key argument that the UCB/Charpentier legal team (and Carol Greider and I as an expert witnesses) used in pursuing the claim of interference by the Broad with UCB/Charpentier patent applications. The patent office board chose instead to rely on a small number of statements, made in 2012 prior to the actual demonstration of CRISPR utility in eukaryotic cells, that pointed out that the experiments had yet to be done. This did not mean that researchers in the area doubted that those experiments would succeed. In fact, several groups jumped in immediately with the obvious expectation of success, using completely standard methods.

    It is indeed sad, as you say, Mike, that so much time, effort, money and spleen has been spent on what should be open access to this powerful technology. Sadder still, even the case of applications to eukaryotes is not closed. Berkeley/Charpentier may appeal, and there are other groups that have plausible claims to have preceded Zhang/Broad in demonstrating use in human cells.

  7. Gerald Barnett
    Posted February 21, 2017 at 10:38 pm | Permalink

    Great article and comments.

    Despite popular accounts of Bayh-Dole, there’s no good evidence that prior to Bayh-Dole government-owned patents were not used. It’s just that most were not licensed exclusively for commercial development. Many were defense-related inventions which defense contractors had passed up the opportunity to patent and for which there was no market other than selling to the government (see the work by David Mowery and Rebecca Eisenberg). There was no point in commercial licensing, then, nor for any one contractor to have an exclusive position with regard to technology developed at the federal government’s request and for its use.

    In the developing area of medicinal chemistry, in the early 1960s, the pharmaceutical industry staged a boycott of compounds with potential therapeutic effects because the Department of Health, Education, and Welfare as a matter of agency policy refused to give companies monopolies in these compounds. Clearly, those inventive compounds were not being used by the pharmaceutical industry–but the matter not the presence of a government (or university) patent rather the lack of a private monopoly to control development and maximize income.

    From 1968 to 1978, NIH operated an “Institutional Patent Agreement” program that permitted nonprofits to sign a master agreement under which they could own inventions made with agency funding, provided the institutions decided to patent the inventions and make the inventions available to industry. The IPA program was an end-run around HEW policy and with a lot of hand-waving about non-exclusive licensing and limits on exclusive licenses allowed universities to set up shop to feed NIH-funded inventions to the pharmaceutical industry as monopolies, getting on the order of 1% royalty for their role in circumventing public policy. After a decade of effort, universities reported a 4% “commercialization” rate–about the same rate they derided the federal government for. But the federal government wasn’t even trying to “commercialize” most patents it held, and for many more inventions that it had rights to, it chose to “dedicate” (to use the term from the Kennedy patent policy of 1963) these inventions to the public.

    HEW reviewed the IPA program and found that universities always granted exclusive licenses that ran against HEW policy. So HEW shut down the IPA program and a Senate committee blocked an NIH effort to make the IPA program government-wide. Bayh-Dole was the patent brokers’ response (led by NIH’s patent counsel) to overturn HEW’s policy for making university inventions made with federal support broadly available.

    The Bayh-Dole effort was to overturn executive branch patent policy by statute and make the effect government-wide, so as to hide the primary purpose, which was to restore the monopoly-making pathway worked out in the IPA program for the pharmaceutical industry. The behavior of Harvard’s patent administrators is just another manifestation of the past thirty-five years of university complicity in this grand bargain made at the expense of science, the public domain, and public interest.

    There is nothing about the CRISPR technology that requires patents before it can be “developed” for commercial use. Patents here reveal nothing that is not already published. Nor are patents needed to “call forth risk capital” that otherwise would not be available. Patents here permit a university to shake down industry, introducing complexity, delay, and expense. The sad thing is that UC and Harvard’s dispute is over who should get the right to conduct this shake down and pocket the gold from doing so. Neither institution, as far as I am aware, is proposing to license CRISPR royalty-free or to secure rights to ensure no-one is blocks further development of the platform while enjoying the benefit of access.

  8. Ashley Stevens
    Posted February 22, 2017 at 7:27 am | Permalink

    You totally understand that there are two parallel tracks at work here:
    * an academic track, where the results are freely disseminated and available to all to build on; and
    * a commercialization track, where scientific results are turned into intellectual property that has normal property rights that can be licensed or sold and turned into products that help people

    • Posted February 22, 2017 at 7:31 am | Permalink

      I understand how the system works. My point is that it is not working. The commercialization track is not dominating the academic track, and the way in which universities are commercializing things is an obstacle to them being turned into products that help people.

  9. Anand Ray
    Posted February 23, 2017 at 4:50 pm | Permalink

    Interesting viewpoint but lacks insight about the value of patents to scientific progress. There is a huge valley of death between the bench and the shelf. Without patent protection and exclusivity most discoveries will be lying by the wayside with no investors or entrepreneurs willing to take the risk. As it is even with patent protections most startups or new development projects fail due to enormous risks. It has been my experience that obtaining grants for university basic research from NIH, NSF is much much easier even in these hard times than finding investments for most entrepreneurial efforts.

    • Posted February 23, 2017 at 8:23 pm | Permalink

      You make these assertions without evidence. It may be true that some product ideas do not receive investment, but it does not follow that if they did they would have made good products or successful businesses.

      • Anand
        Posted February 23, 2017 at 9:59 pm | Permalink

        I agree that not all product ideas are good. But the question here is how much more difficult would it be for a good product idea to become successful without IP protection. Well there are many successful contemporary biotech companies out there big and small, some perhaps started up by faculty or postdocs in Northern California Universities. Just ask them how many would have ever taken off without some IP protection, or business secret or other barrier to entry. Based on my experience even non – profit foundations like the Gates Foundation want their inventors to get IP. Even if it were for low cost products in Africa.

  10. Kaleberg
    Posted February 23, 2017 at 8:17 pm | Permalink

    John von Neumann not only invented the modern digital computer, he also published to pre-empt any patents of his ideas. Eckert and Mauchly, who later founded Univac, were pissed, but thanks to JvN’s quick action we have a modern computer industry today.

    Patents are just as likely to delay technology as to advance it. Look at the Wright Brothers with their incredibly broad airplane patent that effectively kept the US well behind in the aviation industry into the 1920s.

  11. Posted February 26, 2017 at 2:29 pm | Permalink

    This is a very rich post, and set of comments by highly knowledgeable commentators.

    Just a few points about the history.

    The judges here did not have a lot of leeway, given patent case law. It might, as a matter of science, and as Eisen, Dana and Greider (as well as the UC lawyers) be true that it was both “obvious to try” and likely to work in eukaryotes. The problem is that the patent document needs to describe the invention in sufficient detail that someone of ordinary skill can make and use the invention. This is usually quite close to the intuitive meaning of “possession,” meaning if you want to claim exclusive rights to something, then you have to describe it in detail, which means you have to have done the work. Doudna and Charpentier did incredible work, and reallly focused attention on CRISPR/Cas9, and I hope they get the Nobel for it. But that’s now what the judges were faced with. The UC folks could probably have won the race to eukaryotic use, which was foreseeably where the $ would be made, but it’s certainly understandable that they did not, since that was not their core interest.

    But the mistake here, in my view, was UC’s very broad claim language, combined with a decision (by Broad and UC and U Vienna) to exclusively license therapeutics and several other uses to startups. That’s really a problem for a broadly applicable platform technology, since the startups will quite predictably (and with good reason) focus on high-value, quick-return applications. But this technology is vast in its scope of applications and many applications won’t be terribly lucrative, but might be incredibly important for environment, agriculture, public health and other domains where returns are much lower than for human therapeutics.

    In contrast to Stanford’s licensing of Cohen-Boyer, which was nonexclusive licensing on very reasonable terms to as many folks as would take licenses, and Columbia’s licensing of the Wigler-Axel patents (which NIH compelled to be nonexclusive, despite Columbia’s request to exclusively license to a single firm; but it was pre-Bayh-Dole). Both UC and Broad have, instead, opted for exclusive licensing of therapeutics to startups. Moreover, they decided to go to war rather than come up with a transparent and open licensing template like Cohen-Boyer. That’s too bad, and probably even a financial mistake, since most of the $ from this technology will be a decade or more into the future, and will come from products and services that simply cannot be predicted and probably only a tiny fraction of which will come from any of the licensed startup firms. There will be sub-licensing but it will be slower, more complicated, and less extensive than it could have been with a transparent cross-licensing framework.

    Denying patents to universities is not a very good solution. Patents do serve useful purposes in attracting private investment in R&D. And they also package rights that can allow commercialization, when commercialization is a good way to get something into the world quickly. It’s not simple “patents good” or “patents bad,” but rather knowing what you’re doing and having a strategy. It’s hard to argue for CRISPR that we have such a strategy, and this may be as case where patents could hinder as much as help; time will tell (and maybe we’ll never know). But for sure, this case argues for doing a lot better job of tracking what happens when patents derive from public (government and nonprofit) funding. Strengthening the rules and conditions under which federally funded inventions are licensed, and increasing awareness of how exclusive licensing can be abused, esp. for platform technologies, would be a better policy choice. But as Ashley Stevens notes, patents really do induce private investment. Startups really are important in the innovation ecosystem; just not a great choice to give broad rights for a broadly useful technology.

    As Eisen notes, the chilling effect on scientific communication really is a big deal. And it truly would be tragic if this case leads to university licensing offices muzzling their scientists, which is a foreseeable outcome. But my take on that problem is somewhat different. It would not be to counsel scientists not to be honest about uncertainty, but rather to constrain patent claims to work that has actually been done. That is, be more careful about what you claim in patents so it does not conflict with what the scientists will say, and especially stop claiming broad applications that have not been demonstrated in practice. And exclusively license with a product or service is on the horizon that demands substantial investment, but not when applications are many and far off.

    The claiming and patenting strategy that UC pursued got them into trouble, and my reading of the case law is that the unanimous decision of the PTAB three-judge panel is unlikely to be overturned, and even if it were, the Supreme Court might well step in to reverse the Court of Appeals for the Federal Circuit, since UC’s main claim is just the kind of very broad claim that covers broad expanses of science that has driven the Supreme Court’s quartet of (unanimous) decisions about patentable subject matter, reining in the appeals court and patent office. UC was playing the patent game by the old rules of Monopoly, but the rules have changed, and generally for the better.

    UC blew it and shot themselves in the foot. The problem was not in the science, because Doudna and Charpentier clearly did fabulous work, but the legal strategy that led to a decision that left UC worse off than they were in 2012, having caused tens of millions of dollars of investors’ money to be spent on paperwork instead of science and delayed getting needed certainty about who should license what from whom, when in fact the most important thing is to get this technology into as many hands with as few restrictions as possible. At least Broad has a website. Just try to find out how UC plans to license. Even Jon Cohn can’t figure it all out, despite asking lots of questions and following the story closely.

    • Arti Rai
      Posted February 28, 2017 at 10:33 am | Permalink

      As usual, Bob Cook-Deegan has done an excellent job of identifying a key issue that is often overlooked. Over the last 20 years or so, the patent law rules have indeed changed, so as to require claiming only of what one has done, not one plans to do. Interestingly the key requirement policing narrow claiming, written description, first emerged in a 1997 case also involving UC. (UC v. Eli Lilly) Since 1997, as the requirement has gained broader acceptance, it has also evolved, so that it is bit less formulaic. In the main, however, written description frowns upon claiming anything until one has done it.

  12. Posted March 3, 2017 at 12:04 am | Permalink

    It would be interesting to chart parallel history from 1970’s with and without public research funding patents (from restriction cloning onwards). One clue could be in the monoclonal antibody case – no patent from Cesare Milstein, everybody could use it, public and private, with great benefits.

    One little genetic example: we wanted to include a SNP in a panel, the university wanted $250,000 up front and 5% of the price for royalty. They would not budge an inch – this was for one SNP on a multigene panel, it was completely uneconomical so we couldn’t use it. All the work had been publically funded.

    One big genetic example – BRCA: a) how may women, worldwide, have died because it was not possible to screen them, b) if there were no patents would the screening technology have developed in the same way. My answers a) undetermined, probably lots b) yes, and how, it would have developed much more quickly, it would now be a $50 test, if that

  13. Jonathon Keeney
    Posted March 6, 2017 at 9:32 am | Permalink

    You raise some excellent points. Some of which have a little similarity to what I’ve written about regarding the differences between “blue skies” research and for-profit research (

    But I disagree that patents are the cause of all this trouble. Full disclosure: I occasionally serve as a Technical Specialist for a law firm (meaning I draft biomedical patents for them).

    From my perspective, the cause is clear: competition. People like me had planned to make their entire careers in academic science but were ultimately scared away from it because of the state of the field. When this happens en masse, academic science is enriched for individuals who thrive on aggressive competition.

    Indeed, I think that this is exactly the “worrisome consequences” of a hypercompetitive culture that Alberts et al. warned of ( As graduate programs are incentivized to move warm bodies through training programs and further compound competition for dollars, the individuals who succeed are bound to have a more aggressive stance when it comes to funding. The merits of a research proposal are simply not enough anymore. We are quite literally selecting for an aggressive attitude.

    This often can mean prioritizing matters of career over matters of scientific curiosity. That’s a very different state of mind, and often means a very different kind of person that’s amenable to that state of mind.

    All of this drives a deep malaise. The desire to understand the Universe might just be our way of coping with the human condition. Those of us who can’t help but to think like this are bonded together by this spirit of inquiry, despite our differences of opinion. Increasingly, the spirit of competition seems to be displacing that attitude and erecting barriers.

    On the other side of all this, people like myself are trying to find other ways to be involved in research outside of academia. With the incentive to be a professor rapidly drying up and the academic system blasting through a Malthusian limit with no signs of slowing, the private sector seems to be the only other option.

    The fundamental concept behind a patent is to incentivize people to freely publish their ideas in exchange for a limited monopoly. You don’t need to enforce the patent, but at least you’ve got something to build value around. It’s a different way of fund raising. As someone who’s started a business in the private sector in the hopes of staying involved in biomedical research, I see no other realistic option.

    It may also be competitive in the private sector, but at least your publications — patents — have explicit value. You can quite literally turn them into funding if they’re strong ideas and well written. It may not be the wide open, blue skies sort of academic paper I would love to publish, but it is an overt acceptance of the state of science, and its inextricable link to funding. Much like the evolution of a new trait, it isn’t ideal, but it works.

    I also realize that there’s another dimension: with mass production comes reduced quality. As the number of graduates increases, it’s tempting to think that only the most talented people are the ones sticking around in academic science, and that folks like myself are nothing more than a case of sour grapes. But the truth is that many motivated, hard working, talented individuals (including those of us who don’t care too much about “long hours” and little pay at this stage) are leaving science — just not the cutthroat competitive individuals that thrive in an alpha system. That was once the domain of business or law, but is now forcing individuals like myself who have a background in science to chart new courses if we hope to be involved in research.

    There’s certainly nothing wrong with some competition, but it’s not a great environment for the sort of blue skies thinking you describe. Excessive pressure doesn’t encourage creativity. Indeed, this is the point of the tenure system: psychological safety to pursue anything. This is where the best ideas come from.

    I’m not pointing any fingers with this, I realize it’s more complex and nuanced than can be captured in a quick dialogue, and that there are also external factors (e.g. public funding). I just have a hard time believing this is anything other than exactly what we’ve been headed towards for years now.

    Just like any other species starved of resources trying to find a new niche, novel strategies and overtly aggressive behaviors should be expected. Put simply, although you make very valid points, patents are the symptom of the current academic climate, not the cause.

  14. Posted March 19, 2017 at 3:29 pm | Permalink

    For a general discussion of the interest of having patents in the first place, see the free ebook ‘Against intellectual monopoly’ by the economists Boldrin and Levine.

    Their conclusion is that patents are harmful. I am not aware of any refutation of their arguments, whereas they of course do refute the standard arguments in favour of patents.

5 Trackbacks

  • By SynBio Newsreel, February 2017 | GeneMods on February 24, 2017 at 1:24 pm

    […] in vitro and in bacteria. Berkeley professor and soon-to-be Senate candidate Michael Eisen wrote a scathing critique of the decision, and of University patent rights […]

  • […] Patents are destroying the soul of academic science […]

  • By CRISPR and Bayh-Dole | Research Enterprise on February 26, 2017 at 1:06 pm

    […] Eisen has an insightful discussion on his It is NOT Junk site on the CRISPR patent dispute between the University of California and the […]

  • […] Science that is publicly funded is patentable by universities, but it hasn’t always been that way. Following last week’s patent ruling on CRISPR-Cas9, Michael Eisen, a geneticist running for Senate, argues “The soul of academic science is being destroyed, one patent at a time” (Michael Eisen). […]

  • […] academic IP“), reports on a talk given by Michael Eisen a few days ago. Eisen has published an article at his blog on the patent fight between UC Berkeley and The Broad Institute (and its backers MIT and Harvard). […]