|
Items in cart [0] |
Questions? Call 888-624-8373 |
PAPERBACK + PDF PAPERBACK PDF BOOK PDF CHAPTERS Free Resources |
||||||||||||||||
|
||||||||||||||||
|
|
|||||||||||||||
| ||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 17
2
Defining "Large-Scale Science"
in Biomedical Research
The term "large-scale science" is defined and used in many differ-
ent ways (National Research Council, 1994~. The concept can vary
greatly across fields and disciplines, or even across funding agen-
cies; what is "large" for biology, for example, may be quite modest for
space science or high-energy physics. Similarly, a large project in cancer
research may pale in comparison with the Human Genome Project. The
concept may also vary over time, in part as a result of technological ad-
vances. For instance, because of enormous advances in DNA sequencing
technology, the time and cost of sequencing a mammalian genome are
now considerably lower than was the case when the Human Genome
Project (HOP) was launched; thus such projects are becoming le.. likely
to be viewed as exceptional, large-scale undertakings.
T T ~ . . ~ . ~ . ~ `~ ## 1 `` ~ ~ ##
_ _~ _
Untortunately, tne concepts ot large and small science are often
stereotyped in discussions of relative merit. Yet inaccurate ~eneraliza-
J r con
.. . .. .. . .. ~ .. . . . .. ~ .. . . . a.
lions belle tne complexity ot tne terms. it IS tneretore essential to cletlne
clearly what is and is not meant by large-scale science within the context
of this study. For the purposes of this report, a project may be character-
ized as large-scale if it serves any or all of the following three objectives:
· Creation of large-scale products (e.g., generating masses of related
data to accomplish a single broad mission or goal)
· Developing large-scale infrastructure (e.g., generating databases
and bioinformatics tools, or advancing the speed and volume of research
through improved instrumentation)
17
OCR for page 18
18
LARGE-SCALE BIOMEDICAL SCIENCE
· Addressing large and complex but focused problems that have a
broad impact on biomedical or cancer research and may require interac-
tions or collaborations among multiple investigators and institutions
Biomedical research projects are not easily classified as either small-
or large-scale because there is considerable overlap among the attributes
that could be used to define them. Each attribute can be characterized
along a continuum from what is typical for conventional small-scale re-
search to what is typical for a very large-scale, collaborative endeavor (see
Figure 2-1~. Any given project may have a combination of attributes that
fall on different points along this continuum. Large-scale projects tend
to be very resource intensive (where the term "resource" may include
Conventional small-scale research ~ Large-scale ~ Very large-scale collaborative research
Smaller, more specific goals ~ Broad goals (encompassing an entire field of
inquiry)
Short-term objectives ~ Requires long-range strategic planning
Relatively shorter time frame ~ Often a longer time frame
Lower total cost, higher unit cost ~ Higher total cost, lower unit cost
Hypothesis driven, undefined deliverables
Small peer review group approval sufficient
Minimal management structure
Minimal oversight by funders
Problem-directed with well-defined
deliverables and endpoints
Acceptance by the field as a whole important
Larger, more complex management
structure
More oversight by funders
Single principal investigator ~ Multi-investigator and multi-institutional
More dependent on scientists in training ~ More dependent on technical staff
Generally funded by unsolicited, ~ Often funded through solicited cooperative
investigator-initiated (Ret) grants agreements
More discipline-oriented ~ Often interdisciplinary
Takes advantage of infrastructure and
technologies generated by large-scale projects
May or may not involve bioinformatics
Develops scientific research capacity,
infrastructure, and technologies
Data and outcome analysis highly
dependent on bioinformatics
FIGURE 2-1 The range of attributes that may characterize scientific research.
There is no absolute distinction indeed there is much overlap between the
characteristic of small- and large-scale research. Rather, these characteristics vary
along a continuum that extends from traditional independent small-scale projects
through very large, collaborative projects. Any single project may share some
characteristics with either of these extremes.
OCR for page 19
DEFINING "LARGE-SCALE SCIENCE" IN BIOMEDICAL RESEARCH
19
money, space and equipment, and personnel); thus they require collective
agreement or buy-in from the larger scientific community, rather than just
a small number of experts in a subspecialty. To achieve such agreement,
large-scale projects must be mission or goal oriented, with clearly defined
endpoints and deliverables that create infrastructure or scientific capacity
to enhance future research endeavors. Such infrastructure may include
products such as databases and new technologies that could be used as
research tools by a significant portion of the scientific community and
would provide a common platform for research. In other words, a major
intent of such projects is to enable the progress of smaller projects. Tech-
nological advances have created a need for data-rich foundations for many
cutting-edge research proposals that are investigator initiated and hy-
pothesis driven. Thus, many large-scale projects can be thought of as
inductive or generating hypotheses, as opposed to deductive or testing
hypotheses, the latter being more commonly the realm of smaller-scale
research. Large-scale collaborative projects may also complement smaller
projects by achieving an important, complex goal that could not be ac-
complished through the traditional model of single-investigator, small-
scale research. In either case, the objective of a large-scale project should
be to produce a public good an end product that is valuable for society
and is useful to many or all investigators in the field.
Unlike traditional investigator-initiated projects, research within
large-scale projects may be conducted by many investigators at multiple
institutions or sometimes even in numerous countries. Such research is
also often multidisciplinary in nature. Thus, the work may require exter-
nal coordination and management of various complementary compo-
nents. It can also be very challenging to analyze the resultant masses of
data and to evaluate the outcomes and scientific capacity of such collabo-
rative research. Furthermore, these unconventional projects have larger
budgets than most projects undertaken in the biomedical sciences, so it
can be difficult to launch them using the traditional NIH funding mecha-
nisms. In principle, however, the unit cost of collecting data in a large-
scale project should be lower. These projects also often have a longer time
frame than smaller projects, and thus require more strategic planning
with intermediate goals and endpoints, as well as a phase-out strategy.
Within the context of this report, the definition of large-scale bio-
medical science does not include exceptionally large laboratories that are
headed by a single principal investigator who is simply funded by mul-
tiple grants obtained through conventional funding sources. Nor does it
include traditional program (PO-1) grants, in which multiple investigators
are provided funding for independent but somewhat related small-scale
projects. Unlike some other fields, large-scale biomedical science usually
does not entail very large research facilities, such as the Fermi National
OCR for page 20
20
LARGE-SCALE BIOMEDICAL SCIENCE
Accelerator Laboratory for research in high-energy physics. In addition,
large-scale biomedical science is not defined by whether it is basic, trans-
lational, or clinical research, but could entail any of these categories. For
example, cancer clinical cooperative groups may be seen as a form of
clinical large-scale science. The NCI, unlike other NIH Institutes, has set
aside a sum of money to support a large infrastructure to carry out clinical
studies.
Ultimately, the distinction between small- and large-scale biomedical
science is determined by the needs and difficulties entailed in achieving a
given research goal, and by the current capabilities in a particular field.
For example, many traditional investigator-initiated projects in biomedi-
cal research focus on improving our understanding of genes or proteins
that are thought to be of biological interest. In contrast, unconventional
large-scale projects take advantage of economies of scale to produce rela-
tively standardized data on entire classes or categories of biological ques-
tions. Thus, as noted earlier, they may reveal novel areas of research for
follow-up by smaller science projects, and they also provide essential
tools and databases for subsequent research. Large-scale projects may be
the most suitable approach for biological questions that can be addressed
more effectively or efficiently by coordinating the work of many scientists
to produce clearly defined deliverables through the development and use
of advanced technology. Smaller projects are more suitable for address-
ing specific, hypothesis-driven scientific questions, which are essential for
the steady progress and evolution of the field. Such projects are under-
taken by many individual investigators, and often yield unexpected find-
ings that can dramatically alter the course of future research.
Ideally, as noted in Chapter 1, there should be a synergism between
large- and small-scale science in the long term. For example, one of the
frequently cited benefits of the Human Genome Project (HOP) is that it
could facilitate faster, less costly, and easier location and identification of
genes that promote disease when mutated a goal of many smaller con-
ventional science projects. Both large and small science endeavors can
make important contributions to a particular field, and the appropriate
balance between the two may vary over time. Moreover, because bio-
medical research in general is becoming increasingly interdisciplinary
and technology driven, there may be greater opportunities to reap the
benefits of large-scale projects.
EXAMPLES OF POTENTIAL LARGE-SCALE BIOMEDICAL
RESEARCH PROJECTS
Although the number and variety of potential large-scale biomedical
research projects are probably limitless, there are several areas that have
OCR for page 21
DEFINING "LARGE-SCALE SCIENCE" IN BIOMEDICAL RESEARCH
21
been widely discussed and may be more feasible now or in the near
future. In fact, a number of such projects are already under way with
support from a variety of sources, including industry, government, and
nonprofit organizations. Several examples of potential projects in four
areas genomics, structural biology and proteomics, bioinformatics, and
diagnostics and biomarker research are discussed briefly here as a means
of elaborating on the working definition of large-scale biomedical science
used for this report. Some of these examples are discussed in greater
detail in Chapter 3 as models for conducting large-scale bioscience
research.
Large-scale biomedical research differs from many large-scale under-
takings in the physical sciences in the sense that partial completion or
partial success of a project to collect large pools of biological data would
still be useful. As a result, it may be less risky to undertake a long-range,
large-scale project in the biosciences when future budgets are in question.
For example, production of a partial rather than a comprehensive catalog
of protein structures could still be quite useful to the scientific commu-
nity. In contrast, the building of a large-scale facility, such as a super-
conducting super collider or the Fermi Laboratory is useful only if the
facility were completed and then used successfully by members of the
scientific community to generate data. Likewise, the Manhattan Project to
develop the atomic bomb would have been deemed a failure if only par-
tial progress had been made in attaining the ultimate goal.
Genomics
Thousands of people are now working in genomics a field that did
not exist 15 years ago. (For a recent summary of genomics funding, see
Figure 4-3 in Chapter 4~. The completion of the draft sequence of the
human genome is a tremendous achievement, but a great deal of addi-
tional work is needed to realize the full value of this accomplishment.
DNA sequences provide only limited information about a species. Many
additional layers of information, regulation, and interaction must be deci-
phered if we are to truly understand the workings of the human body in
health and disease. Of the many types of biological information, DNA
sequences are among the easiest to obtain but the most difficult to inter-
pret that is, they provide minimal information regarding structure and
function. Thus, the sequence of the human genome in itself does not
reveal the "secret of life," but it is an important tool for answering many
questions in biomedical research.
For example, defining and characterizing the many regulatory ele-
ments in DNA will improve our understanding of how, when, and why
various gene products are generated in both health and disease. The avail-
OCR for page 22
22
LARGE-SCALE BIOMEDICAL SCIENCE
ability of genome databases should facilitate the development of "whole
genome" screens that can be used to assess the expression of all genes in
a given sample or to examine the resulting phenotypes when the genome
is systematically altered to over- or underexpress the genes. There is also
great interest in defining variation among humans with regard to genetic
polymorphisms in disease-related genes and disease modifier genes-
small differences in the DNA sequence of individuals that may not be
directly responsible for disease per se, but may lead to subtle differences
in susceptibility for various diseases, including cancer, or may contribute
to the variability in response to therapies. Polymorphisms can also serve
as markers for locating genes that do directly contribute to disease when
mutated.
Other examples of genomics-related projects include generating data-
bases of full-length cDNAs DNA sequences that are complementary to
messenger RNAs, which actually code for proteins, and thus have inter-
vening "intron" sequences removed. These resources could then be used
as tools to study gene expression and function. This is one of the aims of
NCI's Cancer Genome Anatomy Project (CGAP). There is also great inter-
est in sequencing the genomes of organisms that serve as experimental or
comparative models for biomedical research.
Structural Biology and Proteomics
Structural biology is the study of protein composition and configura-
tion (Burley, 2000~. The term "proteomics" refers to the study of the struc-
ture and function of the "proteome" that is, all proteins produced by the
genome. The expressed products of a given genome can vary greatly across
cell and tissue types, and over time, within the same cell. There are many
opportunities for biochemical modification, regulation, and translocation
between the time when transcription of the DNA into RNA is initiated and
when the final protein product is removed or eliminated from the cells.
Furthermore, proteins do not work alone, but within multisubunit struc-
tures and complex networks; thus there is an immensely sophisticated com-
binatorial complexity to deal with in trying to understand cellular or
organismal function. The pathobiology of disease adds further layers of
complexity that can be quite species-specific. In the case of cancer, for ex-
ample, a great variety of mutations can be found that affect the structure,
interactions, and function of proteins that play key roles in the regulation of
cell growth and survival. Furthermore, the specific mutations present can
vary greatly across different types of cancer, among individual patients,
and even within different tissue layers and cells of a single tumor.
Analogies have been drawn between the HOP and the study of pro-
teomics, but one major difference is the lack of a single objective with a clear
OCR for page 23
DEFINING "LARGE-SCALE SCIENCE" IN BIOMEDICAL RESEARCH
23
endpoint. In the case of the HOP, the goal was simply to obtain a reference
sequence for each of the chromosomes in a human cell. Because there is no
single "human proteome," the endpoint will vary depending on what ques-
tion is being addressed. In the case of cancer, for example, there could be
great value in cataloging and studying the unique proteomes of cancer
cells. Novel forms of proteins, altered interactions among proteins, and
altered responses to normal regulation may be discovered.
Bioinformatics
In many aspects, biology is becoming an information science: many
important questions in biology are now being addressed, at least in part,
through interactions with computer science and applied mathematics.
Scientists can now produce immense datasets that allow them to look at
biological information in ways never before possible. For example, it is
now theoretically possible to study complex and dynamic biological sys-
tems quantitatively (Lake and Hood, 2001~. Once a large resource of bio-
logical data or information becomes available, however, it becomes a chal-
lenge to use that resource effectively. The new field of bioinformatics
aims to develop the computational tools and protocols needed for estab-
lishing, maintaining, using, and analyzing large sets of data or biological
information. Thus, bioinformatics may constitute one key component of a
large-scale research project aimed at generating large datasets that en-
compass an entire field of inquiry. In cancer research, for example, it
would be useful to catalog and characterize the key molecular changes
cells undergo in the transition from a normal to a neoplastic and meta-
static cell. The development of bioinformatics tools and resources could
also potentially serve as a large-scale research project in itself, because the
availability of standardized bioinformatics tools could lead to greater
uniformity and use of data generated within smaller, more traditional
science projects. There is a great need for a common language and plat-
form for many applications.
Diagnostics and Biomarker Research
Much effort has been devoted to identifying and characterizing "mo-
lecular biomarkers" of cancer any change at the biochemical or molecu-
lar level that may provide insight into how a particular cancer will be-
have, how it should be treated, and how it is responding to treatment.
There is also great interest in using biomarkers for early detection, since
some cancerous changes may be detectable by molecular methods before
the cells have had a chance to grow into a tumor that can be detected by
physical methods (usually imaging or palpation). For example, cancer
OCR for page 24
24
LARGE-SCALE BIOMEDICAL SCIENCE
cells can secrete abnormal proteins that might be detected by a blood test.
Many potential markers have been studied over the years, but only a very
few have proven to be clinically useful. However, recent advances in
high-throughput technologies (such as those developed for genomics,
proteomics, and bioinformatics) may make it easier to systematically
search for and assess biomarker candidates.
Patient Databases and Specimen Banks
Collections of archived patient information including clinical data,
family history, and risk factors, as well as patient samples, such as tissue,
blood, and urine can be very useful for studying the genetics, biology,
etiology, and epidemiology of diseases, especially when they are linked.
Such collections of information can also be used to examine the long-term
effects of medical interventions. Once established, these annotated data
and specimen banks can be used to address new questions and hypoth-
eses as they arise. Some of the challenges involved in developing this sort
of research tool, in addition to the high cost, include concerns about scien-
tists' access to the resource, as well as patient confidentiality and informed
consent for future studies. Changing technology can also render older
samples obsolete if the newer methods of analysis require a different
method of sample preservation.
POTENTIAL OBSTACLES TO UNDERTAKING LARGE-SCALE
BIOMEDICAL RESEARCH PROJECTS
Because large-scale science projects may not fit readily into the tradi-
tional molds for biomedical research, there are many factors to consider
and obstacles to overcome in making decisions about whether and how to
conduct such projects in cancer research. A brief overview of these topics
is provided here to elaborate the working definition of large-scale science
in cancer research. Each topic is covered in greater detail in Chapters 4
through 7.
Determining Appropriate Funding Mechanisms and
Allocation of Funds
Buy-in by the leaders of the scientific community as a whole is impor-
tant for the initiation of a large-scale research project, as this mode of
operation is a relatively new concept in biology and has been met with
resistance in the past. There should be some consensus that a large-scale
approach to a scientific problem will add value, and will achieve a given
goal more rapidly, more efficiently, or more completely than would be
OCR for page 25
DEFINING "LARGE-SCALE SCIENCE" IN BIOMEDICAL RESEARCH
25
possible through conventional funding mechanisms. In other words, it
should be clear that to forego a large-scale approach would result in a lost
opportunity to achieve a certain goal, or significant delays and increased
cost in the long run.
Once a large-scale science project has been agreed upon, funding
sources must be identified. The number and variety of potential funding
sources for biomedical research have increased greatly in the last 50 years.
Sources include several government agencies, many private industries,
and nonprofit organizations, each with a different culture, objectives, and
traditions that may cause it to react quite differently to a given idea for a
large-scale project. For example, industry can be expected to take a greater
interest in projects that appear to offer potential near-term profits, whereas
federal agencies are more likely to fund the generation of basic informa-
tion that could be used as a research tool. However, these distinctions are
rarely clear-cut in biology, and thus there is often overlapping interest
and even competition among potential sponsors of large-scale research
projects. In any case, the decision to offer funding and allocation of the
funds are prerequisites for any large and complex project, as the tradi-
tional funding mechanisms in biomedical research were not designed for
such endeavors.
Once funds have been designated for a large-scale biomedical re-
search project, criteria must be established for determining which indi-
viduals, groups, or institutions will be awarded funds for specific compo-
nents of the project. The vetting process for large-scale projects may
require a different set of questions for evaluating the relative merits of
applicants than those commonly raised for smaller projects. In some cases,
more long-term planning than is typical of traditionally funded projects
might be required to define the objectives, feasibility, and expected prod-
ucts of a large-scale project, including intermediate endpoints for measur-
ing progress and assessing accountability. Such long-range planning is
very challenging in a rapidly changing scientific field, and may be some-
what at odds with the nature of scientific exploration and discovery.
Organization and Management
There is no single formula for organizing and managing a large-scale
research project. The approach can vary greatly depending on the goal
and the methods chosen to achieve it. Nonetheless, it can be said that the
organizational requirements of large-scale science projects are likely to be
quite different from those of the more traditional academic approach to
biomedical science. For large-scale projects, the work may need to be
coordinated among multiple public and private research institutions, or
OCR for page 26
26
LARGE-SCALE BIOMEDICAL SCIENCE
across disciplines, funding agencies, and even national governments in
the case of international projects.
The typical U.S. academic research laboratory is headed by a single
principal investigator, who oversees the work of morejunior scientists-
such as graduate students and postdoctoral fellows as well as techni-
cians. There tends to be relatively little hierarchical management of
projects within such laboratories, and little or no management from exter-
nal sources. Because the junior scientists are generally in training to be-
come independent researchers, they should ideally spend much of their
time learning techniques and developing their own independent lines of
research.
Large-scale science projects, in contrast, often require external man-
agement and oversight to some degree so that the work of the participat-
ing groups can be coordinated and kept on track for meeting the program
goals. Once a large-scale project has been launched, it is imperative to
monitor and evaluate its progress against expected milestones, and to
alter course if necessary. When a project requires a multidisciplinary ap-
proach, the potential problems of organization, management, and over-
sight are even greater because of difficulties in communicating across
fields. This situation could make it more difficult to establish priorities
and intermediate endpoints or milestones, which are essential for attain-
ing the ultimate project goal. The ideal manager for external oversight
would thus have extensive experience in all the relevant disciplines; how-
ever, such individuals may be rare because most current training pro-
grams tend to focus on a single discipline.
Because of these challenges, the industrial model of biomedical re-
search may have much to offer large-scale research projects, even when
they are undertaken with public funds at traditional academic institu-
tions. In industry, projects generally involve many layers of oversight,
and teams often specialize in and are responsible for particular methods
or stages of the work. Yet, even large-scale science projects undertaken in
collaboration with industry or through industry consortia may experi-
ence organizational difficulties if they require groups to mesh dissimilar
organizational schemes and cultures.
lo, .
Personnel Issues
The challenges involved in organizing and managing a large-scale
science project include questions of staffing and the training of junior
scientists working on the project. As mentioned above, much of the work
in academic research laboratories is done by graduate students and
postdoctoral fellows who are striving to build a scientific reputation and
career. This is viewed as a mutually beneficial arrangement because stu-
OCR for page 27
DEFINING "LARGE-SCALE SCIENCE" IN BIOMEDICAL RESEARCH
27
dents and fellows provide an inexpensive but highly effective labor pool
in exchange for training and future career opportunities based on profes-
sional recognition for the publications they produce. However, this aca-
demic career track may not mesh well with the goals, products, and
timeframe of many large-scale projects. Students who are assigned to
work on a small piece of a large project may spend many years making a
valuable contribution, but emerge without a significant publication record
on which to base their career advancement. They may also fail to derive
the crucial breadth of training or experience students obtain by working
on and developing a smaller, independent project. As a result, it may be
most appropriate to rely more on technical staff than on students when
undertaking a long-term, product-oriented large-scale project.
Information Sharing and Intellectual Property Concerns
The success of big-science projects in fields such as high-energy
physics has been attributed in part to the fact that the products of the
research have no commercial value, and thus the scientists involved in a
project are quite willing to share results and information (Kevles and
Hood, 1992~. In contrast, many large-scale projects in biomedical re-
search have substantial commercial value, making it less likely that data
and reagents will be freely shared. The potential profits to be gained in
developing new drugs or other medically applicable technologies are
enormous, and many products of large-scale projects can be used as
tools in the development of such drugs and technologies. However,
when many different research tools are needed to develop a clinically
applicable product, aggressive enforcement of patents and pursuit of
licensing revenues associated with those tools could potentially hamper
the progress of research. As a result, there have been many debates
about access to biological data and the merit and appropriate use of
patenting and licensing of research tools in biomedicine (NIH, 1998;
Helter and Eisenberg, 1998~. The challenge is to strike a balance between
patent protection and public access so that institutions are willing to
take the risks and make the commitments necessary to develop new
products with medical and commercial value without significantly im-
peding the progress of research in the field as a whole.
Projects funded by federal agencies may be more apt than those funded
by private industry to rapidly place results into publicly accessible data-
bases and to forego the potential revenues of patenting and licensing the
products of the research. Again, however, there are no absolute distinctions
between publicly and privately funded research with regard to these issues
(Eisenberg, 2000~. For example, some projects funded largely by industry
consortia (such as the Single Nucleotide Polymorphisms [SNP] Consor-
OCR for page 28
28
LARGE-SCALE BIOMEDICAL SCIENCE
tiuml) have policies regarding the creation of unencumbered public-domain
database resources similar to those of the publicly funded HOP. On the
other hand, legislations passed in 1980 has encouraged academic scientists
and others with federal funding to patent their findings in order to facilitate
commercial development of the research.
SUMMARY
The ultimate goal of biomedical research, both large- and small-scale,
is to advance knowledge and provide useful innovations to society. De-
termining the best and most efficient method for accomplishing that goal,
however, is a continuing and evolving challenge. A review and assess-
ment of large-scale science in biomedical research is warranted at this
time because it is a relatively new concept in bioscience in general, and
there is great interest in applying this scientific approach to address ques-
tions in the study of cancer. For the first time, scientists now have the
ability to develop a large infrastructure upon which to base future re-
search. The availability of genome sequences (human as well as model
organisms, such as bacteria, yeast, worm, fruit fly, and mouse) allows for
gene identification, examination of the regulation of gene expression,
cross-species comparisons, and the study of polymorphisms in popula-
tions. Messenger RNA profiles can be generated to study the normal
function and pathology of different tissues. Technology is available to
study the structure and function of proteins, and their dependence on
chemical modification and location within cells. Further improvements in
experimental technologies and the informatics tools needed to process the
information they generate will likely continue to enhance the speed and
scale with which these resources can be generated and put to use.
When is a large-scale approach suitable for biomedical research, and
what can we do to facilitate such efforts when they are deemed appropri-
ate? The goals of this report are to examine the potential contributions of
large-scale science to biomedical research, to identify impediments to ap-
plying the large-scale approach effectively, and to recommend ways of
improving the process for future endeavors should they be undertaken.
1 The SNP Consortium is composed of the Wellcome Trust and 11 pharmaceutical and
technological companies: APBiotech, AstraZeneca PLC, Aventis, Bayer AG, Bristol-Meyers
Squibb Company, F. Hoffmann-LaRoche, Glaxo Wellcome PLC, IBM, Motorola, Novartis,
Pfizer Inc., Searle, and SmithKline Beecham PLC. See .
2 The Bayh-Dole Act and the Stevenson-Wydler Technology Innovation Act encouraged
organizations to retain certain patent rights in government-sponsored research, and per-
mitted the funded entity to transfer the technology to third parties. For more detail, see
Chapter 7.
Representative terms from entire chapter:
science projects
