Toward an Interdisciplinary Ph.D. Training Experience for the 21st Century

How will the daily life of a research scientist, a research engineer in 2015 differ from that today or ten years ago?

There is a growing consensus (1, 2, 3) that an increasing fraction of the more exciting, pivotal, biomedical research both in academe and industry will be interdisciplinary and will be undertaken by multidisciplinary teams. Answers to important biomedical questions are increasingly within the reach of such team efforts, and beyond the resources of even the most accomplished individual investigator.

Advancing technology drives science, and in so doing provides an expanding menu of relatively lower cost experimental options. They, in turn, expand the range of questions that scientists and research engineers can ask. Today, a majority of the important research in physics is done by large teams. That pattern is beginning to pervade other fields, and is already evident in biomedical research. The independent individual scientist with several dedicated minions has served as the icon of science, especially academic science, for a couple of hundred years. It has served us well and will continue to do so. In the future, however, it will no longer be the norm. The most exciting opportunities, in industry and soon in academe, will be for scientists having interdisciplinary training, and who are comfortable and efficient working in teams. They will have the skills, including critically needed communication skills, to fill a variety of problem solving roles.

Completing an interdisciplinary Ph.D. program within five years, so that it seamlessly launches one into the professional arena in the desired direction (e.g., industry, academe) is a complex, multistep process. It needs to be planned and coordinated with other activities. Milestones need to be set and achieved in a timely fashion if the target is to be reached within the allotted time. The process is not unlike a decathlon that has been designed to span several years. As with the athlete in a decathlon, the emerging scientist needs a coach. The Ph.D. Mentor (defined below) fills that role.

Unfortunately, most biomedical science and engineering Ph.D. programs today are still funded and organized to train graduate students to be classical, independent scientists. Typically, the new graduate student in such a program must select a single faculty member to be their ?research advisor.? S/he then develops a highly specialized research project around an important question selected by or in consultation with their research advisor. After the thesis is completed and accepted, the new Ph.D. scientist typically migrates to a new laboratory at a new location, a University, a company or a research center, to serve and learn under the direction of another individual scientist.

Prof. Hunt and others are implementing a new, quite different research training opportunity, one that builds multidisciplinary expertise along with essential collaborative and communication skills. We aim to foster scientific vision and independence rather than dependence. We aim to sharpen the trainee's creative and problem solving skills. This evolving plan envisions a logical multi-step process:

  • The student, aided and supported by his or her Ph.D. Mentor, select an initial multidisciplinary area as a starting point, likely one that fits with the student's emerging interests. The area may be represented by aspects of current science and engineering that the student feels is most appealing to them. Or it may be the multidisciplinary area represented by aspects of the research and interests of three or more faculty members. For example, the areas of interest may be molecular bioengineering, computer science and pharmacology, or drug metabolism, genetics and cancer.
  • Once the area overlap is clear, the student then identifies questions or problems that are most interesting, most appealing, or most worthy of answering. Specifically, the student builds, revises and narrows a list of interesting research questions that are expected to require knowledge and expertise in the three designated areas. Faculty members, fellow graduate students and postdocs are all typically willing to help. Rotation projects facilitate this process. Experience shows that from such a process at least one broad, exciting research area will emerge.
  • In parallel with the above, one must identify faculty collaborators. During the preceding process two, three or more faculty collaborators will likely emerge. One will be the Ph.D. Mentor. Those identified may be within the same institution (e.g., they are all UCSF faculty members). One, however, may be from outside: a scientist working at a local biotech company or a faculty member at another University.

The next step is to transform the identified, broad problem area into a research proposal that will become the research focus for one's oral qualifying examination, and ideally for one's thesis project as well. Typically that transformation requires undertaking some preliminary research. As soon as one begins the research, one is immediately faced with the realities of being a project manager. Even though the graduate student is the junior partner in the emerging collaboration, s/he must become the project manager. For that role the student needs a degree of independence from each senior collaborator. The successful project is one where each faculty collaborator sees the merits of the project and is thus willing to materially contribute resources (space, reagents, computer hardware &/or software, stipend support, etc.). The student's Ph.D. Mentor will be the science management tutor, and will help insure that all necessary resources are available, and that a reasonable, achievable timeline is available.

No scientist today can have the depth of knowledge and expertise in all of the specialized sub-fields that are relevant to his or her research. For the graduate student following the plan outlined above, it is the research, the science, and the emerging questions that will dictate the areas of specialized knowledge that the student will need to acquire. The Ph.D. Mentor will assist, guide and advise. Collaborators will contribute what is needed but is missing. In so doing they become a partner, a steak holder, in the project. Clearly, the faculty collaborators will have been selected because they are expected to bring considerable additional knowledge and/or resources to the project. At times, additional specialized expertise may be needed.

Traditionally, the new graduate student selects a research advisor toward the end of his or her first year. Next, s/he becomes a member of the new research advisor's laboratory, selects one of the on-going research projects within the lab, and then begins doing research, usually following the directions of a senior graduate student, a postdoc, or the new research advisor. The thesis project typically emerges from research already underway within the laboratory. It is common for the student to become completely dependent on their research advisor. As the end of the Ph.D. process approaches it is that advisor who determines when you have finished your research. The Ph.D. Mentor plan, on the other hand, strives to foster a much different learning and training experience. The Ph.D. Mentor's goal is to work with the trainee (and the trainee's academic program advisor) and the collaborating scientists to develop a research and training plan that extends beyond the Ph.D. program to include the first, critically important, post-Ph.D. career move. The Ph.D. Mentor's role is thus larger than that of the traditional Ph.D. research advisor.

C. Anthony Hunt, PhD
The University of California, San Francisco


The premise of the agent paradigm, its related theory and methodologies together with advances in multilevel modeling of complex systems of interactions opened new frontiers for advancing the physical, natural, social, military, and information sciences and engineering...