The emerging field of systems biology provides a conceptual framework on which to build an integrated computational model of the complex genetic network that mediates an individual's state of health. The tools to construct such an integrated biomedical systems model are being provided by new high-throughput "omics"-based methods followed by application of computational algorithms for forward and reverse engineering to the systems parameters captured by these molecular profiling technologies. The goal of this biomedical systems modeling is to provide a practical guide for predictive, preventive, and personalized medicine. It has been suggested that this systems approach represents a paradigm shift and that it will take 10-20 years of development and transition to achieve the ultimate goal of personalized medicine . At the heart of this conceptual change is the understanding of normal and patho logical processes as distinct states of genetics-based hierarchical networks . One important insight that has already been realized is that biological networks from different levels of organization (e.g., metabolic, protein interactions, regulatory, cell interactions, tissue and organ interactions, and even populations of individuals) share the same global architecture [78, 79]. The small-world property of biological networks, the high degree of connectivity between nodes, has profound consequences for our understanding of drug targets and of intervention strategies to correct disease states. From this systems perspective, combinatorial therapeutic strategies, designed to repro-gram cellular responses by changing the local architecture of the genetic network, are likely to be more effective than traditional single target-based drug interventions. Conversely, the effects of a given drug will be dependent on its interactions with the specific genetic network state of a given individual, that is, pharmacogenetic interactions will mediate the efficacy of drug treatment in personalized medicine.
Genetic network states are dynamic, both in response to environmental stimuli and as a result of stochastic noise in the genetic circuitry [80, 81]. Thus comprehensive time course data on the changes in gene expression profiles for healthy versus disease states will need to be collected to support predictive, dynamic network models of disease progression and for prognosis for therapeutic interventions. For example, numerous attempts are ongoing to create such databases of gene, regulatory, and biochemical networks for cellular signaling processes, for example, the Alliance for Cellular Signaling  and the Signal Transduction Knowledge Environment (http://stke.sciencemag. org/index.dtl).
Integrative genomics therefore has the potential for ultimately mapping causal associations between gene expression profiles and disease states of the underlying cellular networks . The etiologies of complex adult diseases, such as cancer, diabetes, asthma, and neurodegenerative conditions are determined by multiplex gene-environment interactions. Thus the promise of toxi-cogenomic modeling of risk factors for environmentally responsive disease is of particular importance and has resulted in the establishment of the Environmental Genome Project by the National Institute of Environmental Health Sciences, to provide a focus for development of a systems toxicology perspective for personalized medicine . This paradigm shift for the field of drug safety evaluation and predictive risk assessment is reflected in the significant efforts to fund research that integrates toxicology, systems biology, and genomics, including initiatives that extend from the database level up to the exchange of information between different fields .
One could argue that the increase in publications on systems biology is a result of the increased funding and conceptual interest in this rather than the success of groundbreaking new discoveries . However, initial studies in synthetic biology  and in pathway engineering of mammalian genetic networks , albeit only in the discovery stage of development at present, promise to provide practical tools for reprogramming of the cellular networks that underlie health and disease avoidance. If tailored to the specific phar-macogenetic state of an individual , such a set of network intervention tools would define a new class of therapeutic drug and would enable a practical implementation of personalized medicine.
Current progress in systems biology suggests that predictive, network-based analytical approaches will continue to be developed at the interface of chemoinformatics and bioinformatics, generating a broad spectrum of applications ranging from drug target selection through clinical data analysis. Given the emphasis that we are currently seeing on systems approaches within academia and the pharmaceutical industry, systems-based technologies are likely to have an increasingly important role in enabling future advances in drug discovery and development.
Was this article helpful?