Project 2:
Open Source Studies and Collaborative Approaches to R&D
Principal Investigator: Arti K. Rai
This project explores, both empirically and analytically, the desirability and feasibility of different models of open and collaborative production in genomics research and development. By open and collaborative production, we mean projects that operate outside the traditional small lab structure of biomedical science and also disavow, at least to some extent, the traditional sorts of exclusionary property rights. Using the knowledge gained through empirical and analytical exploration, the project investigators and consultants are developing guidelines to assist funding agencies and universities in determining whether particular genome-related projects should operate under an open and collaborative model. The papers resulting from this work will also contribute to the growing economic literature on institutional organization of research and development, particularly the literature that focuses on "open source"
Project 2 employs quantitative, interview-based, and legal/analytical/economic approaches in its examination of how alternatives to traditional patenting and licensing are being deployed, or could be deployed, to stimulate collaborative R&D.
Some key findings from Project 2:
University Software Ownership
We conducted a systematic investigation of university ownership practices towards software. Our patent data, which span the period from 1982 to 2002, reveal that software patents represent a significant, and growing, proportion of university patent holdings. In about one-third of the cases, the patented software has biomedical applications.
Controlling for R&D (both computer science R&D and other R&D), one factor that is strongly correlated with a university's tendency to seek software patents is its tendency to seek patents in other areas. For those cross-sectional analyses where we have data on the size of technology transfer offices, we also find a strong correlation with size. The quantitative data indicate that most universities make decisions on patenting based not on the economic characteristic of the technology at issue but on the private marginal costs of patent acquisition.
This one-size-fits-all approach is particularly troubling to the extent that it may be driving a finding we observe in our review of software patent litigation to which universities are parties. In most of the cases that we have reviewed (admittedly, a biased sample in that the disputes led to litigation), the university patent appears to be deployed as a mechanism for ìholding upî the development efforts of successful commercializers. In many of the cases, the patents in question have also been questioned by the courts for obviousness or undue breadth.
Rai, Allison, and Sampat, University Software Ownership and Litigation: A First Examination, 87 North Carolina Law Review 101 (forthcoming 2009).
Synthetic Biology
Like software, synthetic biology represents an area of biological research that has a strong digital component. The economic considerations at issue in synthetic biology may thus be different from those in more conventional biological research.
The methodology employed in our research is primarily legal/analytical/economic. Because synthetic biology operates at the intersection of biology and software, historically lax legal standards for determining patent validity in both biotechnology and software have allowed broad patents on foundational technology to issue. These patents could create problems even if, under some very recent Supreme Court case law, they would not ultimately be upheld by the courts. Additionally, both broad and narrow patents could create problems to the extent they read on the standards that the synthetic biology community aims to enunciate. As experience in the information technology industries has shown, secret infringement - a common technique for avoiding liability in biological science research - is not a viable option for standards. Finally, synthetic biology patents may suffer from some of the same problems of inadequate disclosure historically associated with software patents. For example, inadequate disclosure appears to be a salient problem in the platform technology of zinc finger proteins.
Kumar S., Rai A. Synthetic biology: the intellectual property puzzle. Texas Law Review. 85:1745 - 1768. 2007. Available at here.
Chandrasekhran, S., Kumar, S., Valley, C., and Rai, A., Proprietary Science, Open Science, and the Role of Disclosure. Nature Biotechnology. 2009 27(2): 140-4.
Collaborative R&D for Drug Development
In this paper, we apply principles of incomplete contracting theory to propose a novel collaborative public-private partnership for accelerated drug development. This partnership would combine the academic sector's expertise in assay design with the pharmaceutical industry's currently secret libraries, so as to allow many more drug/assay screens than is currently possible. Creation of a standardized agreement for such screening would be facilitated by the fact that participants would bring only relatively low-value information to the table ex ante. However, the agreement would provide for clearly established property rights (and possible liability rules) ex post, at the point the information became valuable.
Rai A., Reichman J., Uhlir P., Crossman C. Pathways across the valley of death: novel intellectual property strategies for accelerated drug discovery. Yale Journal of Health Law, Policy, and Ethics 8(1):53-89. Winter 2008. Available here.
"Open Source" Studies
Closely related to the question on data sharing is the use of open source principles in the design of biological software, databases, and wet lab projects. Our study of various successful open source software and database projects (most notably Bioperl, a prominent bioinformatics program) suggest that open source-type principles could gain some traction in those areas, particularly to the extent that such principles are supported by relevant funders. The need for significant up-front expenditures, as well as concerns about credit/publication, may make wet lab research a less promising arena for open source.
Our proposal for using principles of open source research for upstream work on tropical diseases is currently being implemented in a number of different contexts. (These websites link to associated papers: www.tropicaldisease.org, www.osdd.net, www.thesynapticleap.org.)
Patent System Reform
In general, collaborations are more likely to succeed if they operate against the background of a patent system that yields clearly defined, high-quality patents. Thus the work we have done on using principles of administrative law to design a better patent system is relevant.
Benjamin S. M., Rai, A. Who's afraid of the APA: what the patent system can learn from administrative law. Georgetown Law Journal. 95: 296 - 336. 2007. Available here.