Modeling the human cardiome in silico.

The past two decades have seen impressive success for reductionist biologic science, generating new experimental data at an unprecedented rate. The list of organisms with completely sequenced genomes is growing each month. In June, the completion of a draft sequence of the entire human genome was announced amid much fanfare. Molecular mechanisms for fundamental processes such as ligand-receptor interactions and signal transduction are being elucidated in exquisite structural detail. However, as attention turns to the next phases, such as cataloging protein structures (proteomics), it is clear to biologists that the challenge is much greater than assigning functions to individual genes. Most cell functions require the coordinated interaction of numerous gene products. Metabolic or signaling pathways, for example, can be considered the expression of a “genetic circuit,” a wiring diagram for cellular function.1 And the layers of complexity do not end at the plasma membrane. Tissue and organ functions require the interactions of large ensembles of cells in functional units and networks.2 No amount of biochemical or cellular detail is sufficient to completely describe a ventricular arrhythmia, for example. Ventricular anatomic properties are equally important. To identify the comprehensive approach that will be needed to reintegrate molecular and genetic data into a quantitative understanding of physiology and pathophysiology in the whole organism, Bassingthwaighte3 coined the term “physiome.” Literally, the term reflects an integrative approach to physiology, meaning life (“physio”) as a whole (“ome”). The Physiome Project (http://www.physiome.org/) was conceived as a multicenter, international collaboration to coordinate integrative studies of quantitative physiologic function. The project was endorsed in 1997 by the International Union of Physiological Sciences at a conference in St. Petersburg, Russia, called “On Designing the Physiome Project.” In addition, the Physiome Project is recognized by organizations such as the American Institute of Medical and Biological Engineering. In April 1998, the first Microcirculation Physiome Project Working Group Meeting was held as a satellite to the “Experimental Biology ‘98” meeting, and more than 100 scientists attended. In September 1999, a similar number of participants attended the “Physiome Symposium: Integrated Biology of the Heart” in Seattle, Washington. The Physiome Project has two major goals: the cataloging of physiologic information into databases for online query and retrieval and the systematic organization of those data into quantitative models of integrated systems.3 The long-range goal is to improve the understanding of human physiology and pathophysiology by integrating quantitative information both vertically across scales of biologic organization from genome to organism and horizontally across interacting physiologic functions and processes. Projects such as the Human Genome Project and its spin-offs have generated hundreds of databases of molecular sequence and structure information such as GenBank (http://www.ncbi.nlm.nih.gov/Genbank/) and the Protein Data Bank (http://www.rcsb.org/pdb/). These databases in turn have generated demand for online tools for data-mining, homology searching, sequence alignment, and numerous other analyses. One of the best entry points for those interested in the burgeoning field of bioinformatics is the National Center for Biotechnology Information Web site (http://www.ncbi.nlm.nih.gov/). In contrast, a major obstacle to the progress of the Physiome Project is the lack of databases for the morphology and physiologic function of cells, tissues, and organs. Although there are, for example, some excellent databases of metabolic pathways such as the Metabolic Pathways Database (MPW) (http://wit.mcs.anl.gov/MPW/) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) (http://www.genome.ad.jp/kegg/), there are not yet comprehensive public databases of myocyte ion channel kinetics or coronary vascular structure. This is one reason that the Physiome Project has focused on developing integrated theoretical and computational models. Models, even incomplete ones, can provide a formal framework for classifying and organizing data derived from experimental biology, particularly EDITORIAL