Systems biology reveals biology of systems



In the last decades, genomic and postgenomic technologies obtained a great amount of information on molecular bases of cell physiology and organization. In spite of this, the knowledge of cells and living organisms in their entirety, is far from being achieved. In order to deal with biological complexity, Systems Biology uses a new approach to overcome this inadequacy. Despite different definitions, Systems Biology's view of biological phenomena highlights that a holistic perspective is needed to integrate and understand the huge amount of empirical data which have been collected. This is one of the aspects that makes Systems Biology so interesting, from a theoretical and epistemological point of view, and that renders it a useful tool to help students approach living beings' dynamics within a comprehensive framework of their biological features as well. © 2010 Wiley Periodicals, Inc. Complexity, 2010


The paradoxical effect of all the very powerful molecular analysis methods developed in the last decades is that they completely falsified the premises they were based on while allowing to open a completely different perspective to biological sciences. The so-called omics tools allowed, for the first time, to get an highly parallel view of the biological systems at the molecular level in which the expression level of thousands of genes (or the concentration of hundreds of metabolites) was simultaneously known on the same biological specimen. These highly dimensional views highlighted a picture completely different from what expected by the molecular biology implicit paradigms1: biological phenomena cannot be interpreted as sequential chains of activation/inactivation of specific genes in a way similar to the instruction manual of our wash machine, on the contrary, the genome (and so the metabolome and the proteome) were discovered to act “as a whole” envisaging a still mysterious but fascinating “biological statistical mechanics,” so asking biologists to come back to the old physiological (the important is the process not the detailed structure) way of thinking. At odds with 19 century physiologists that had to deal with two or three variables at time, nowadays biologists are submerged by massive amounts of data.

A cross-disciplinary field, called Systems Biology, is growing to collect and use the large amount of data produced by new methods for which an interdisciplinary approach and training is also required2. In the last few years, universities and institutions such as Harvard and Massachusetts Institute of Technology have created cross-disciplinary departments devoted to Systems Biology. Around the year 2000, Institutes of Systems Biology were established in Seattle and Tokyo as well. New scientific departments might represent a way to put together scientists having different backgrounds and to increase the interests of the researcher's potential carried out in the department.

Systems Biology studies are posited at three basic levels of definition. The first one is a direct consequence of the sudden increase of biomedical information following the mapping of human genome. Vast genomic databases and technical know-how used to design detailed maps of genes and their transcripts have been created through the huge work of molecular biology in the last 50 years, determining a strong development of the genomic approach in biomedical research especially in the form of differential gene expression studies (DNA microarray) giving rise to the field of transcriptomics. Proteomics is often considered the second step in the study of biological systems. Its focus on proteins with their post-transcriptional regulation processes identifies a complexity level not recognized in the classical molecular biology approach. Finally, the third level has been determined by the study of the metabolome, which represents the metabolic fingerprints of a certain biological system or process and can be considered a sort of “ultimate phenotype”: the representation of the system as a chemical reactor.

Mathematical tools are required to check how the proposed interactions could generate new functions and behavior in a biological system. It was the case when acetaldehyde was proposed to be “responsible” for the synchronization of glycolytic oscillations between individual yeast cells3. A mathematical approach was needed because the relations among molecules are so complex that an immediate intuitive prediction is impossible and because new systemic behaviors also depend on the particular magnitude of the parameter values4.

Therefore, a challenge for Systems Biology has been to integrate omics information to give a more complete picture of living organisms. It is worth stressing the fact that for integration, we do not intend the construction of more and more complicated and refined relation structures between the system elements but the definition of mesoscopic scale observables translating many different microscopic configurations of the system into the same general behavior in analogy with statistical mechanic approaches5. This point is of utmost importance given in many cases (see, for example, Ref.6) the detailed knowledge of the relation structure of a system is of no or little help to predict its behavior.

Computational and bioinformatics disciplines also contribute to allow the study of gene–protein, protein–protein, and metabolite–protein interactions simultaneously, giving rise to a new area in Systems Biology modeling, known as interactomics. This technologically based perspective of Systems Biology gives proof of existence of many simple regulatory circuits that have been demonstrated to be abundant in biological networks7, so that reconstruction of cellular networks and their mechanistic explanations are at the heart of Systems Biology, as conceived by most researchers. Creating predictive model systems will generate “virtual cells” useful to discover subcellular mechanisms of regulation8 or to develop more effective drugs without testing them on patients in the first phases of their validation. Systems-based drug discoveries, for anticancer therapy, for example, allow to discover the location of key elements in the whole process and to observe the global effect of this kind of perturbations through simulations9.

However, the challenge will continue with the integration of this complex and highly diverse information into a conceptual framework: holistic, quantitative, and predictive10. How do entire signaling pathways feed into dense 50 or 60 protein signaling networks with their complex cooperative inputs, outputs, and parallel processing? How do these networks sense, react, and keep our cells in working order? It is even clearer that not genes, nor proteins, nor metabolites but their connections are the key points of main questions addressed by biological science in the last century. Depending on their hierarchical organization, clinical and experimental evidences reveal the complexity of important diseases and biological systems11. Systems Biology might then provide a new outcome, which is not merely a more refined picture, addressing a different level of understanding. From this point of view, Systems Biology can be considered a holistic approach, having the aim to build dynamic models that can simulate and predict physiological behavior or pathology of a complex biological system, moving from empirical data collected through the reductionist approach carried on at the cellular and subcellular levels. That is why, there are those who also stated that Systems Biology is not just a combined application of the aforementioned disciplines to living systems, it has its own unique foundations and methodology, enabling this science to emerge from other disciplines. Its premise is that there is something to be explained in living systems' properties that cannot be understood by molecular biology alone. “Reaching such understanding will require a system biology that is defined as the science that deciphers how biological functions arise from interactions between components of living organism. It studies the gap between molecules and life”12.

Although a sort of contraposition between a methodological and an epistemological approach is often evident, a common feature is shared among Systems Biology's perspectives that highlight the need for rediscovering the systemic dimension of biological phenomena, not only from a methodological point of view but also from an epistemological one. For both these reasons, we suggest that also in the educational field, Systems Biology is a useful approach for underlying the complexity of living organisms as well as for exemplifying how the progress in scientific knowledge periodically leads to open new perspectives, which are based on the previous discoveries but which often require new interpretative paradigms.


The historical background of Systems Biology seems to be quite instructive as it shows how, from different disciplines, the biological problem leads to the building of the same framework: the system vision is revealed to be the central tool to understand living beings. Formal study of Systems Biology, as a distinct discipline, was launched by systems theorist Mihajlo Mesarovic in 1966 with an international symposium entitled “Systems Theory and Biology.” At that time, biochemists were thoroughly studying enzymes and their kinetics, with the aim of understanding the behavior of biochemical pathways on this basis. This can be considered as the first evidence of the need for an integrated approach to biological processes. Metabolic control theory opened a door to comprehend metabolism as a network, which is more complex than the sum of its parts. This helped to overcome simplistic concepts, such as the rate-limiting step, and to examine the contribution of individual components to the network performance.

The theorist who can be seen as a precursor of Systems Biology is Ludwig von Bertalanffy with his General Systems Theory. According to his theory, the dynamics of any system can be explained by showing the relations between its parts and the regularities of their interactions. To fully understand this, it is necessary not only to look at biological system as a unity operating in its internal dynamics but also to see it in its circumstances. From this point of view, the manner in which von Bertalanffy defines the ideas of regulation and feedback in living beings is quite interesting. From his point of view, order is realized by a dynamic interaction of different processes13, overcoming the feedback concept, which is mainly linked with cybernetics and defined as a type of circular action between the different parts of a dynamic system14.

Although mathematicians and physicists always paid attention to the biological world, their involvement in biological research remarkably increased since Crick—fascinated by the structure and properties of the informational molecule of DNA–and Schrödinger—surprised with the apparent inconsistency of life with the second law of thermodynamics—launch themselves in a probable unpopular reflection about biological life and its specific features15. Moving away from the biochemical area, Reinhart Heinrich always taught to search for the principle behind observation, looking for different perspectives, and connecting the abstraction with biological evidence, working on the forefront of System Biology. Besides many other objectives, he developed theoretical approaches for the description and quantitative investigation of signaling pathways16 and developed a metabolic control theory, already in use when this term was not yet coined17. However, a holistic approach, which would have taken into account biological system as a whole, would have to wait a few decades more.

Ahead of its time, in the studies on living systems, Systems Biology has been brought to the forefront with the tour de force efforts of John Sulston and colleagues who, in the late 1970s and early 1980s, set out to determine the entire cell lineage of C. elegans. Work on the worm had the advantage of looking for something “systematic” from the beginning, as exemplified by the conscious choice of that worm as a model18. This systematic modeling approach would be the driving force of many researches also in the field of molecular biology throughout the 1980s, and it continues to be one of the characteristic of Systems Biology's methodology. From then on, researchers adapted many concepts from systems theory to give rise to new interpretation of the experimental results. Relation seems to be the key word that successfully explains biological organization and behavior and concepts already common in biology, terms such as dynamics and feedback loops acquire more relevance. One example of the application of Systems Biology approach for a rereading of results and interpretation obtained by reductionist approach is the work of Ideker et al. on the galactose utilization (GAL) pathway of Saccharomyces cerevisiae. This pathway has been extensively studied over 30 years by a reductionist approach (one gene/protein at a time). On the basis of the results obtained by this approach, the relevance of galactose metabolism genes in this pathway was predicted. By a Systems Biology approach, which integrates data on protein and RNA levels as well as protein–protein and protein–DNA interaction into a single model, agreements and inconsistencies between RNA and protein levels were identified. On the basis of the obtained information, Ideker et al.19 were able to suggest new hypothesis on the regulation of GAL pathway, which was experimentally verified afterward.

This new interdisciplinary approach adopts a different perspective with respect to classical reductionistic one. Instead of trying and dissecting a given problem into its constituent atomisms hoping to individuate the “crucial step” of the studied process, the challenge is to take a frankly holistic approach in which the solution comes out from the clarification of the emergent properties coming out from the interaction pattern, in what has been dubbed Systems Biology. The idea emerging from Systems Biology, that the sum of the parts generates a collective quality than none of the constituent parts alone possess, reveals the limits of the reductionistic approach in understanding processes that happen in a biological system and their dynamical behavior in their entirety. Organisms are clearly much more than the sum of their parts, and the “behavior of complex physiological processes cannot be understood simply by knowing how the parts work in isolation”20. Genes determine the amino acid sequence of encoded proteins that in turn perform specific functions supporting cell physiology and organism development. On the other hand, gene products do not act in isolation but in a complex network whose interconnections are important to specify phenotypes. Biological functions can only rarely be attributed to a single molecule, and one phenotype may result from several different molecular and epigenetic mechanisms, while the same molecule may be involved in different phenotypes. Actually, even in the same organism, a protein may have different functions in different cells21; or signal pathway effectors may induce various differentiation programs in different cell lineages22. Moreover, in multicellular organisms, single cells do not have an existence independent from the whole organism, they are ontogenetically linked. This means that the usual way of thinking about organisms as made up of cells that relinquished their independence is inaccurate or misguided. A great number of biologists have been insisting that explanations of life should always be sought for gene or gene product level, regardless of the level of organization at which the phenomenon of interest is observed. In this view, genes are the only units of selection23 and development is just the unfolding of a genetic program; genes are the building units of the organism24. However, experimental evidence has challenged these pronouncements: molecular biology's dogma failed as it was evident that phenotype is not completely determined by genotype and that genetic markers are not so exclusively responsible for hereditary diseases. The statement underlining that “the human genome, like a good teacher raises at least as many questions as it answers”25 clearly illustrates the shortcomings of this reductionism. In relation to the postgenome sequencing era, reductionism has then been defined as “the attempt to explain complex phenomena by defining the functional properties of the individual components” of a system26. A good example to illustrate the need to integrate reductionistic approach with a systemic vision comes from cancer research. Defined a systems biology disease, understanding cancer complexity does not seem a matter of alternative theories but requires, first and foremost, a “systemic” methodological approach11. The point is that complexity seems to be not only a system's feature but, possibly, the most evident characteristic of any organism of life. That is why Systems Biology is not only a valuable instrument for this challenge but it is a challenge in itself from an intellectual point of view. The statement is that knowledge goes beyond information and its organization into a framework; it demands new concepts and vision of biological phenomena.


The capability of Systems Biology to analyze complex data from multiple experimental sources using interdisciplinary tools and to integrate them depends on the use of technological platforms and on the development of high-throughput data collection technique. The use of these new platforms was pivotal for gaining knowledge of the status of cellular components at any time, while complex programs helped to determine how and when molecules interact.

From this point of view, Systems Biology provides a new general paradigm, able to take into account different levels of biological complexity. High-throughput omics data sets are generally semiquantitative and specific to a particular experimental system. Networks can integrate multidimensional data in a framework able to give a self-consistent compendium of systemic data to foresee and preview reactions and behavior of complex biological systems. Systems Biology's synthesis allows the development of holistic approaches and provides a good tool for a mathematical interpretation and description of complex biological problems10. A fine example of how the coupling of empirical data with mathematical analysis permits the identification of previously unknown signaling mechanisms comes from a study on NF-kB signaling module, where the importance of IkB isoforms in feedback loops has been deciphered27.

From the data, new hypotheses can be formulated and examined by new experiments. Large sets of genome-wide data are collected, and associations between the different molecules are deduced. Systems Biology also aims to understand cell cycle regulation, or other processes, in organisms of different evolutionary complexity, as well as pathological versus physiological conditions by interactively integrating experimental and computational findings28. It is a matter of fact that the discipline has begun to piece together first level detailed maps describing how cells process various biochemical signals. In the future, it could also push more molecular-based disease treatments closer to reality.

Interdisciplinary tools and personnel also define the new strategy of pursuing integration of the amount of data obtained by various experimental approaches. Systems Biology is able to generate mathematical models to comprehensively describe and understand biological systems, through the functional and evolutionary analysis of medium- to large-sized networks. It has also been proposed that the network paradigm in which microscopic level elements are each other related by functional links is a promising metaphor to try and develop a statistical mechanics inspired approach for biological systems [5].

With the emergence of Systems Biology, it may be possible to grasp and understand some fundamental laws that rule biological systems. Soon or later the development of Systems Biology will have a deep impact in the rethinking of the basic principles of physics, pushing basic science to rethink many time honored simplifications (e.g., how the apparent violations of second law of thermodynamics by biological systems accommodates with a sound definition of functional and structural complexity of the studied systems?). In studying the basic properties observed in different biological systems, new concepts emerge that have been already treated in an analytical perspective (robustness, noise, degeneracy, bow tie architecture, etc.). Recently, receptor tyrosine kinases family and signaling networks they are involved in, has been analyzed to exemplify the systems perspective in the context of information relay networks and their relevance to human malignancies29. Once again, it is worth noting that plenty of these new concepts, like robustness, are intimately linked to biological complexity, and that they rule many aspects of functioning systems, sharing many characteristics, and intersecting each other, often becoming one of the individual and part of the other.

Systems Biology, finally, seems to be an open door to biology of systems when it overcomes reductionism's limits. Complex networks are probably incomprehensible without some mathematical structure and Systems Biology may be useful to clarify and to predict some emergent properties at molecular levels of biology systems30. However, the greatest challenge is not to get caught up in a new reductionism by reducing life to math but rather to bring about a change: from an entirely reductionist hypothesis-driven method to a high-throughput data acquisition, rigorous quantization, and mathematical modeling through a systemic perspective. We agree that the limitation of Systems Biology is greatly tied to data modeling but generalization and simplification is something intrinsically related with this methodology. What are Systems Biology's mathematical and informational limits? It is what it is: a mathematical and informational tool. Nevertheless, in our opinion, the idea beyond Systems Biology is rediscovering and highlighting an intellectual challenge: the need for integrating the physical aspects of biological systems with a holistic view of them. Taking into account the whole has always been a perspective of biology, Systems Biology reveals biology of systems, starting from and reaching to their holistic aspect.

The examples above given, from different areas of biological studies, underline how this topic and methodology is spreading out among different fields. Besides the increasing interest in empirical research, the educational interest on Systems Biology should also be supported by its epistemological implications. It is not solely a problem of how to integrate all the information available as much as to be able to manage the complexity; the systemic dimension, which is why its data always need interpretation31, will eventually help our students to develop critical and interpretative thoughts, generating new hypothesis and moving research forward. While it can be usefully complemented by reductionist approach, Systems Biology will be an instrument that can provide new important outputs, taking into consideration the fact that to transform information into understanding is a rational challenge not just a technical one.

This is particularly evident when considering the actual crisis in the development of new pharmacologically active compounds. In a recent paper33, the need for a general recasting of the way we look at drug mechanisms of action and the consequent change in the research and development strategies are clearly outlined. For some decades, the pharmacological research was dominated by the search for a molecular determinant of a given disease that could act as the attack point for the therapy. This strategy reached the postgenomic era under the heading of “druggable genome” making the implicit hypothesis that high throughput technologies could give rise to an unparalleled amount of possible drug targets consequently enlarging the spectrum of pharmacological intervention.

This kind of strategy did not fulfill its promise. Overington et al.32 performed a thorough statistical analysis of the marketed drugs, discovering that ∼130 target families (functional domains of a protein for which a relevant fraction of family members have been successfully targeted by a drug in the terminology of the authors) cover all current marketed drugs. This number is in striking contrast to the estimated number of protein families and folds (10,000 folds and more than 16,000 families). Moreover, the authors estimate that the 76% of the 361 new chemical entities approved by FDA between 1989 and 2000 targeted a previously drugged domain and only 6% targeted a previously undrugged domain, while the remainder have either unknown targets or are believed not to have distinct molecular targets underlying their action. These data suggest the repertoire of protein domains amenable to be targeted by drugs is only a very limited fraction of the entire genome. The authors comment these data pointing to the need of actively exploring the “polypharmacology” way, i.e., “dirty drugs” that allow for a relevant clinical effect by a combined effect on a wide spectrum of different targets. In Systems Biology terms, this implies the need of a completely different view of how a macroscopic level therapeutical effect can emerge from a microscopic stimulus (drug) that only in a minor part of the cases can be traced back to linear causation chains driven by “specific receptors.”

Csermely et al.33 suggest the future of pharmacology is in the exploitation of the network pharmacology paradigm. The authors suggest a shift of paradigm from the search for a very efficient and specific binding to a single molecular target to the development of “weak binders” to multiple edges of biological regulation networks. This is still a very theoretical and hypothetical proposal that in any case holds promises for a more realistic (and effective) approach to pharmacological research.

All in all we can state Systems Biology approach will not only change the style of what a “biological explanation” is but will have a deep impact in terms of therapeutical interventions.