In his treatise elsewhere in this issue, Costello1 has raised a series of extremely important questions that can perhaps best be summarized into one question: “How should we train the current generation of biomedical scientists to make the best use of the tremendous gains in molecular technologies over the last few decades to address complex biological problems?” Costello raises the concern that our current biomedical graduate trainees are too focused on molecular technologies, claiming that our training programs are generating “researchers” or “super technologists” who are narrowly focused on data accumulation through technological approaches. True “scientists,” Costello argues, are a dying breed whose research skills are complemented by a broad perspective of the origin and history of science and the organs/systems necessary for significant scientific discovery, as was expected of trainees several decades ago. He further proposes that the impact of a narrowly trained investigative community has serious implications on the future direction and funding of biomedical research. Our response to Costello’s call-to-arms begins with an assessment of the requisite skills to function as a competitive biomedical scientist in today’s research environment.
Nurturing a Scientific Attitude
To cultivate true biomedical scientists, we must endow our graduate and postdoctoral trainees with an intellectual toolkit that complements cutting-edge technologies and fosters scientific curiosity along with the ability to establish collaborative teams to solve important biomedical problems. We must challenge our trainees to think creatively and broadly, ask important questions, and think outside the box, and we must teach them to be fearless in their approaches to answering those questions. By so doing, we will make them scientists, and not researchers or super technicians, as Costello contends, that can effectively navigate an unprecedented technological environment.
Indeed, funding agencies support the scientist mentality. The National Institutes of Health (NIH) has recognized the need to counter the historically conservative approach of NIH study sections. The NIH’s New Innovator Award program encourages the broader qualities of a scientist over those of a narrowly focused researcher. This program is being further extended by the Transformative RO1 program, which supports “exceptionally innovative, high-risk, original and/or unconventional research.”2 The question to the research community is, How can we train our students and postdocs to fit this mold? We can begin by challenging our trainees’ thought processes with frequent informal and formal interactions that motivate them to participate actively in thought-provoking scientific discussions. This does not require additional coursework for students or unduly infringe on time constraints of already overburdened faculty. It simply means that mentors must nurture an environment conducive to scientific inquiry inside and outside the laboratory and classroom.
In this context, we challenge the reader to reflect on how much time he or she spends probing the thought processes and scientific reasoning skills of trainees. Simple activities, such as having trainees outline the logic of an experimental design in lab meeting or asking them the more general significance of their research to other fields, help create an intellectual environment that allows individuals to understand the larger context of their studies. In fact, not challenging one’s students and postdocs in such a manner is abdicating one’s most basic mentoring responsibilities. Challenging students also fosters a mindset receptive to exploring the potential significance of unexpected outcomes that have the potential of generating new paradigms. To paraphrase Isaac Asimov, new discoveries are more often associated with “That’s funny” rather than “Eureka!” We must ask ourselves how much time should be spent in the classroom learning about a multiplicity of subjects versus working in a laboratory, analyzing data, and learning how to be a scientist. It is debatable whether the classroom is the most effective environment for affecting scientific training, as many students view coursework as the beast that must be slain, as a detour that confounds progress in their research. What better way to circumvent and change this attitude than to incorporate “science”—the broad historical, philosophical, and biological context of trainees’ research projects—into the trainees’ everyday activities?
However, intellect without tools is insufficient for testing a hypothesis experimentally. It is essential that our trainees have a firm grasp of the strengths and limitations of contemporary molecular and cellular biology technologies to understand their potential to help solve biomedical problems. However, these tools are not the Holy Grail of hypothesis-driven investigation but, rather, robust techniques to help one address the molecular underpinnings of an important biomedical problem. This is not to imply that acquisition of fundamental molecular knowledge is unimportant or on the decline but that, even on the molecular level, there is a necessity to think beyond the immediate data to determine the higher-order relationships (e.g., regulatory networks and pathways). For example, in the early days of targeted gene knockout mouse models, many would conclude that a gene was unimportant in the absence of overt phenotypic alterations. This simplistic interpretation is characteristic to the supertechnologist way of thinking, but the fact that there is now a better understanding of molecular complexity and redundancy is testament to the strength and persistence of scientific influences.
Costello has raised another issue of concern to all of us in the biomedical research community—the lack of an understanding of physiology, pathophysiology, and systems biology. This deficiency is highlighted even more by the decline in the number of classically trained physiologists and a decline in pure physiology departments within universities. We agree that our trainees need to understand, for example, how the mutation of a single nucleotide can have such a profound effect on the entire human body or how the body integrates all of its signals into a functioning organism. Is there a solution to the problem of such knowledge gaps? The Howard Hughes Medical Institute recognized this problem several years ago and issued a request for applications (RFA) entitled “Med to Grad Programs,” which was focused on creating graduate training programs that combined the basic sciences with the medical sciences. From that RFA arose a group of programs that represent a set of approaches to broaden trainees’ understanding of how to translate a fundamental discovery into a better understanding of a disease process or bring about a new therapeutic approach. That program was so successful that the institute has recently reissued the RFA. Similarly, the NIH under the current leadership has recognized a need to train PhD students to take a broader approach to biomedical research. These initiatives are most laudable and should clearly be continued.
Is there anything in our training programs that might conflict with these goals? We train our students to be highly focused in their pursuits to ensure that they can design appropriately controlled experiments that allow for meaningful conclusions. This might be misconstrued as endorsing formula-type science, whereby a technology dictates the research question, rather than adopting a discriminating use of technologies that best address the problem at hand. Nobel Laureate Dr. Joseph Goldstein, in his 1986 presidential address to the American Society of Clinical Investigation, spoke of the investigator that developed a new technique and then proceeded to ask the same question in different species using the same technique. This investigator published many papers using the technique but never made a significant impact or moved the field forward. Goldstein coined the term “PAIDS” for Paralyzed Academic Investigator’s Disease Syndrome.3 Such an approach will seldom move a field forward and potentially has catastrophic implications on how the data are interpreted, not to mention the catastrophic implications for the individual’s career opportunities. Goldstein’s PAIDS individual may qualify as the stereotype of Costello’s supertechnologist.
A Possible Approach
After identifying the requirements for cultivating a scientific attitude, what is feasible in the training of PhD students? We propose that the impetus for finding a solution to the observed inconsistencies of true scientific training will come from funding agencies and professional associations encouraging the development of multidisciplinary research teams. Indeed, the Graduate Research Education and Training group (comprising the faculty and administrative leaders of biomedical PhD, MD–PhD, and postdoctoral programs) of the Association of American Medical Colleges recognized that the training of research teams is a critical approach to being able to conduct the best possible high-quality research. As a result, they organized a session at their annual meeting in October 2008 entitled “Training Research Teams.” Thus, we would propose that working in and/or leading a research team be considered an essential additional skill in our students’ toolkit. Building research teams will spawn what can best be described as scientific synergism. Such teams will bring together individuals who have significant expertise in their discipline and who can view a subject from different perspectives and broaden the approach. It will bring about critical-thinking skills leading to outside-the-box solutions to pressing biomedical questions.
The era of the individual researcher is undergoing a transformation. If multidisciplinary teams are the wave of the future, as is likely, our graduate students and postdocs must learn to appreciate alternative fields of research and develop the terminology and understanding to be able to communicate across disciplines.
If one agrees that the toolkit for our trainees should contain a mix of scientific techniques and sociologic and philosophic skills, then who shall provide the latter skills? We must begin by example. Faculty should include the historic basis of fundamental discoveries in their lectures. Principal investigators should function as mentors to trainees, occasionally handing out classical or paradigm-shifting papers for discussion at lab meetings. Faculty should undertake giving a lecture or two outside their areas of expertise at least once a year. Mentors should also encourage their students and postdocs to attend clinical grand rounds if affiliated or proximal to an academic medical center. Administrators should provide opportunities for graduate students and postdocs to teach and should encourage faculty to attend and actively participate in student defense presentations. Qualifier exams can be designed to encourage a trainee’s appreciation of systems and networks. These steps would require little loss of critical time, would not detract from current activities, and would go a long way toward inculcating a “systems” appreciation in students’ research.
The question of whether our trainees are equal to, not as well prepared as, or better prepared than those of past generations, as Costello poses, is impossible to answer—today’s funding, academic, and technological environments are too dissimilar to those of the past. However, it is apparent that current trainees are different from their forerunners as a reflection of evolving training programs and powerful technologies. It is up to all of us, from the level of individual faculty to central administration, to ensure that our trainees are given the necessary opportunities and guidance to help them function effectively in a multidisciplinary teamwork environment with scientific reasoning and logic at its nidus. Then, we can say with conviction that our students are trained as PhDs—doctors of philosophy—with the intellectual and technologic skills to aggressively pursue scientific challenges in today’s environment.