Some of the instruments to which I refer include amino acid analyzers, peptide synthesizers, spectrometers, and protein sequencers. In addition to the high cost for the purchase of these types of instruments, there were significant costs associated with their operation and maintenance. Further, this was accompanied by the need for a significant degree of technical expertise to operate these instruments at optimal levels. It became clear to both external and internal funding sources that the existing paradigm of purchasing an expensive instrument with significant capacity for throughput and placing it in an individual’s laboratory was not an effective use of either funds or the instrument. Generally, the data required by any one individual was usually very specific and limited in the number of experiments that needed to be performed. Thus, it was not in the interest of the research faculty to master the technology, given the time and difficulty such a process demands. As a result, there began the development of “shared resources,” “cores,” or “facilities” whose function was the operation of these instruments, typically for a select group of scientists, with the aim of generating a maximal amount of data from the instruments for this group.

By the late 1970s and early 1980s, another wave of instrumentation development significantly influenced the biological and biomedical sciences; these instruments included the solid phase and gas-phase protein sequencers, DNA/oligonucleotide synthesizers, and DNA sequencers. These instruments, in addition to providing analytical and reagent support for targeted scientific fields, were also comparatively high-throughput platforms that solidified the concept of “shared resources” to broaden the cost-basis for operation by utilizing maximal throughput. Over the past two decades, there have been a variety of enhancements to those technologies as well as new technologies and platforms, such as instrumentation for gene expression profiling via DNA/oligonucleotide microarrays, proteomics utilizing mass spectrometry, and numerous front-end sample preparation technologies (e.g., 2D SDS-PAGE, orthogonal chromatographies), protein array technologies, single nucleotide polymorphisms (SNP) analysis and realtime quantitative PCR, just to name a few. All of these technologies were well suited for exploitation under the shared resource paradigm (i.e., high initial cost, high operational cost, high throughput, sophisticated operational expertise required) and further solidified the concept of institutional shared resources or cores in academic and industrial research settings.

The staffing of shared resource facilities in academic settings was generally by non-tenure track research faculty or research associates who were rarely directly involved with the research process other than to perform their specific tasks. Over time, given the everincreasing technological sophistication required for the effective operation of instrumentation, researchers were less able to critically understand the technology, and thus, became more and more reliant on the expertise of the shared resource staff. Interestingly, there seemed to be some reluctance of researchers to effectively “partner” with shared resources and their staff. An analogy of the process is of a sample anonymously pushed through a window in the wall and anonymously the resultant data returned to the investigator without significant interaction between the two parties. Furthermore, there was often little exchange between shared resource facilities within an institution and they often functioned in a scientific and technological vacuum. In addition to this organization scheme not being a particularly effective use of resources, it was not uncommon for duplication of services and facilities to be found within some institutions.

Another difficulty from the human resource standpoint was that many individuals who served in these settings felt like “second class citizens.” Many of these staff, particularly those with higher professional degrees, were left with a feeling that, although their technical skills were generally being appropriately utilized and appreciated, their scientific skills and training were not. Needless to say, this is not an effective management approach for yielding optimal productivity from staff or for generating an environment conducive to job satisfaction.

New age biology, new age cores–advent of systems biology

In 2001, Ideker, Galitski, and Hood published a review that described a new approach for thinking about and investigating biological systems that they termed “Systems Biology.”1 The concept is that biological systems should be investigated in a holistic manner by analyzing “the gene, protein, and informational pathway responses; integrating these data and ultimately, formulating mathematical models that describe the structure of the system and its response to individual perturbations.”1 The genesis for this concept is unclear, but one can speculate that given the very close relationship and appreciation that Hood has had in the development of instrumentation and its role in the generation of reagents for the investigation of molecular systems as well as their analysis, he must have felt that the fields of biomedical and biological sciences now had the analytical armament to mount a systems approach to studying biology, and hence, it was appropriate to launch this field.2,3 Embarking on such ambitious investigations demands effective evaluation of the projects from a multidisciplinary approach, often involving scientists with very different backgrounds. Such studies that utilize a systems approach require the design of experiments that move beyond the traditional boundaries of the typical institutional shared resource core. If institutions and their investigators are going to pursue a systems approach to study biological and biomedical questions, the current, often observed, paradigm of the shared resources functioning in the institutional research process by generating high throughput datarich information as isolated technological and intellectual islands must evolve. Such an evolution must address several key features.

Shared resource integration, computational biology and bioinformatics 

In most large research institutions, the day of “mom and pop” cores with one or two staff members operating in a scientific and administrative vacuum is no longer economically or scientifically viable. As biological studies shift from a reductionist to constructionist approach (i.e., the prelude to systems biology), there is an obvious need to integrate data from a number of technological platforms and experimental arenas.4 Biologists need to collaborate widely with other scientists with different training in order to thoroughly address biological systems. Consider the advent of “-omics” – genomics, proteomics, interactomes, etc.; most biologists at the cutting edge of their field of interest in fact often approach their investigations exploring a variety of these “-omic” representations of a biological system. The obvious next step as outlined by Spence and Aurora4 is the integration of data generated from the technological platforms that analyze the various “-omes.”

Enter the disciplines of computational biology and bioinformatics. As the pressure to generate value-added information from these data-rich resources grows, many shared resources are beginning to develop computational and bioinformatics capabilities within their laboratories. This is an excellent initial approach to the problem, but as one can see, it is only temporary in that a second order value-added approach is ultimately required to support systems biology. This involves true integration across data generating platforms, and hence, across institutional shared resource boundaries. Thus, there is a need for a centralized computational biology/bioinformatics resource. These resources would function as a conduit from the shared resources to manage large databases, explore effective means for their integration to relational databases, and generate new, value-added/synergistic data from these to support a systems approach to biological studies.

Institutional role in evolving shared resources to support a systems approach to biological studies

As mentioned above, shared resources, particularly in academic settings, have traditionally functioned under the administration of the founding department or institutional research consortium, or in some cases, as a loose confederation of cores administered centrally from the institution’s office for research. Recently, there have been a number of factors that have favored the movement to a more centralized, integrated approach to shared resource administration and management. First, and not surprisingly most effective in this process, are fiscal pressures. Most academic and industrial institutions are well aware of the critical importance of research infrastructure in the pursuit of innovative science. Research infrastructure plays a key role in investigator recruitment and retention in that most scientists recognize the importance of having such technological capability as key to the successful funding of their research proposals and execution of the experiments proposed. As noted, instrumentation for biomedical and biological research is becoming increasingly expensive to purchase and operate. Furthermore, institutions that are heavily vested into scientific instrumentation and technology are faced with the never-ending requirement of instrumentation up-grading and replacement on a four to eight year cycle. On the academic side, funding for such instrumentation is becoming increasingly difficult to secure. The traditional governmental sources for instrument funding, such as the National Center for Research Resources within the NIH and similar instrumentation programs at the National Science Foundation, have seen their ability to fund instrumentation decline due to increasing demand, increasing costs for instrumentation, and only modest increases in their budgets for such programs. Thus, institutions are forced to seek additional sources for funding instrumentation. All of these pressures may serve to galvanize institutions into revisiting their organization, administration, and management of shared resources so that they are more cost effective and are positioned to provide a systems approach for their researchers to investigate biological and biomedical problems.

One approach to integrating institutional shared resources is to consider an organizational structure for the shared resources without borders. What is meant by this, is that rather than an organizational structure represented by technological quantum that is somewhat isolated from another, a more fluid vision of the cores be taken that self-organizes shared resources with the experimental approaches investigators may take for a systems-type project. A simple example to demonstrate this concept would be the strategic combination of two traditional cores, such as proteomics and the other functional genomics utilizing microarray technology. Both make use of platforms that generate tremendous amounts of data. From the biological perspective, most investigators would have interest in knowing the relational status of a system’s proteome and transcriptome. 5 In this situation, these cores can be virtually integrated by computational and bioinformatics tools to generate relational databases and value-added output for the investigators. Other integration points between additional platformbased shared resources can be imagined. The overall anticipated outcome for these approaches is two-fold: a synergism of expertise found in shared resources and an enhanced valueadded product for the investigators, providing them with a more integrative, constructionist view of the biological system they are exploring. Furthermore, from a fiscal perspective, one can envision such approaches would lower costs in terms of shared resources, including staffing, equipment, instrumentation, management, etc., as well as potentially lowering the costs for providing such services to client investigators, a key factor in this time of static real-dollar support for biomedical research.

Effective shared resource staff utilization 

As noted above, integration, where feasible, of shared resources includes staffing. Rather than developing hardboundaries around cores represented by organizations, management, staff, and technologies, institutions should explore integration sites among all of these factors. Fortunately, there does appear to be some changes in attitudes held by institutional investigators toward the scientists and technologists who staff shared resources. Researchers are becoming increasingly aware of their limitations in understanding the complex technological nuisances associated with many of the instruments used in current biomedical and biological investigations. Often, investigators do not fully appreciate what specific instruments can or cannot do, what appropriate scientific questions can be addressed by such instruments, what the format of the data may look like, and how that data can be effectively utilized in the resolution of the scientific question they are studying. Furthermore, most investigators simply do not have the time to invest in learning about the diverse set of instruments they would like utilized in their studies. Thus, many investigators are beginning to integrate shared resource staff, and consequently, their knowledge and expertise into the research process employed in their laboratories. The most effective use of such staff captures their expertise in the design of experiments, specifically in their development and execution, analysis of data, and computational and informatic resources. Ironically, this use of resource staff in a research process stream such as that described above gives rise to new ways of thinking about the management of resource staff and how they and their services are compensated.

Traditionally, many shared resources operate on a “fee for service” basis with compensation for staff directly based on their contribution to the specific service or product. The paradigm described above, where there is an increasing intellectual input into the scientific process by the shared resource staff member, presents a more difficult situation in terms of compensation. In the academic world, intellectual input into a process or the institution is highly esteemed but difficult to evaluate, thus, the fee for service model begins to breakdown.

One approach for resource staff as they become more actively engaged in the research projects of principle investigators is to have compensation for their efforts appear in the budgets of investigators’ research proposals. A different approach would be for the institutions themselves to recognize and support the role of resource staff in terms of their close participation and integration into investigator’s research projects and thus provide compensation for them not covered under the traditional fee for service model. Whatever the case, the traditional role of shared resource staff both in terms of their function in the research process and the mechanisms for supporting that role is changing; institutions should explore methods that compensate and validate resource staff in these changing roles, which leads to my final point, resource staff career tracks.

Developing the research resource career track

The value of career research resource staff seems to be better appreciated in the realm of industrial research compared to academic research. In industry, resource staff usually have a clearly defined research track that outlines what the career track is and how one can progress in that track. Most often, that is not the case in academia. Resource research staff, even those who hold higher degrees, generally do not have faculty appointments. Often they are on a non-faculty research appointment that usually has no real career track associated with it and no defined procedure or expectations for promotion. Perhaps in the past this has been an acceptable staffing model for scientists and technologists in resource laboratories, but with the increasing scientific, technological and intellectual demands placed on staff in shared resources, I would argue that, at best, it is outmoded and, at worse, institutions with this approach to resource staff will not attract or retain the best personnel, nor will they be able to achieve the type of integrated core structure that is critical for a systems approach to biological studies. One model for addressing these issues is the one my institution, the University of Virginia, has adopted for the staffing of senior leadership positions in our shared resources. Approximately ten years ago, the University of Virginia recognized the importance of the scientists and technologists in our shared resources to the research mission of the institution and that to fill these slots, a meaningful career track, in addition to equitable compensation, was needed. In light of this, our institution developed a Research Faculty for Service career track ( Although this is a non-tenure track, it is different from the institution’s research faculty career track with different expectations and milestones for promotion that are specifically tailored to the job functions faculty have when employed in shared resources. Certainly one can imagine other career track models that would be effective in shared resources, but whatever those may be, I would argue that any model which does not provide a dynamic career track with clear job expectations and guidelines for promotion will prove difficult to populate with engaged, vibrant scientists and technologists who enjoy and gain satisfaction from employment in a shared resource environment and wish to play an active role in the institution’s research process.

National research resource centers

The National Institutes of Health National Center for Research Resources (NCRR) supports approximately 40 biomedical technology resource centers throughout the United States along with its Biomedical Informatics Research Network (BIRN). The concept behind these programs is to offer the best technology and intellectual support to biomedical researchers; not unlike the philosophical underpinning of most academic shared resources. Thus, this calls into question the possibility of duplication of effort, resources, etc. Often, it is true that the resources available at the national centers may surpass those of the individual institutions, both in regard to instrumentation and perhaps expertise and one may question the value of the local efforts to develop and sustain institutional research resources in the face of such national centers. However, most institutions have come to realize the value of proximity in terms of such support. There is as yet, no totally satisfactory substitution for readily accessible, local technology and expertise provided under the collegial paradigm of an academic institution. Certainly the national centers are very important for some types of projects, most notably large scale or extremely complex investigations, but for the most part local support appears to be preferred by investigators. Therefore, in the conceivable future one may expect continued research support that is provided by both national centers and local shared resources. An interesting consideration is how interactions between such research support providers are integrated with the overall enhancement of product from both types of shared research facilities. What first comes to mind is integration at the level of computation, informatics, and databases. It will be interesting to follow these developments in the arena of the national centers. It is unlikely that institutions will close their shared resource support in favor of some distant national center. On the other hand, as new complex, extremely expensive instrumentation becomes available, it is likely that the paradigm of the national centers will remain. However, one might argue that these centers, in addition to providing support to investigators involved in complex investigations, should also be reaching out to provide support for local shared resources, perhaps an equally valuable function and a further justification for their existence.


Shared resource laboratories in both academic and industrial settings have played a very important role in the biological and biomedical sciences for the past 40 years. Traditionally, these resources, within institutions, have been somewhat insular in regard to their interactions with both other resources as well as the investigators to whom they provided services and reagents. With the advent of high-throughput platforms that generate large data streams from a variety of biological molecules and biological experiments, a systems approach to exploring medical and biological questions is becoming possible. However, for that to become practically feasible there must be an evolution in shared resource organization, management, and staffing that takes advantage of integration sites among shared resources for data integration, analysis, and value-added output in a systems format. Institutional leadership that wishes to promote a systems approach for their biological and medical research missions should begin to explore how to provide an organizational infrastructure for shared resources to evolve such that they can effectively become part of the scientific process, and hence, support such types of experimental approaches.

For more information on shared biomolecular research resources, the Association of Biomolecular Resource Facilities provides an excellent entry point into this arena of biomolecular sciences and shared resources (http.//

Acknowledgements: I would like to thank Dr. Nancy Denslow, University of Florida, for her suggestions and comments in writing this article.

  1. Ideker, T., Galitski, T. and Hood, L. (2001) “A new approach to decoding life: Systems Biology.” Annu. Rev. Genomics Hum. Genet. 2:343-72.
  2. Hunkapiller, M. W. and Hood, L. E. (1980) “New protein sequenator with increased sensitivity.” Science, 207:523-5.
  3. Kaiser, R. J., MacKellar, S. L., Yinayak, R. S., Sanders, J. Z., Saavedra, R. A., and Hood, L. E. (1989) “Nucleic Acids” Res. 17:6087-102.
  4. Spence, P. and Aurora R. (1999) “From Reductionist to constructionist, but only if we integrate.” Trends in Pharmainformatics (Suppl). pp. 37-39.
  5. Stein, L. (2003) “Integrating biological databases.” Nature Rev. Genetics 4:337-346.

Jay W. Fox, Ph.D. is Professor of Microbiology, Asst. Dean for Research Support, at the University of Virginia, Department of Microbiology, PO Box 800734, University of Virginia, Charlottesville, VA 22908-0734. Jay can be reached at email: