This section includes relevant background text, definitions and examples, policy statements, video, and expert commentary. It should be read by those looking for a thorough understanding of collaborative science.

»Foundation Text«

Table of Contents

  1. The trend toward more collaborative research, and examples of it.
  2. What is collaborative research?
    1. Collaboration usually involves two or more researchers within an institution or in different institutions, working either in the same field or in different scientific fields or sectors of the economy
    2. What is spurring the increasingly collaborative and multidisciplinary nature of research?
      1. Funding sources
      2. Researchers need complementary skills
      3. Ease with new telecommunications technologies, e-mail
      4. Technology-transfer between academia and industry fostered by the Bayh-Dole Act
      5. Evidence that collaborations will improve progress
  3. What are some of the potential problems with collaborative research?
    1. Difference in style of investigators
    2. Difference in style of research across disciplines
    3. Differences between academic and industrial research with respect to sharing of data and results
    4. Ethical considerations may affect research across institutions and nations; there may be differences in issues such as disclosure of conflict of interest and standards for clinical research
  4. What are the ways to enhance collaboration?
    1. Communication first, second, and throughout
    2. Discussing in advance who will do what in a project; understanding that this might change as the research evolves
    3. Discussing authorship in advance
    4. Discussing data in advance
    5. Discussing intellectual-property issues in advance
    6. Managing accountability
  5. What is the institutional role in the collaborative process?
    1. Technology-transfer offices
    2. Grants-and-contracts offices
    3. Clinical-trials offices

I. The trend toward more collaborative research, and examples of it

High-energy physics. The human-genome project. The social impact of computing. Telemedicine. As the nature of the questions that researchers ask becomes more complex, investigators have come together, bringing their unique expertise to find the answers. Researchers working in these fields—and, more and more, those working in all academic disciplines—are collaborating in large groups, and in many cases will work with researchers who are educated and skilled in different subjects. Physicists throughout the world pool resources to understand the subatomic particles that compose matter. Molecular biologists, engineers, computer scientists, lawyers, and investment analysts work together to develop machines, methods, software, and new companies to mine the information contained within human genes to solve health problems. Physicians and computer scientists develop computerized systems that allow patients in rural areas to connect via the Internet for a virtual physical examination. Sociologists study how advances in computing technology change human interactions, relationships, and commerce. But, as the trend for more interdisciplinary research continues, researchers are going to have learn to speak languages outside the realm of their own expertise. Such communication can take time and create conflict.

Benefits of Collaborative Research

The Problems of the World



II. What is collaborative research?

A. Collaboration has been intrinsic to the research process for the past 50 years, but collaboration per se usually refers to researchers who work within the same discipline, either within an institution or in different institutions. When a biochemist solves the crystal structure of a protein for a molecular biologist working on the regulation of the protein, that is a collaboration within the biological sciences. Multidisciplinary research is a form of collaborative research that involves researchers working across disciplines, either within an institution or in different institutions. A physician working with an engineer to manufacture a new imaging device, or an epidemiologist working with a political scientist on a tobacco-control initiative, is an example of a cross-disciplinary research project. When the pharmaceutical industry works with a medical center to perform a clinical trial of a new drug, it is a collaboration across industry and academia. Each of these interactions creates different expectations and requires a variety of modes of communication to ensure that the collaboration is successful.



B. What is spurring the increasingly collaborative nature of research?

  1. Funding sources: The federal government, including the National Institutes of Health and the National Science Foundation, now supports projects that ask for researchers in different disciplines to work together. The NIH’s road map, a project that is determining the agency’s future extramural and intramural research mission, has defined certain areas for study, such as nanomedicine and structural biology, which will involve research within and across disciplines, and, in some cases, with industry. The new programs support research as well as the education of next-generation scientists to look across disciplines when doing research. The NSF also has “cross-cutting/interdisciplinary programs” that seek new and multidisciplinary approaches in research, education, earth systems, and organizational structures. Private foundations also will fund centers at universities that bring together expertise in different areas to solve a particular problem. Finally, universities have set up funding streams to support innovative approaches within their institutions.
  2. Researchers need complementary skills A chemist developing a nanotube to deliver a drug that will regulate the expression of a gene to prevent the growth of cancer cells or to kill existing cells has many technical problems to solve. The main problem is creating the nanostructure that has the intelligence to respond to the changes in the cellular mileu. That researcher does not necessarily have the expertise to understand what impact the technology he or she is developing will have on the recipient. If the government says in a grant application that special attention will be given to a grant that looks at the social effects of medical nanotechnology, a multidisciplinary collaboration between the chemist and a social scientist might be born. As scientific technologies, such as cloning, are changing the way human beings think about themselves, the public may be interested in understanding the impact that these new kinds of research will have.
  3. Ease with new telecommunications technologies, e-mail E-mail and Web-based technologies have changed the way many people in research-oriented countries interact, and scientists are among the beneficiaries of the new communications technology. While the wider use of fax machines, in the 1980s, allowed documents and data to be sent across phone lines, shared computerized databases at that time were more difficult to manage. Today, Web-based technologies allow researchers to input and manipulate data in shared databases with ease. Web-based telecommunications systems also allow people from across the world to communicate, simulating face-to-face meetings.
  4. Techology transfer between academia and industry, fostered by the Bayh-Dole Act The Bayh-Dole Act of 1980 allowed universities to have control of the intellectual property, such as patents, that is generated from federally funded research; with a patent in hand, universities could exclusively license the patent to businesses. For the past 25 years or so, many universities, including Columbia, have benefited from the licensing revenue. Recent inventions, in the form of new drugs and computer technologies, have also helped the public. The law has encouraged new relationships between academic researchers and companies. Recently, Congress went a bit further and enacted what is called the CREATE Act to foster intellectual-property protection for researchers collaborating across institutions. But critics have charged that Bayh-Dole promotes universities’ selling out their interests to industry rather than relying on raising money from tuition and other sources. Problems with this, such as academics having a conflict of interest in exploiting research results for private gain, will be addressed later.
  5. Evidence that such collaborations improve progress Research collaborations, according to Francis Macrina1, allow researchers to ask questions they wouldn’t be able to ask if they worked alone. In the biomedical sciences, for example, researchers can test the same hypothesis using genetic and biochemical methods, with data from both validating the findings. According to Macrina, the identification of the colon-cancer gene came, first, from findings with medical geneticists working with patients who had the disease. Repair mechanisms in yeast led to an understanding of how the gene was working in higher organisms. With the gene in hand, and with an understanding of its action, researchers have been developing new diagnostics and better treatments for the disease.



III. What are some of the potential problems with collaborative research?

A. Difference in style of investigators

As in any relationship, people have different styles of relating. Some people are more formal, while others are more laid-back and relaxed. Likewise, in science some researchers have collaborations in which they develop a project over a beer and a handshake at a conference and the tenure of the tie remains informal throughout. Others require more documentation and rigorous enumerations of responsibilities. However, even if a researcher works easily with another researcher, shared grants, data, and materials require more formal written agreements involving grants-and-contracts offices at their respective universities.



B. Difference in style of research across and within disciplines

One collaborator believes that peer-reviewed papers should be short and should use a limited amount of data. Another collaborator believes that more data should be collected and the “story” of the research should be developed before anything gets published. This kind of disagreement can occur with collaborators in the same field or in different fields. Disciplines also suggest, in different ways, who should be an author on a paper. In certain fields, people who have not contributed substantially to the intellectual process of the research are not included, while, in other fields, people get authorship if they participated in doing the research at any level.

Different research disciplines also have varied approaches in work habits. Biomedical laboratories can run 24 hours a day because of the nature of performing experiments, but other disciplines may have more routine, 8- to 10-hour days. Also, different types of work may follow different timetables. Statisticians working on analyzing data may move faster than the social-science researchers surveying hundreds of people in a population for data. Imaging a region of the brain involved in a drug addict’s high may generate quicker results than elucidating the socioeconomic pathway that the drug took to come into the addict’s hands.

Researchers also speak different languages. Technical jargon exists within subspecialties within a discipline and across disciplines. It can be challenging for researchers to create a language understood across many disciplines.

Several Kinds of Concerns

The crucial point is to presume nothing and to put everything on the table for discussion as early in the relationship as possible.



C. Differences between academic and industrial research with respect to sharing of data and results

The free exchange of information at scientific meetings and in publications is the ethic and lifeblood of academia. However, in commercial enterprises research data could have financial repercussions, so data is carefully vetted before it is published, if it is ever published. When academics and business researchers work together on projects, each party has to come to an agreement about how data and materials will be shared. Institutions such as Columbia University do not allow hindrance of publication in research collaborations with industry. Other universities are willing to forgo the freedom in exchange for funding, access to industrial ideas, and opportunities to train students in commercial types of research endeavors.

The issue of industry sponsoring drug trials at academic research centers and not allowing the recipients to publish papers about the results of the trials has become front-page news. The New York Times reported in late 2004 that medical-school researchers funded by the pharmaceutical industry had sought access to unpublished data in an antidepressant trial to determine whether the drugs increased the risk of suicidal behavior in children. Drug companies denied the questioning researchers access to the data and would not allow them to communicate with other researchers who had participated in the same study at another institution. Drug companies perform clinical trials at multiple locations and keep the data centralized, with each institution not aware of results from elsewhere. An editorial in the Times said that it may now be time for all institutions to negotiate contracts with drug companies that would “ensure researchers’ access to data and prompt publication of results.”2



D. Ethical considerations may affect research across institutions and nations

Universities throughout the nation and the world may have different standards for the nature of disclosure of potential financial conflicts of interest. While one academic medical center would not allow a researcher who developed a drug to be involved in the clinical trials of that drug, another might permit it as long as safeguards were in place that would prevent the researcher from knowing the progress of ongoing trials. International collaborations raise concerns about differing standards of treatment for research subjects in developed versus developing nations. Reader K. Lie, in his presentation of “Ethical Issues in Collaborative Research,” points out that the quality of care given to subjects in poorer countries perhaps should be equal to that of the richer ones, even if some say that different standards of care are essential to identify results.

The Problems of the World



IV. What are the ways to enhance collaboration?

Macrina3 points out six key components of a successful collaboration. They are:

A. Communication first, second, and throughout

No one in a collaboration should assume anything. Establishing, maintaining, and even terminating communication is important for the project to continue. If two researchers exchange data, personnel, or materials without a formal collaboration in place, perhaps they need to address whether one should be established. Once a collaboration is formally created, then discussion about data, ideas, and personnel issues should occur. Researchers need to communicate effectively, whether the other person is across the hall or on the other side of the globe.

Collaborative Relationships

Communication is particularly important in collaborations between academia and industry. Special requirements may be imposed on the publication of material or on the invention and patents. Whether a graduate student participates in such an academic-industrial project must be resolved early on if the research may not be published in a timely fashion. Also, patent lawyers, technology-transfer administrators, and marketing personnel from industry need to establish a common ground for communication.



B. Discussing in advance who will do what in a project, while understanding that the research may evolve

Parties in a collaboration should define goals in such a way that they could not have been established without the collaboration. Setting goals leads to expectations and outcomes. Who will take charge of the collaboration also has to be defined. As multiple laboratories or groups of researchers may be involved, coordinating the effort among the participants requires management (and communication). When a research project changes direction, how that will impact participants needs to be addressed. Authors may be added or eliminated. Finally, researchers have to determine when a collaboration is over.



C. Discussing authorship in advance

Different disciplines have varying standards for determining authorship. The criteria for authorship among collaborators has to be established beforehand so all know what to expect. But with authorship comes responsibility, so collaborators need to determine how they will deal with the differing expertise levels of each author. Who will actually write the manuscript and be responsible for the input from collaborators has to be established. If the research changes direction, someone expecting authorship might be disappointed, so the evolution of a project has to be considered. Finally, who will be included in acknowledgments should be addressed.



D. Discussing data and material management in advance

Macrina offers good examples of what can happen among researchers who share resources and data. Laboratory A, for example, has purified a protein and prepared antibodies to the protein. Laboratory B will screen an expression library to find the clone. Laboratory B will get the monoclonal antibody and the clone will be shared. But will Laboratory B also get the cell line that makes the monoclonal antibody? How such a question is resolved affects the ability of the laboratories to replicate work and to perform independent work at the end of a collaboration. The issue of who owns data is governed by the type and source of funds used to support research. The NIH and the NSF allow grantee institutions to own data, a regulation with implications for research done off-site. Investigators and institutions also have rules for the custody and retention of data, to which all parties must adhere.

Also, the transfer of materials among collaborators is subject to so-called “material transfer agreements,” or MTAs, developed by administration offices. They include:



E. Discussing intellectual property issues in advance

All investigators want to be able to protect results that might have potential commercial application. Disclosing results early could prevent collaborators from being able to obtain patent protection. All parties should know of institutional and granting-agency policy regarding intellectual property and patent procedures.



F. Managing accountability

Each institution has to abide by certain regulations, policies, and laws. Researchers working with animals, humans, or hazardous substances have to conform to the appropriate regulations, policies, and laws. Basic research scientists might have access to patient data from the clinical arm of a study and must be aware that they need to maintain the confidentiality of patients. Also, clinicians should inform bench researchers of the potential hazards of certain human tissues. Researchers also need to inform one another of any potential conflict of interest that they might have in the project.



V. What is the institutional role in the collaborative process?

A. Technology-transfer offices

Science and Technology Ventures, or STV, http://www.stv.columbia.edu/, is the technology-transfer office for Columbia University, and is an example of analogous offices in universities throughout the country. The responsibilities of STV include identifying and patenting new inventions and copyright materials, including software. The office helps inventors develop the necessary documentation for patents and other kinds of protection. Although the university owns the intellectual property, the technology-transfer office works with the principal investigator to develop the best possible deal to benefit the university and the inventor, both of which receive licensing revenues.

Personnel in the office also interact with industry representatives to set up collaborative research agreements, to inform them of new inventions, and to negotiate license agreements. The office also advises venture-capital groups of new inventions and facilitates the start-up of new companies, some of which may establish themselves in the Audubon Business and Technology Center, a Columbia incubator. The office will help researchers to develop a business plan that will be attractive to potential investors. STV also has some discretionary funding to support faculty who might develop new patentable technologies

STV provides guidance about how researchers can protect their inventions. If results of research are made public without first being protected -- with a confidentiality agreement, materials-transfer agreement, or patent application -- the monetary value of the invention is substantially reduced.



B. Grants-and-contracts offices

Columbia University has two offices for dealing with grants-and-contracts administration – the Office of Projects and Grants http://www.columbia.edu/cu/opg/, on the Morningside campus, and the Office of Grants and Contracts http://www.cumc.columbia.edu/research/ogc-pers.htm, on the Medical Center campus. Grants-and-contracts offices in other universities perform similar functions. The Columbia offices are authorized by the Trustees of the University to submit sponsored project proposals to all agencies (whether governmental or private) and to negotiate and accept awards based on these proposals. The Columbia grants office negotiates grants with industry, except for the intellectual-property clauses, which the technology-transfer office manages.

If a Columbia researcher enters a collaboration with a researcher in another department while a grant is ongoing, they do not have to formalize the relationship through the grants office. If, however, a Columbia researcher with an ongoing grant enters into a collaboration with a researcher at another institution and money is involved in the transaction, a subcontract is written, which the grants office manages. If collaborators within an institution apply for a grant together, they are both included in the personnel section of the grant. If collaborators from different institutions apply together for a grant, a decision about who will be the prime institution and who will be the secondary institution, obtaining a subcontract, has to be made. Material-transfer and intellectual-property agreements also come into play.

Obstacles in International Research



C. Clinical-trials offices

The mission of Columbia University Medical Center's Office of Clinical Trials, http://www.clinicaltrials-nyph.org/, which opened in 1992, is to further promote the excellence, scope, and efficiency of clinical research activities at the Columbia University Medical Center, the New York Presbyterian Hospital, and the New York Presbyterian Healthcare Network, which includes Cornell University. The office has served as a model for clinical-trials departments at other academic medical centers throughout the country. The office provides many support services to clinical investigators, freeing them to focus on their research.

The major responsibilities of the office include:

Bridging institutional lines, the Office of Clinical Trials represents the network institutions and is the administrative unit that pharmaceutical and diagnostic companies should deal with when exploring the clinical-trial possibilities and when negotiating a Clinical Trial Agreement.



Continue to the next section: → Resources













Reference Notes:

1 Macrina, F. Scientific Integrity: An Introductory Text with Cases. Second Edition. American Society for Microbiology Press. 2000.     back

2 "Free the Academic Drug Tests," New York Times, editorial, Nov. 30, 2004.     back

3 Macrina, F. Scientific Integrity: An Introductory Text with Cases. Second Edition. American Society for Microbiology Press. 2000.     back