Ensuring Access to the Outcomes of Community Resource Projects

The assistant professor was very excited. She had only been in her faculty position for a few months and had several grant proposals pending. One of them was to the Plant Genome Research Program (PGRP) to develop a novel method for silencing genes. It had been about 4 months since her application was submitted, so a voice mail message from the Program Director asking her to call could mean only one thing: She was going to be funded!

Her hands shaking, she picked up the telephone and dialed the number. However, it was not that simple. Although the Program Director was calling to let her know that her proposal was under consideration for funding, some intellectual property (IP) issues had to be sorted out before a final decision could be made. The assistant professor was mystified. After all, she had not signed any material transfer agreements (MTAs) and, because she wanted to share her ideas with everyone, did not plan to patent any of her research. The Program Director seemed to be very concerned about the promoters she planned to use and asked where they came from. Apart from the freezer in her postdoctoral advisor's laboratory, the young assistant professor wasn't sure...

This is not a true story so we can give it a happy ending. Her postdoctoral advisor had cloned and patented the promoters many years before. His institution, which actually owned the IP rights, prepared an MTA that enabled her to use them without restriction. The assistant professor was pleased to discover that patent protection did not prevent her from using the promoters. She was surprised to learn that an institutional representative would have to sign the MTA. Fortunately, her institution's technology transfer office accepted and signed the agreement. The Program Director was satisfied by the outcome, and an award was made.

This story only scratches the surface of IP issues, which can be unexpected and multilayered, especially for projects that are funded to develop and distribute resources or tools to the research community. Many academic researchers are not aware of the IP issues associated with their research, either with the starting materials or with the outcomes. They often view technology transfer offices as the potential cause for delays in their research and publications and avoid contact. Unfortunately, this can lead to the very problems that they are attempting to avoid. The purpose of this article is to review some of the issues that arise during the review, funding, and management of biological community resource projects at the National Science Foundation (NSF). However, many of the issues that are raised are common to projects funded by other federal and private agencies.

THE NSF

The mission of NSF, set out in the NSF Act of 1950 (Public Law 810507), is: “To promote the progress of science; to advance the National health, prosperity, and welfare; to secure the National defense; and for other purposes.” Funding from the NSF accounts for about 65% of the federal support to academic institutions for basic research in nonmedical biological sciences at academic institutions in the United States (http://www.nsf.gov/od/lpa/news/media/fsnsf.htm). In 2002, the NSF Biology (BIO) Directorate made 3,494 research awards totaling over $505 million (http://www.nsf.gov/bio/bio_bdg04/bio_bgt04_toc.htm). Until recently, almost all of these awards would have been made to single investigators or to small collaborative teams for focused research projects. In most cases, no specific data or material release requirements were attached beyond the general expectations stated in Section VIH of The Grant Proposal Guide (http://www.ns-f.gov/pubs/2003/nsf032/032_6.htm). These expectations state: “... that the research findings will be submitted for publication promptly, and that investigators will share with other researchers, at no more than incremental cost and within a reasonable time, the data, samples, physical collections and other supporting materials created or gathered in the course of the work.” Grantees are also encouraged to share software and inventions, once appropriate protection for the IP has been secured, and otherwise act to make the innovations widely useful and usable.

THE NATIONAL PLANT GENOME INITIATIVE (NPGI) AND THE NSF PGRP

The advent of genomics has changed, at all levels, the way in which academic biological research is funded, conducted, and managed. Development of genomics resources often involves multiple institutions and can be funded by multiple agencies. The NSF is a participant (along with the U.S. Department of Agriculture [USDA], Department of Energy, and the National Institutes of Health) in the NPGI. This initiative was established in 1998 to exploit advances in genomics to improve the useful properties of plants that are important to humanity. The initial 5-year plan (http://www.ostp.gov/NSTC/html/npgireport.html) included guiding principles for the operation of participating programs; one guiding principle was to make all information accessible to the broader community, including industrial and international researchers. The new 5-year plan (http://ostp.gov/NSTC/html/NPG12003-2008.pdf), released in January 2003, also includes these important principles.

The NSF PGRP was established in 1998 as part of the NPGI. In its first 5 years, this program has invested about $250 million in 101 projects, many of which are collaborative efforts aimed at generating collections of resources, or tools, for the broader community (http://www.nsf.gov/bio/dbi/dbi_pgr.htm). Through the NPGI, the NSF has also funded projects (in partnership with Department of Energy and USDA) to participate in international projects for the sequencing of the Arabidopsis and rice (Oryza sativa) genomes. More recently, the NSF and USDA have established the joint Microbial Genome Sequencing Program (http://www.nsf.gov/bio/pubs/awards/microbgen0201.htm). Additional programs have been established within the NSF BIO Directorate to support development of genomic resources, including the Arabidopsis 2010 Project (http://www.nsf.gov/bio/progdes/bio2010.htm) and The Bacterial Artificial Chromosome (BAC) Program (http://www.nsf.gov/bio/pubs/awards/bachome.htm).

The broad scope of these programs, from sequencing to tool development, along with associated IP issues, has led to expanded project review and management at the NSF. In all cases, the challenge is to balance the goal of making NSF-funded resources broadly available while preserving the right of funded institutions to seek patent protection for the IP generated. Because the PGRP has handled the largest number of community resource projects to date, its guidelines are the most comprehensive. However, the expectations are generally consistent for all NSF BIO community resource projects, regardless of the program providing the funding.

THE IMPACT OF THE BAYH-DOLE ACT

IP is an issue for community resource projects in the United States in part because of changes in the law about 20 years ago. The passage of the Bayh-Dole Act in 1981 permitted universities and small businesses to own the IP rights to discoveries made through federal funding and to become involved in commercialization. Before enactment of this bill, there was no government-wide policy for assigning ownership of inventions made by contractors and grantees. The lack of a consistent and coherent policy resulted in poor transfer of new technologies resulting from this funding to the private sector. In 1980, the government held title to about 28,000 patents, yet fewer than 5% of these were licensed to industry for commercial development (Council on Governmental Relations Report; http://www.cogr.edu/docs/bayh_dole.pdf).

Universities are now obliged to have written agreements with faculty and staff that require disclosure of inventions and assignment of associated IP. Each new invention must be disclosed to the funding agency within 2 months of notification by the investigator. After notification of the funding agency, the university has 2 years to decide whether to retain the title to the invention, unless the data have been published or otherwise used publicly, in which case it has 1 year. A university that elects to retain the title has 1 year to file a patent application in the United States. After filing for a patent, the university has an additional 10 months to notify the funding agency of its intention to apply for a foreign patent. If the university does not intend to file foreign patents, the funding agency may elect to do so on behalf of the United States Government.

The degree of impact of IP issues on management of community resource projects depends on the nature of the resource generated. Major resources impacted by these provisions include cloned promoters and selectable markers. Since 1981, many cloning tools, whether obtained from an industrial or academic source, are shared only if the requestor signs an MTA delimiting the terms under which the item can be used. An MTA may include terms that assign any IP associated with discoveries or inventions made using the resource to the owner of that resource (“reach-through rights”). Academic researchers sometimes rely on the assumption that patent rights are not generally enforced when used for basic research. However, broad and unimpeded distribution of the publicly funded derivatives to industrial and academic researchers may not be possible, even if patent rights were not enforced during their development. In addition, a recent ruling (Madey v. Duke No. 01-1567; http://www.ll.georgetown.edu/federal/judicial/fed/opinions/01opinions/01-1567.html) may effectively end research exemption for use of patented technologies. If it does, end users of such technologies will need to proceed with caution.

SHARING OF COMMUNITY RESOURCES

Sequence information from public large-scale genome sequencing projects in the United States (and internationally) is usually deposited immediately into GenBank because these projects adhere to a set of principles established at the first International Meeting on Human Genome Sequencing held in Bermuda in 1996. The participants of this meeting, which included representatives of laboratories involved in human genome sequencing and of funding agencies, issued a statement that all genomic DNA sequence information should be “freely available and in the public domain in order to encourage research and development and to maximize its benefit to society” (http://www.sanger.ac.uk/HGP/policy-forum.shtml). These guidelines were used for NSF-funded Arabidopsis and rice projects and are in use for the maize (Zea mays) and microbial sequencing projects.

Because a central mission of the PGRP is to support generation of community resources, specific guidelines for data and materials release have been created and are provided in the annual program solicitation. Investigators are asked to describe how project outcomes will be managed in their proposals. In the current Program Solicitation (NSF 02-187; http://www.nsf.gov/pubsys/ods/getpub.cfm?nsf02187), highlighted items to be addressed include the nature of the materials to be shared, the timing and means of release, and any constraints on release. In the case of a multi-institution proposal, the institution of the principal investigator is designated the lead. A lead institution is responsible for coordinating and managing the research outcomes and IP resulting from the award. Institutions of the coprincipal investigators are strongly encouraged to formulate a coherent plan before submission of a proposal to avoid problems once the project is under way.

Guidelines for disclosure and release are specific for each type of resource. It is expected that sequences resulting from high-throughput, large-scale sequencing projects will be released to public databases according to the currently accepted community standard as soon as sequences are assembled and checked for quality. For genome sequence, the quality standard used is the Bermuda agreement, and the sequences are required to be deposited in GenBank. Projects funded to produce community resources (biological materials, software, etc.) are encouraged to make these available as soon as the quality is checked against prestated specifications. It is expected that the timing of release will be stated clearly in the proposal. Any resource produced must be available to all segments of the scientific community, including industry. A reasonable charge for filling requests is permitted provided the fee structure and terms are detailed in the proposal. Where resources will likely be in great demand, projects are encouraged to deposit plant materials in public stock centers such as the Arabidopsis Biological Resource Center (Ohio State University, Columbus; http://Arabidopsis.org/abrc/) or the Maize Genetics Cooperation Stock Center (University of Illinois, Urbana-Champaign; http://w3.aces.uiuc.edu/maize-coop/). If accessibility is to differ between industry and the academic community, the differences must be clear. If an MTA is to be used, the terms must be described in detail, and the ability and willingness of any proposed stock center to handle restricted distribution materials must be documented. When a project involves the use of proprietary data or materials from other sources, the data or materials resulting from NSF-funded research must be readily available without any restrictions to the users of such data or materials (no reach-through rights). It is expected that the terms of any usage agreements will be clearly stated in the proposal. It is also expected that any active agreements such as MTAs, the use of which could adversely affect the outcomes of PGRP-funded research, will be disclosed to the Program Directors before funding of a project.

Large community resource projects require more intensive management than smaller, single-investigator awards. Awards are often made as cooperative agreements in which the funds are released at intervals (often yearly), contingent on satisfactory progress toward a prestated set of yearly deliverables. The progress toward planned data release and compliance with stated goals and terms are evaluated through written annual progress reports and at site visits. These projects usually have an Advisory Committee that is advisory to the project rather than NSF. Face-to-face meetings occur at least one a year, and the resulting reports, along with the project team responses to any issues raised therein, are sent to NSF as part of the annual progress report for the project. Finally, any newly arising MTAs, software licenses, or other agreements that may affect distribution of project outcomes are reviewed and approved by the PGRP staff before implementation. Clearly, there is no “one size fits all” set of specific guidelines for managing community resource projects. Each is reviewed, awarded, and managed with regard to its specific goals and outcomes. Although these additional reviews create a substantial amount of work for both the funding agency and grantees, the benefits to the community are substantial.

INTERNATIONAL AND INDUSTRIAL COLLABORATIONS

International and industrial collaborations have been encouraged from the outset of the NPGI and the PGRP. International groups do not normally receive funds from NSF but support their portion of a project from internal resources, participating as collaborators. Industrial participants can participate as collaborators or as subcontractors. In some cases, a company may have a unique tool, method, or resource not available in academia. Alternatively, the company may provide a service such as high-throughput sequencing faster, at a lower cost, and at higher quality than an academic counterpart. In these instances, a project may elect to pay a company to provide these services through a subcontract. This kind of fee-for-service relationship is quite distinct from the collaborations mentioned above. All services or materials paid for by the project must be available to the community without any IP issues that hinder distribution. The use of proprietary technologies by a company need not impact the delivery of unencumbered materials, and in most cases is not an issue.

In collaborative projects, the industrial participants are a part of the intellectual direction and, as such, often bring in-kind resources to the project. As detailed in the Program Solicitation, any IP issues associated with these industrial contributions have to be clearly addressed in the proposal and are considered during the review process. Signed, detailed agreements are required before any award is made. Reach-through rights are not permitted on the NSF-funded project outcomes, even if they derive from materials provided directly from the company. Long delays of project data release to allow privileged review by the industrial collaborator are also not allowed. Throughout this discussion, industry and academia have been discussed as discrete entities. However, the boundaries between industry and academia have blurred with the emergence of “the corporate academic,” who may well hold an academic appointment and have additional industrial research interests. In some cases, an academic researcher may have started a company in partnership with his or her institution to capitalize on discoveries made through NSF funding. In these situations, the onus of ensuring that there is no resulting conflict of interest lies with the institutions submitting proposals. Institutional provisions are considered carefully during the review process, and, if necessary, additional management plans are added to ensure that funds are expended in accordance with the NSF guidelines.

DATABASES

Management of project databases and the resources therein has presented new challenges. A common issue is exactly what will be posted publicly on the World Wide Web and the terms under which data can be accessed. In most cases, it is expected that the raw data (for example, from microarray analyses) and derivatives and compilations will be made available on the Web site or upon request if the datasets are large. The goal of providing immediate public access to project data once it has met appropriate quality control standards may need to be balanced with the need to allow students and postdoctoral researchers on the project sufficient time to analyze data for publications. Limited delay in release of subsets of data may be permitted so that a publication can be completed. However, such agreements generally do not apply to all project data and include a default date by which data must be released.

Projects that develop a public database into which all project data are released immediately often face the issue of receiving appropriate credit for their work. Obviously, anyone should be able to access and analyze the data. It is up to the project team to conduct and publish prompt analyses of their data. However, the desire to have the first opportunity to perform all analyses is not sufficient justification to withhold project data for prolonged periods of time. Some projects have posted the results of their analyses on their Web site and then discovered unattributed use of their results. One way in which this type of concern is being addressed by some projects is by posting guidelines or agreements governing use of, and citing results of, these analyses.

CONCLUDING REMARKS

In summary, collaborative community resource projects, particularly in genomics, are driving the development of more coherent guidelines for sharing data and materials at U.S. funding agencies, including NSF. These guiding principles will need to remain flexible to accommodate the wide range of data, material, and tools covered, as well as new technologies. It is increasingly important for academic researchers to make sure that they are aware of possible IP issues that could affect their project. In the end, the NSF's goal in funding resource development projects is to foster production of tools, materials, and results that are available on similar terms to all members of the broader scientific community. The maximum benefit to the scientific community will only be realized if this actually happens.