Synthetic Society/Ownership, sharing and innovation
See also, Frank's notes. [editorial note: may want to merge them with this material.]
These notes were taken by Ken Oye and edited/posted by Reshma.
- Create incentives to innovate
- Limit restrictions on diffusion of fruits of innovation
- Minimize conflicting claims that wreck both incentives and innovation
- Maintain health of field
- promote learning
- avoid (premature) enclosure
- answer the question of whether current IPR laws and TLOs will impede progress in future?
What is patentable and/or copyrightable?
- Broad biological functions
- Specific sequences
- Specific uses
Answering this question requires an inventory of what has been patented, what has been challenged and what are the outcomes of adjudication. Since the record in synthetic biology is incomplete/nonexistent, we will need information on other areas that may frame decisions here like pharma, electronics, biologicals, software.
What are the positions of existing companies in synthetic biology on the IPR issues? How do they think they will play out?
- Sri will check with people at Duke Center for the Public Domain, Science Commons and the Public Patent Foundation.
- Gautam will check with venture capitalists.
[editorial note: the material below is merely intended to provide substrate for discussion. In all likelihood it contains errors. Please feel free to revise, edit, delete as you see fit.]
The questions below stem from the premise that in engineering industries (like aircraft parts) in which the technology is well-understood, intellectual property rights (IPR) are readily circumvented. Therefore, it is more worthwhile to share IPR rather than to enforce your own IPR. In the pharmaceutical industry, the technology is not well-understood and so in some sense, your IPR is worth more because it is far more difficult (and possibly impossible) to circumvent. The question is to which industry is synthetic biology more similar? Currently, we may be in a place where synthetic biology is more similar to the pharmaceutical industry but one could imagine that as the field matures, it might become more similar to the engineering industries.
What are effects of alternate definitions of what is patentable and copywritable on
- development of field?
Technical and economic issues
What are the costs of innovation?
[disclaimer: I am not sure if this is the type of information needed for this discussion or whether these answers are even "right" or not]
Currently, the costs of innovation in synthetic biology are quite high because
- There are a lot of fundamental problems in synthetic biology that aren't solved or even studied in a sytematic way. Many of the research projects underway at MIT and elsewhere are designed to address these issues. However, one problem we face is that many of the critical pieces of work are not necessarily appropriate as a student research project. It is not clear whether they can be solved in an academic setting or whether they need to be addressed elsewhere ... be it a company, a nonprofit, a government lab or something like the Broad Institute. A key question is how to make it worthwhile for people to tackle these basic problems and how to encourage them to allow others to make use of their work.
- It is hard to fabricate synthetic biology systems right now but this is expected to change soon. For genetically encoded systems, it is simply either time-consuming or costly to make several versions of the same system which is critical to finding one that works. Molecular cloning ... the process by which most genetically encoded systems are currently made ... takes a lot of time and effort. DNA synthesis, the alternative, is still sufficiently expensive to not be routine. However, most expect that this problem should be solved soon by improvements in DNA synthesis technology. Some predict it will be solved in the next 5 years: see Rob Carlson's article on The pace and proliferation of biological technologies.
Viability of workarounds and substitutes?
The ease with which a particular patent can be circumvented depends on how broad that patent is. I have no idea whether the following kinds of patents could or have been awarded (or are enforceable), I just enumerate them here to illustrate different scenarios. It is probably useful to try and find different example patents for each of these categories (if they exist) to better ground the discussion in reality. I've put some initial links to existing patents that might fall under a particular category (they need to be read more carefully to see exactly how broad they are). Of course, just because these patents were issued, it does not mean that they were or will necessarily be enforced by the assignee or would be upheld if enforced. Please add more types of patents as you think of them.
- Patents on fundamental ideas in synthetic biology
Example: A patent on the idea of a biological part: a piece of DNA with specific function that can be combined with another part in a predefined fashion.
Such a patent would be impossible to circumvent. It represents a fundamental concept that underpins synthetic biology.
- See Stanford patent on System and method for simulating operation of biochemical systems. United States Patent 5914891
- Patents on fundamental biologial functions
Example: A patent on a genetically-encoded inverter (i.e. a device that takes an input signal and produces and inverted output signal encoded in DNA).
Such a patent would be almost impossible to circumvent because it represents a basic biological function that is of use in a range of synthetic biological systems.
- See US Dept of Health patent on Molecular computing elements, gates and flip-flops. United States Patent 6774222
- See Boston University patent on Multi-state genetic oscillator. United States Patent 6737269
- See Boston University patent on Bistable genetic toggle switch. United States Patent 6841376
- See Boston University parent on Adjustable threshold switch. United States Patent 6828140
- Patents on classes of biological molecules with a particular function
Example: A patent on the use of zinc finger proteins to bind a specific sequence of DNA.
Zinc fingers proteins are a family of proteins found in nature that are known to bind specific target sequences of DNA with high affinity and specificity. They have been the subject of study for many years and work has been done to engineer non-natural variants of zinc fingers that bind to nearly every possible target DNA sequence. Hence, they represent a very useful family of proteins for use in synthetic biological systems. Such a patent is not impossible to circumvent because there are other proteins that bind DNA and that could be engineered to bind new sequences. However, due to the previous effort that has been invested in zinc finger proteins, it would require a considerable investment of time and money to find a substitute solution. Additionally, zinc finger proteins are widely acknowledged to represent the most elegant solution to the general problem of having proteins that bind to every possible DNA sequence and thus circumventing such a patent might involve having to pursue a "second-rate" solution.
- See MIT patent on Poly zinc finger proteins with improved linkers. United States Patent 6903185
- See Scripps Research Institute patent on Zinc finger binding domains for GNN. United States Patent 6610512
- See Sangamo Biosciences, Inc. patent on Regulation of endogenous gene expression in cells using zinc finger proteins. United States Patent 6607882
- Patent on a particular biological molecule.
Example: A patent on the sequence of a particular protein that senses light and transmits a signal into the cell.
Such a patent would likely be fairly easy to circumvent because there are probably a few amino acids that could be changed in the protein such that it would it would still be functional yet not have the exact same sequence as specified in the patent. Note that there are of course exceptions to this rule: there are some proteins that have been so optimized for a specific function that any mutation in the sequence supposedly leads to less functionality (for instance, the drug Ziconitide which is a peptide). However, such proteins are reasonably rare exceptions.
Nearterm and medium term uses and applications of synthetic biology?
These are examples of ongoing research that have (mostly) obvious commercial applications. Such applications are likely to be realized on approximately 5 year timescale.
- Using synthetic biological systems to study existing naturally occurring biological systems.
- A genetically encoded counter to count cell division events.
- A genetically encoded oscillator to provide in vivo oscillatory input signal to a particular pathway.
- Such a system could be marketed by biotech companies to academic and industrial researchers. For example, Invitrogen could sell a cell cycle counter to anyone interested in studying cell division and replication. In all likelihood, this system would be vulnerable to hacks: i.e. individual researchers sequencing and manufacturing the system on their own to avoid paying the (usually) high costs of buying such "kits" from companies.
- Using synthetic biological systems to manufacture compounds.
- Jay Keasling's work on an antimalarial precursor
- Kristala Jones Prather's work engineering bacterial to produce nonnatural compounds by constructing new metabolic pathways either by engineering new enzymes or putting together existing enzymes in a novel way. research summary
- Such systems can be used by companies like Dupont to provide either environmentally friendly or cheaper means for synthesizing valuable chemical compounds. Such processes are likely to remain trade secret?
- Using engineered bacteria for sensing
- Chris Voigt's work on engineering bacteria to target cancer cells. This is useful for delivering things to a particular location. pubmed
- Homme Hellinga's work on engineering bacteria to detect low level explosives like TNT.
- Such systems have medical applications since they would enable targeted treatment of diseases potentially leading to lower side effects. These systems would probably require significant investment of resources to survive clinical testing. Systems used to monitor environmental conditions and check for toxins and/or particular DNA sequences would be useful to the military.
- Using biological systems to assemble and/or pattern materials
- Such systems enable manufacturing of small-scale structures.
Legal and technical issues
Are we in a potential anticommons situation? stovepiping?
Legal and risk assessment
Liability banes that may come with property rights boons?
- Reshma and Jason and Frank will begin to identify knowns and unknowns on this.
- What strategies to change definitions of patentable?
- What strategies to clarify what the definitions are?
- What strategies GIVEN definitions of patentable?
- Sri will sort previously brainstormed strategies into relevant subcategories.
- Drew Endy's talk at OSCON 2005
- Meeting notes from Nov 2005 SynBio conference at Duke Law's Center for Study of the Public Domain [available, email firstname.lastname@example.org]
- Innovation & Incentives by Suzanne Scotchmer.
- Notes & FAQ from the pre-historic days of the BioBricks Foundation.
- Arti Rai and James Boyle article on legal status of Synthetic Biology [draft pending, email email@example.com]
- 17.310J class materials
- Arti Rai's Publications
- Rai, Arti K. and Eisenberg, Rebecca S. (2003) Bayh-Dole Reform and the Progress of Biomedicine. Law & Contemporary Problems 66(WtrSpr):289-314. Link
- Rai, Arti K. (2005) Proprietary Rights and Collective Action: The Case of Biotechnology Research With Low Commercial Value, in Maskus, Keith E. and Reichman, Jerome H., Eds. International Public Goods and Technology Transfer in a Globalized Intellectual Property Regime, pages 288-306. Link
- A discussion of "technology specificity" and patenting -- particularly, the obviousness doctrine in an emerging field
- Biotechnology's Uncertainty Principle; Mark Lemley and Dan Burk; UC Bekeley Public Law Research Paper No. 125; Minnesota Public Law Research Paper No. 03-4
- (Mostly) Against Exceptionalism; R. Polk Wagner; U of Penn, Inst for Law & Econ Research Paper 02-18; U of Penn. Law School, Pub. Law Research Paper No. 07
- Weapons of Business Destruction from Slate, posted on Slashdot.
- This article makes an interesting point (probably made elsewhere too) that patents only work well when it is easy to figure out whether you are violating a patent or not. In the pharmaceutical industry where you are patenting things, it is pretty easy to determine whether you are violating a patent. In the software industry, it is very hard to determine whether your algorithm is violating someone else's algorithm. In the domain of synthetic biology, I would imagine that it would be pretty hard to assess patent violations. --RS