The workshop discussions covered much ground and many beliefs about the advantages and disadvantages of different strategies for protecting intellectual property. Many common assumptions about the economic benefits of patenting do not withstand close scrutiny, and none applies across the board to research tools in molecular biology. Although in many peoples' minds protection of intellectual property raises a question of whether to patent, controversies over licensing agreements and protection of unpatented intellectual property seem to be more common. The consensus among workshop participants was that the most important issue is access to, not patenting of, research tools. Participants from the three sectors—academe, industry, and government—made it clear that they are not opposed to patenting of research tools. By and large, they expressed the view that broad access to research tools is important for the continued vigor of the research enterprise and that research in molecular biology is in the public interest because of its potential to contribute to human health. At the same time, the specter of patents that could stifle research permeated many of the discussions. Even though they have not been patented en masse, the patenting of espressed sequence tags (ESTs) seems to represent the worst fears of many participants. Indeed, the issue was raised in almost every case study and repeatedly in general discussion. Many participants raised the concern that patents on genes or gene fragments could result in fragmented ownership that would severely disrupt scientific progress, or as Bennett Shapiro put it, "restricting access this way could Balkanize the human genome."

Discussions among the workshop participants made it clear that one should not expect to identify a single guiding principle for the protection of intellectual property surrounding research tools. The differences in circumstances surrounding the various case studies argue against easy generalizations. Even the recognition that a particular innovation is an important research tool does not define clear strategies. Not only does the category have no legal status, it is not even a discrete category. A technology might be a research tool in the hands of some users and a means of production in the hands of others. Even as a research tool, the technology takes different forms. Participants drew a distinction between research aids and discovery tools. That is, a research tool can be more like a screwdriver—designed to turn screws of a specific size and shape—or more like a telescope—a discovery tool useful for scanning horizons. Describing something as a research tool generally describes only particular instances of its use and almost never defines its full range of uses or values.

Leon Rosenberg noted that the research-use exemption is used by many companies and described it as a "rational forbearance," which he hoped would continue. Many workshop participants, especially those working in universities, expressed surprise that the research-use exemption was not an established legal principle. However, until this exemption is upheld in a court of law, it will remain simply a general practice with no legal protection and will be sustained primarily by individuals' self-interest.

All the speakers at the workshop associated with major pharmaceutical companies—Bennet Shapiro, Thomas Caskey, and Leon Rosenberg—mentioned that it was in their organizations' self-interest to disseminate their research tools broadly. In contrast, speakers from smaller biotechnology firms noted that their survival depended on their ability to disseminate research tools on a more restricted basis. Both sectors claim a common goal—to develop useful applications from advances in molecular biology—but the means by which they can achieve this goal, given their business constraints, are quite different.

Finally, the variation in the technology of different research tools argues against the wisdom of searching for universal guiding principles. Perhaps the only guiding principle that one might derive from categorizing a particular innovation as a research tool is the observation that the innovation has the potential to foster scientific progress. Certainly, none of the workshop participants questioned the principle that scientific progress in biomedical research should not be impeded, no matter how the intellectual property rights are distributed. Instead, the disputes about rights of access to research tools lay in the details. Does the delay in disclosure that results from patent application inhibit scientific progress? Is access to research tools broad enough to facilitate scientific progress? Do the benefits of a protected environment for further development outweigh the costs of excluding others from this research tool?

Indeed, rather than providing answers, the most practical contribution of the workshop might have been to suggest questions. What sorts of issues might technology transfer managers consider in deciding how to manage research tools? The following questions emerged from our distillation of the common themes raised by workshop participants and from comparisons of the case studies. They might be useful for informing decisions about appropriate strategies for protecting intellectual property that constitutes important research tools.

•   Is this technology critical to researchers? Are there other ways to do the same thing?

•   Is private investment necessary to make the research tool more useful? Is the size of the potential market sufficient to warrant the amount of extra investment?

•  Does or might this research technology have alternative uses for which commercial markets exist, and would private investment be necessary to develop those markets?

•   Are multiple dissemination strategies feasible (such as different licensing terms and different fees)?

•   Where substantial further investment is necessary to develop the technology, will further development yield additional patentable inventions sufficient to motivate investment?

•   Is this invention likely to be only one of many pieces to a complex puzzle, each of which could be separately owned?

Those questions encapsulate themes raised by many of the speakers. Many participants spoke of the problem of uncertainty, which is likely to be a continuing problem in setting policies about proprietary decisions. To some degree, it seems that uncertainty is an inevitable byproduct of the recent advances in molecular biology. Molecular biology has already presented the patent system with new categories of patentable material, for which case law—which of necessity looks to the past for guidance—cannot always be expected to provide adequate guidance. One can expect advances in molecular biology to continue to pose new and unexpected challenges to our intellectual property traditions. As science continues to present new challenges, case law will inevitably lag behind the science. Because the useful applications of advances in molecular biology are so closely tied to basic research, it is particularly important to strike the right balance between the twin goals of fostering the growth of scientific knowledge and promoting private investment in the development of applications for new technology. In many cases, those goals can be encouraged by the same strategies; in others, the strategies might conflict. The balance is not likely to be achieved through any single policy and clearly requires the close and continued attention of people concerned with the science and application of molecular biology.