Intellectual Property & Science

In today's class we will look at how it is possible to own scientific results, and what the different types of ownership mean to the conduct and value of science. There are three basic issues to discuss:

• What is the law that covers intellectual property?
• How do IP practices influence science?
• What is changing in the world of IP & Science?

We will have a guest discussant, David Poticha, from the UC Technology Transfer Office, join us for the first part of class. Please read the following and think about good questions for him.

What is Intellectual Property, and what laws cover it?

In American law, there are two ways to own scientific results: copyright and patents. Copyright gives authors of an original work (e.g. a publication or piece of software) an exclusive right to determine in what way the work may be published, produced, sold or distributed. As the name implies, the right is restricted to copying the original. Patents offer a broader set of rights around an invention. Those rights exclude others from making, using, offering for sale or selling the invention. Collectively, copyrights and patents are referred to as "intellectual property" or IP.

US law recognizes three kinds of patents:

• Utility: a new and useful process, machine, article of manufacture or composition of matter, or an improvement of an existing one
• Design: a new, original and ornamental design for an article of manufacture
• Plant: an invented or discovered distinct variety of asexually reproducing plant

Courts have defined the limits of what can be patented, thus far it has been held that the laws of nature, physical phenomena, and abstract ideas are not patentable subject matter. There has been controversy over whether animals can be patented, but patents on transgenic animals (and other organisms) have been issues by the US patent office since the 1988 oncomouse. Other countries have reached different conclusions, with Canada refusing to allow animal patents, and the European patent office in the middle. In a case related to a patent of a human-animal chimera inteded to set a legal precedent, the US Patent office has ruled that human beings (or organisms similar to them) are not patentable.

How do IP practices influence science?

Ownership of IP has brought a lot of money into Universities and supported research.

The Bayh-Dole act of 1980 gives recipients of federal assistance (e.g. universities that get research grants) the right to patent inventions that arise from such funding. In return, any invention created with the use of those funds must be reported to the funding agency, patent applications must be filed, and the recipient has to promote and attempt to commercialize the invention. University technology transfer offices were largely set up to meet their obligations under the Bayh-Dole act. The emphasis on commercialization of results arising from federal funding of scientific research has generated some opposition.

A few academic institutions have made huge windfalls on patent royalties, and used that income to support science. Relatively few inventions ultimately lead to patent applications, relatively few patents are licensed, and relatively few licensed patents produce a lot of income. In the most recent (2006) survey published by the Association of Univerity Technology Managers, there were 18,874 invention disclosures filed at US universities in 2006, 11,622 new patent applications filed (and about 4300 appeals or revisions of old applications), and 3,255 patents issued. In the same year, universities executed 4,192 new licenses or options on their IP portfolios. The total licensing income in 2006 was nearly $1.25 billion, most of which went to relatively few schools. The University of California (whole system) had nearly $200 million in income, NYU had more than $150 million, and Stanford was third with $61 million. (The University of Colorado was 13th, with a bit over $21 million in revenue.) Most of the income for each school comes from a relatively small number of "blockbuster" licenses. The University of California's coffers increased by $100 million in 2006 due to a settlement of a case involving Monsanto and a recombinant DNA patent relevant to making a dairy cow growth hormone. The California system also has longstanding revenue from patents on a hepatitus B vaccine. In 2005 Emory university got more than $500 million for the sale of royalty rights to the anti-HIV drug Emtriva.

The factors influencing whether a university can gain significant revenue through blockbuster patents are complicated (see this economic study that suggests its a combination of volume of basic science research and networks of experts able to make good analyses of their likely commercial importance).

IP restrictions can interfere with scientific research at universities.

There are several ways that IP could interfere with scientific research. Some patents touch on fundamental matters of inference in such a way as to threaten the conduct of certain scientific activities. Patents on gene sequences have been held up as particularly burdensome to scientific research. A recent analysis has suggested that these concerns may not be problematic in practice, although that other issues around commercializations (non-patent related) and a failure to share among scientists for competitive non-commercial reasons might reasonably be said to impede science. [A longer article in this area is optional reading.]

Until 2002, it was generally believed that university research was immunized against patent infringement based on a claim that such use was "experimental." The Madey vs. Duke case made clear that universities could indeed be sued for patent infringement even for experimental use of a patented invention. This has raised concerns that intellectual property law and practices could interfere with scientific research.

Changes in IP approaches in science may have an effect on openness in scientific research

Over the last decade or so, efforts have been made to create intellectual property rules that encourage openness, and bioinformatics has been at the forefront of some of these efforts.

Open approaches to copyright are generally called open access, has been used to ensure that the scientific literature is available to the world without expectation of payment. In 2008 it became law in the US that "all investigators funded by the NIH submit or have submitted for them to the National Library of Medicine’s PubMed Central an electronic version of their final, peer-reviewed manuscripts upon acceptance for publication, to be made publicly available no later than 12 months after the official date of publication" The Organization for Economic Co-operation and Development adopted a communique that encourages all members to adopt similar approaches. This trend is controversial, and efforts have been made to overturn the US law.

Another approach to ensuring openness aims at patents, particularly software patents. Software patents are a relatively recent phenomenon, that have engendered a large amount of opposition in some parts of the software development community. While some have taken a moderate approach to trying to fix aspects of the patent system that seem inappropriate for software, others developed an alternative: Open Source software. The idea of open source is to create legal restrictions that ensure that others can modify a program as they see fit, and distribute such modifications freely. The most popular open source license, the gnu public license (GPL), also requires that anyone who creates a derivative product must also distribute that under the same license.

Probably the most important open source systems in the world are the family of linux operating systems and related code. However, Open Source software is playing an increasingly important role in science, particularly in bioinformatics. The Open Bioinformatics Foundation supports development in many of the most important tools used in our field, e.g. BioJava, BioPython, and BioPerl. The R statistical software package, and the derivative BioConductor project are also widely influential in our field. Search for the phrase "open source" results in more than 1,000 biomedical journal articles in PubMed. However, the adoption of open source software in bioinformatics is also controversial; the ISCB finally approved a software sharing policy statement after years of wrangling among the membership.

The role of openness and intellectual property in science is likely to continue to be a controversial subject. The AAAS has studied it carefully, and the Royal Society (in the UK) has also published a set of recommendations. A Harvard Business School study on the value of openness in scientific problem soving shows pretty conclusively that in an industrial environment, scientific problems are solved faster and better through an open process than through a secretive one, yet that is far from current practice in most companies.