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 25
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research 2 The Integral Role of Theory in Biology He who loves practice without theory is like the sailor who boards ship without a rudder and compass and never knows where he may cast. —Leonardo da Vinci (http://www.brainyquote.com/quotes/authors/l/leonardo_da_vinci.html) This chapter describes several different ideas about scientific theories, emphasizes the diversity of theoretical activities throughout biology, and discusses ways in which theory is integral to each specific kind of scientific activity, including experimentation, observation, exploration, description, and technology development as well as hypothesis testing. Biologists use a theoretical and conceptual framework to inform the entire scientific process, and they frequently advance theory even when their work is not explicitly recognized as theoretical. Explicit recognition of the many entry points of theory into the scientific enterprise may provide greater opportunity for developing new concepts, principles, theories, and perspectives in biology that would not only enhance current scientific practices but also facilitate the exploration of cross-cutting questions that are difficult to address by traditional means.
OCR for page 26
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research THEORY AS PART OF THE PROCESS OF SCIENCE A good scientific experiment, like a good story, has a beginning, a middle, and an end (Galison, 1987). It is satisfying to describe the scientific method as a linear narrative beginning with hypotheses to be tested and then proceeding to experimental design, execution (funding, equipment and material procurement, set-up and manipulations, measurement and data collection, compilation of results), evaluation of evidence, and formulation of new hypotheses. In the occasional blockbuster scientific story, this process culminates in the emergence of a transformative new insight into nature—the recognition of the cell as the basic unit of life, of mitochondria and chloroplasts as evidence of past symbioses, of plants’ ability to turn CO2 and sunlight into O2 and sugars. This is rarely the way it happens, however. Real empirical practices turn out to be a good deal more complicated and a good deal less linear. The traditional story of scientific method leaves as a mystery the important question “Where do new hypotheses come from?” But like a bad television screenplay, the mystery is dissipated by focusing the plot elsewhere, on the problem of confirming or falsifying hypotheses—the logic of justification—rather than the psychology of discovery (Popper, 1959). Each of the steps in this narrative is treated as a black box, when in fact both historical contingency and scientific judgment (in other words, the theoretical and conceptual framework within which the scientists are operating) are at work throughout the narrative, connecting the testing of hypotheses with the generation of new theory. For example, the technologies, protocols, and instruments that are chosen as means of experimentation also appear to have “life cycles.” Their endings or disappearance, like experimental methods in the broad sense, can come from anything from a change of interest, to new discoveries that render them obsolete, to new inventions or procedures that replace them. Decisions to use new instruments, to carry out experiments in new ways, or to take notice of odd or puzzling results do not come out of nowhere but instead are informed by the scientists’ theoretical framework. The ways in which experimental approaches evolve again hints at more complexity than the standard plot allows. Scientific observation is likewise complex, although it is often thought of as no more than merely “looking.” To count as observation in science, “looking” usually requires a sophisticated approach, involving instruments and elaborate protocols embedded in technical practices that frame and shape both the observations and the reports of the results (Hacking, 1983). The things scientists want to observe are rarely easy to see, hear, taste, smell, or touch unaided by instruments or concepts. The things biologists want to observe are not only complex in their own rights but are embedded in complex structures or communities. Indeed, merely choosing what to
OCR for page 27
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research observe—and how—is, in fact, profoundly affected by the theoretical and conceptual framework of the observer. Scientific observation is, in other words, as much a matter of thinking in the right way as of looking in the right direction. “The early bird gets the worm,” first and foremost because she had the idea to get up early to see if the worms might be more plentiful then. Indeed, observation is fully as active and interventionist as experiment and, in the right context, observation can be experimental because the essence of experiment is not manipulation but rather comparative judgment (Bernard, 1865). Experiment, technology development, and observation all seem to be clearly and familiarly embedded in complex social and technical practices involving people with varied skills, interests, and backgrounds and can appear to be divorced from theory. Theory seems to be different and abstract, the product of purely conceptual work to formalize empirical knowledge achieved by science, rather than a living part of the material practice and process of science. Indeed, theory is often described in opposition to practice. The word “theory” can be used to describe many different things. It can mean an idea behind a hypothesis or the status quo to be challenged; a speculative glimmer of an idea before anyone has tested it; or a well-confirmed, authoritative idea that expresses nature’s laws and provides explanations, unification, and means of control after a community of experimenters, observers, and technologists have done their work—but it is infrequently seen as an integral component of each step of the scientific process. Despite this common impression that science is a process and theory its product, however, theory does not merely describe, codify, and enshrine scientific knowledge. It does all of that, and much more, but it cannot be easily dissected out from the body of the scientific enterprise. The many uses of the word “theory,” in science as well as in popular culture, not only suggest that theory involves a rich set of practices and processes but also reflect the complexity and variety of theoretical work in science and its value to society more broadly. A TALE OF TWO THEORIES The word “theory” serves so many purposes in the English language that confusion is almost inevitable. While anyone who has taken a high school science course has been taught that the word “theory,” when used in science, means more than a hunch or an unproved idea, there is nevertheless the tendency to think that some scientific “theories” are more established than others. For example, theories that include mathematical equations and describe a range of physical phenomena that most people have experienced, such as those describing motion or the behavior of gases, are sometimes seen as rising above the designation “theory” and achiev-
OCR for page 28
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research Box 2-1 Stephen Jay Gould on the Theory of Evolution Well, evolution is a theory. It is also a fact. And facts and theories are different things, not rungs in a hierarchy of increasing certainty. Facts are the world’s data. Theories are structures of ideas that explain and interpret facts. Facts do not go away when scientists debate rival theories to explain them. Einstein’s theory of gravitation replaced Newton’s, but apples did not suspend themselves in mid-air, pending the outcome. And humans evolved from apelike ancestors whether they did so by Darwin’s proposed mechanism or by some other, yet to be discovered. Moreover, “fact” does not mean “absolute certainty.” The final proofs of logic and mathematics flow deductively from stated premises and achieve certainty only because they are not about the empirical world. Evolutionists make no claim for perpetual truth, though creationists often do (and then attack us for a style of argument that they themselves favor). In science, “fact” can only mean “confirmed to such a degree that it would be perverse to withhold provisional assent.” I suppose that apples might start to rise tomorrow, but the possibility does not merit equal time in physics classrooms. Evolutionists have been clear about this distinction between fact and theory from the very beginning, if only because we have always acknowledged how far we are from completely understanding the mechanisms (theory) by which evolution (fact) occurred. Darwin continually emphasized the difference between his two great and separate accomplishments: establishing the fact of evolution, and proposing a theory—natural selection—to explain the mechanism of evolution. He wrote in The Descent of Man: “I had two distinct objects in view; firstly, to show that species had not been separately created, and secondly, that natural selection had been the chief agent of change…. Hence if I have erred in … having exaggerated its [natural selection’s] power … I have at least, as I hope, done good service in aiding to overthrow the dogma of separate creations.” SOURCE: Gould (1994). ing the status of “laws.” Thus the “theory” of evolution, which describes a process of change that is ubiquitous but less often recognized as part of everyday experience than is the steam from a kettle or the acceleration of an object falling to the ground, is seen to be somehow less demonstrably true or scientific than the “theory” of gravity. However, from a scientific point of view, the two theories have equivalent goals in the sense that both seek to explain and interpret a set of facts. As Stephen Jay Gould memorably wrote (see Box 2-1), “Facts do not go away when scientists debate rival theories to explain them.” The phrase “evolution is just a theory” reflects this tendency toward invidious comparison with well-established laws of “real” sciences like physics. Such a view of evolution might have been apt in 1838, soon after
OCR for page 29
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research Darwin’s return to England from his five-year voyage on HMS Beagle. At that time, Darwin wrote his private “D Notebook” on transmutation while reflecting on the implications of Malthus’s idea that population growth inevitably outstrips food supply, a full 20 years before publishing On the Origin of Species (Darwin, 1859). “Just a theory,” “a hunch,” or “an educated guess” can certainly mark the beginning of a theoretical enterprise, which in Darwin’s case blossomed in a wealth of investigations from the 1830s to the 1870s, followed by the work of evolutionary biologists for more than a century since his death in 1882. Darwin’s core principles of his theory of “descent with modification,” that is, his mechanism of evolution by natural selection—variation, fitness, and heritability—were first articulated in his “E Notebook” on November 27, 1838 (Barrett et al., 1987). Together with Malthus’s principle of population, they form the conceptual core of a theory as profound, as central to biology, and now as well established as Newton’s theory of motion. Scientific and public reactions to Darwin’s theory upon its publication in 1859 took it to go “beyond the facts,” as was Newton’s widely attacked “occult” principle of gravity after its publication in 1687. But evolutionary theory has moved beyond Darwin’s early insights, just as physics has moved beyond Newton’s. Curiously, Newton’s “laws of motion” are no less celebrated (nor less useful) for having turned out false (in the wake of relativity and quantum mechanics), while the scientific credentials of Darwin’s theory continue to be doubted despite its continuing success in guiding empirical research in a wide variety of biological sciences. It is interesting that at least some physicists no longer describe physical theories in terms of “laws of nature,” noting that even such a “well-tested and well-established understanding of an underlying mechanism or process,” as the standard model in physics unifying strong and electroweak interactions among fundamental particles, “can never be proved to be complete and final—that is why we no longer call it a ‘law’” (Stanford Linear Accelerator Center, 2007). Though dismissive claims about major scientific theories still play a role in popular debates about the place of science in society and culture, they have little influence on theory development in the sciences, other than as warnings against rash speculation, hasty generalization, and delusions of grandeur at the beginning of a line of theoretical work. It is necessary to look beyond common usage and popular stereotypes to understand the role of biological theory in contemporary science. To improve our understanding of the role of theory in biology, the view of theory needs to be expanded beyond the traditional concept of a “law of nature” to one that illustrates how the variety of theoretical practices and modes of representation, explanation, and prediction in biology reflect the complexity and diversity of the phenomena that the theory studies. It is important to have a rich concept of theory and the theoretical enterprise
OCR for page 30
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research in order to understand the many roles of theory in the advancement of biological science, in facing the grand challenges of 21st-century biology, in evaluating how best the biological sciences can be integrated with other sciences, and to ensure that students are able to comprehend and appreciate the patterns and processes behind the wealth of biological facts that are accumulating at an accelerating pace. VARIETY OF MEANINGS OF THE WORD “THEORY” The two extreme definitions of the word “theory”—a speculative idea or a mathematical “law of nature”—both serve poorly as descriptors of the role that theory plays in the science of biology. Equating the word “theory” with “hypothesis” is another source of confusion. More broadly useful is an emerging definition of the word “theory” to mean a family of models. This alternative understanding of the word captures the diverse relationships among theories, laws, hypotheses, and models in modern biology and makes it easier to see that biology is a deeply theoretical enterprise, but not one in which theory is understood in opposition to practice, experiment, or observation or focused narrowly on developing a set of master equations. Theory as Speculation The view that theory is untested speculation is often accompanied by the view that once “proved,” theories turn into facts. Some think of Darwinian evolutionary theory, for example, as mere speculation on grounds that it hasn’t yet proved, by experiment or observation, that natural selection has produced new species of organisms. Others judge evolution to be pseudo-science, claiming that it cannot provide such proof and that, when properly explored, is found inconsistent with the laws of better theories, such as thermodynamics. “[T]heories do not,” however, “turn into facts by the accumulation of evidence” (NRC, 1998, p. 6). Nor should the claim that evolution is a theory (speculative or not) be confused with the claim that evolution is a fact. The fact that life is genealogically organized by descent, with modification, from a common ancestor should not be confused with the theory that the pattern of diversification of life is primarily due to natural selection. Statements about nature state facts if they are true, regardless of whether humans have proved them to be so or not. As has been seen, however, it is not at all obvious that successful scientific theories, such as Newton’s, must be true in order to succeed and be useful. If Newton’s theory is false, then it does not state the facts, at least not in the way popular culture demands. The idea that Newton’s theory is “approximately true,” even while literally false, requires a different account of
OCR for page 31
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research theories than the traditional one in which successful theories state the true laws, or facts, of nature. Theory as Quantitative Laws The idea that mathematical expression is the hallmark of genuine theoretical sciences, while others are simply “less mature,” takes physics as a gold standard to which other sciences must aspire, even though it is not obvious that the aim and structure of successful physical theories are well suited to the phenomena of biology or the social sciences. Examples of important qualitative theories in biology include the circulation theory of the vascular system, the cell theory of living organization, theories of ecological succession, the impact theory of the extinction of dinosaurs, and the theory of evolution by natural selection. Whether qualitative theories such as these win silver or bronze rather than the gold of quantitative theories like Newton’s or Einstein’s is a matter for debate. Nor is it always clear whether mathematical expression of biological theories would better serve science than their qualitative forerunners. The best mathematical biology is strongly driven by clear concepts. The old joke about the theoretical biologist who began a lecture with the words “Consider a spherical cow …” exploits the general lack of understanding of the entry point of mathematical theory. It may or may not be sensible to consider a sphere as a first approximation for the shape of a cow. If the question concerns the phylogenetic relationship of the Bovinae, then the sphere approximation would be laughable, but if the question concerns a calculation of the worldwide release of methane gas due to bovine digestion, then perhaps a spherical approximation might be sensible. Theory as Hypothesis Scientists sometimes use the word “theory” as a synonym for “hypothesis” to mean a claim about nature that is intended for empirical testing. Scientists generally recognize that theories and hypotheses can be well or poorly supported by evidence (facts) and that they must sometimes work with weakly supported theories or hypotheses for lack of something better. A “working hypothesis” is commonplace in science. A theory doesn’t cease to be a theory because it is confirmed, and a bad theory doesn’t cease to be scientific just because it is falsified. More importantly, scientists are well aware of many of the idealizing assumptions they need to make in order to understand, explain, and predict nature and that this means they expect their ideas to be literally false, even if explanatorily productive (Cartwright, 1983; Wimsatt, 1987). Moreover, science is always in process, so scientists can expect theories, hypotheses, and evidence to change over time with
OCR for page 32
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research continued investigation. A static theory is a dead theory, one that no longer drives research. The popular view of theories as final, conclusive, finished, backward-looking codifications of scientific knowledge is inconsistent with the equating of theory and forward-looking hypotheses, since the former are expected to capture the most durable parts of scientific knowledge—laws of nature—while the latter may well fail testing in the next experiment or observation. It is important to clarify the difference between hypotheses and theories. In the traditional understanding of science, one starts with a theoretical framework for the particular system of interest. This framework then provides the starting point for a hypothesis (sometimes an innovative or imaginative or inspired hypothesis) that seeks to explain or predict the behavior of the system of interest. The next step is to observe the system or to perform an experiment. The resulting data are then used to confirm or disconfirm the hypothesis (and perhaps the initial theory). When hypothesis and data agree, the theory is confirmed; when not, the theory is disconfirmed. Theories guide the construction of hypotheses for testing but are not themselves put at risk of falsification by a single observation or experiment. Understanding the role of theories in biology should include the broad organizing function of theories to coordinate and direct whole research programs and provide the basis for explaining broad patterns of empirical phenomena. THEORY AS FAMILIES OF MODELS The limitations of treating biological theories as candidates for universal laws of nature, or grand empirical hypotheses, or even untested speculations can be addressed by adopting a different viewpoint: that theories are collections or “families” of models. A scientific model is a representation of some aspect of nature for a purpose of study (Levins, 1966, 1968; Giere, 1988; Lloyd, 1988; Teller, 2001; Wimsatt, 2007). Most biological systems are too complex to be described by a single model; a family of related models is more appropriate. Modes of representation in models are quite diverse, including verbal, mathematical, visual, and physical. Darwin used words to present evolutionary models, while Robert May used mathematics to formulate ecological models of deterministic chaos. Many molecular biologists and neurobiologists use diagrams to depict causal structure in their models, for example, of how transcription factors regulate gene expression or how neurons interact in brain circuits. Prior to computers, chemists often built elaborate physical models of molecular structures (for more information on modes of model representation, see de Chadarevian and Hopwood, 2004). Models serve as representations because modelers intend them to. This relativity to scientists’ purposes means that models represent nature
OCR for page 33
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research only in relevant respects to limited degrees of accuracy (Giere, 1988, 1999; Teller, 2001). Watson and Crick intended their original wire and metal models of DNA to represent its helical structure in terms of bond angles among the constituent kinds of metal pieces representing the geometrical structure of groups of atoms (purine and pyrimidine bases), but they did not intend to represent the color of atoms as metal gray, the backbone as a continuous, homogenous wirelike strand or made of metal, or the distance between base pairs as several inches (see Giere et al., 2006). Whatever the mode of representation of a particular model, mathematics will frequently be involved in the scientific process of explaining, predicting, or controlling nature. If not in the formulation of the theoretical model itself (or the integration of a family of mathematical models to express a general law), mathematics will be involved in the expression of predictions from the model (as in the use of equations to predict the temperature at which particular DNA sequences will melt into separate strands), or in the aggregation of observations and measurements into useful data sets (as in the population sciences and increasingly in global databases in the molecular sciences), or in statistical procedures to evaluate the test of a hypothesis, or in the design and operation of instruments and computer simulations. A diagram might represent the causal path in a biological mechanism, for example, of the impact of predators on prey in population ecology, or the distribution of characters in a phylogenetic tree, or from a neural circuit to a particular behavioral output. To understand the dynamic operation of such causes, mathematical representations are usually necessary and often mathematics is needed to build a visual representation from data in a database. Increasingly, videography is used to capture dynamic aspects of natural phenomena visually and animation can be used to display dynamic aspects of structural models. At a minimum, mathematical tools are needed to develop and use these visual display technologies, since most are computer based, and to depict empirical data stored in databases. Of all the skills required to do biology, mathematical and computer skills may require the most focused and sustained attention by the K-12 and university education systems and in the continuing education of successful scientists. Quantitative approaches are a critical link between theory and other biological practices. The traditional view of theories, built around the reductionist ideal of the most powerful explanations emanating from the lowest levels, anticipates a single, general, realistic, and precise formal representation in a master equation for a given domain (or even for all of science). There is actually no single best, all-purpose model for any natural phenomenon (Levins, 1968). There can be several, even incompatible, models of the same phenomenon because each can represent separate aspects and our purposes may be quite varied. Teller (2001) points out that physicists sometimes
OCR for page 34
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research model water as an incompressible continuous fluid medium and at other times as a collection of discrete particles. Biologists sometimes model organisms in a population as genetically homogeneous but ecologically variable or, conversely, as genetically variable but ecologically homogeneous (Roughgarden, 1979). Practical purposes and interests typically force scientists into tradeoffs in their models among virtues of accuracy, precision, realism, and generality as well as fruitfulness in stimulating new ideas, testability of hypotheses, and intelligibility of concepts. Accepting such tradeoffs is not a sign of theoretical weakness or limitation, so long as empirical results are tested for robustness to the idealizing assumptions of any given model. Acknowledging tradeoffs, in other words, does not mean that the science is somehow bogus, but rather that the “conceptual engineering” that goes into model building and robust analysis of results is an important and explicit part of the theoretical enterprise (Wimsatt, 2007). Quantitative predictions of the precise abundances of organisms in a model of an ecological community with an unrealistically low number of interacting species might trade off (for reasons of analytical tractability or computational power) against qualitative predictions of increase or decrease with a more realistic number of community members, for example. Computer simulation may bridge that particular tradeoff (facilitating numerical solutions to analytically unsolvable equations and quantitative predictions about many species), but other idealizations in computer programs may limit generality in other respects (e.g., that every simulated member of a given species is assumed to be genetically identical). Computer models of interacting molecular networks that are being developed to understand gene regulation represent a spectrum of approximation methods: from binary state, to Boolean models, to systems of differential equations, to stochastic random models of molecular interactions, and hybrids of all these types. Simplifications are key features of all these models. Levins (1968) conjectured that at most one could maximize two out of three desirable features a model could have: generality, realism, and precision. His point was that our pragmatic interests in biological phenomena, together with our limited ability to work with and understand complex representations, suggest that we may never reach the dream of a “final theory” and that, more importantly, we need to evaluate the conceptual tradeoffs carefully and with much thought if we are not to be led into error. These issues will come up in attempts to construct computational models of the cell that include more and more molecular species, their concentrations, properties, and interactions. Anything can serve as a model for anything else, but whether a model is useful in a particular context depends on the respects and degrees of relevant similarity between a model and what it is intended to represent (Teller, 2001). A fruit fly may (or may not) be a useful genetic model for
OCR for page 35
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research a human, while a Buick may rarely be a useful physical model of a black hole, but either can count as a model, given some specified sense of relevant similarity and some specified or implied degree of accuracy that can guide evaluation of the “fit” of a model to the world. Mathematical models play an especially useful role in most sciences because of their special role in rigorously formulating assumptions and establishing formally the consequences of their operation. Although even very simple mathematical models can exhibit extremely complex behavior—for example, chaos in simple growth models in population ecology or neural network models in cognitive science—often the rigor of mathematical analysis or, increasingly, the power of computation and simulation to extend calculation and reasoning abilities (Humphreys, 2004) is needed to trace clearly the implications of assumptions that cannot be easily interpreted or understood either intuitively or verbally. One virtue of understanding theories as families of models rather than as laws of nature is that models need not be expressed in mathematics nor even in statements, though language and mathematics are two key ways humans have to communicate relevant similarities. One concrete object (e.g., styrofoam balls on sticks) can represent another (the solar system, a molecular structure). Biologists often talk about “animal models” for diseases or for physiological processes. And laboratory systems of organisms exposed to various conditions have often been taken to serve as models for particular biological processes, such as the flour beetle system (Tribolium species) as a model for ecological competition (Park, 1941; see Griesemer and Wade, 1988) or fruit fly systems (Drosophila species) as models for evolutionary, gene transmission, behavioral, or developmental processes. In other cases, a particular phenomenon serves as a model for thinking about and constructing others, as when a particular set of molecular interactions in the promoter region of a gene are studied and used as a basis for exploring genetic regulatory systems in other cases or more generally (see Keller, 2000). Another particularly useful aspect of recognizing biological theory as families of models is that it sheds light on the very fruitful practice of comparing models. In many situations, for example, formal mathematical models can be crucial in helping investigators determine when their qualitative models actually are adequate. Biologists often come up with “word models” about processes which then are shown to be inadequate when one tries to actually implement a formal mathematical model or construct a computer algorithm. “When things get too complicated for human intuition and language, scientists turn to math and models” (von Dassow and Meir, 2004, p. 245). Building formal mathematical models and running simulations is a tool of experimental work that can be useful as one method for testing the adequacy of our understanding and for understanding how interactions
OCR for page 36
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research among components can give rise to system behavior. As the increasing ease of collecting large amounts of data makes it more and more possible to study system-level interactions, mathematical and computational models are becoming increasingly important to many areas of biology. Quantitative approaches, from formal mathematical models, to simulations, to pattern recognition algorithms, have another very important value: By requiring logical discipline and a formal methodology, they can be a powerful tool in hypothesis development and prediction. In some instances, large data sets can themselves serve as experimental resources. One can argue that the field of molecular biology, for example, “has finally inverted the habit of biological inquiry. Instead of using phenomenology and perturbation experiments to deduce some mechanism, and then uncovering facts one by one to support that hypothesis, modern biologists increasingly turn to large-scale exploration (e.g., DNA microarrays, genome sequencing) to generate a mass of facts whose relevance is eventually established by phenomenology and from which mechanistic understanding might hopefully emerge” (von Dassow and Meir, 2004, p. 245). Large-scale methods vary considerably in their ability to deliver reliable quantitative data. DNA sequencing is highly reliable, while large-scale gene expression data are only semiquantitative and most large-scale interaction maps from yeast two-hybrid assays and other methods are not even reproducible from lab to lab. Dynamical mathematical and computer models are some tools for coping with these ever-growing masses of data, and computational methods can often be used to improve the usefulness of data of variable quality. Importantly, not all of these methods demand mathematically tractable models. Computers can enable researchers to test hypotheses without having to come up with master equations. Monte Carlo simulations, for example, can test thousands of complicated scenarios and provide a different kind of demonstration of the “robustness” of a hypothesis than would a mathematical model. Just as biologists’ theoretical and conceptual frameworks drive their choice of experimental and observational strategies, theory will play a critical role in making the best possible use of large data sets. Indeed, the ability to test hypotheses computationally (experimentation in silico) may be one of the most important future sources of theoretical breakthroughs in biology. The accumulation of biological data and its storage, maintenance, and accessibility are challenges today. Theoretical approaches to data analysis are likely to be highly productive but will require scientists, or collaborative teams, that combine biological expertise (both theoretical and experimental) with computational and mathematical competence.
OCR for page 37
The Role of Theory in Advancing 21st-Century Biology: Catalyzing Transformative Research CROSS-CUTTING QUESTIONS In the course of subsequent chapters, it will be made clear that there are many theories, concepts, and principles that operate at the many levels of organization that biologists now study, on timescales from the picoseconds (10−12 s) of vibrational state changes of biomolecules to the 4.5 billion year history (1017 s) of planet Earth, and on size scales from elementary particles such as the electrons (10−15 m diameter) that are exchanged in biochemical reactions to the planet itself (107 m diameter), the physical characteristics of whose surface and atmosphere have been profoundly affected by life, from the evolution of oxygen-generating life forms billions of years ago to anthropogenic climate change today. A model-based view of scientific theories complements the traditional view of (correct) scientific theories as sets of (true) statements of laws of nature, enriching our understanding of the theoretical enterprise and its multiple roles in empirical biology. If there are universal laws of nature, they are as likely to be discovered through study of a variety of models as by a direct search for them. The production of a variety of models to explore a given biological phenomenon from different perspectives creates opportunities, and deep need, for renewed attention to theory and support for theorists willing to question basic assumptions and standard approaches. Support for theoretical work in science, because of theory’s many entry points into biological practice, may require investment in both low-risk traditional as well as high-risk radically transformative approaches, since the robustness of empirical results to the idealizing assumptions of conventional models cannot properly be evaluated without worthy alternatives to compare. This report frames a series of questions about life that cut across established disciplinary perspectives while drawing on shared principles or theories that are central to all biological subdisciplines, including basic principles of evolution (life is descended from a common ancestor and natural selection is a key mechanism of change), of cell biology (all life is made of cells), and of heredity (specific evolved mechanisms of intergenerational information transfer account for genealogical relationships).