During the last decades, molecular biology has identified and characterized most of the molecular parts of the cell and sequenced the human genome. However, it turns out that knowing the molecular components of the cell is not enough to answer many remaining questions about human development and disease. This is because it does not account for the dynamic interactions between the components that regulates cellular physiology and homeostasis.
These interactions and their dynamics are studied in systems biology. Systems biology views the cell as a network of interacting biochemical components (metabolites, proteins, genes, etc.) and studies the topology of these networks as well as the dynamics they give rise to. It translates classical box-and-arrow pathway descriptions to dynamical network models including positive and negative feedbacks.
Because such nonlinear networks go far beyond the capabilities of our intuition, it uses mathematical concepts and computational methods from dynamical systems theory to formally describe the changes in e.g. metabolic fluxes, protein concentrations and gene expression happening inside the cell. Elucidating the dynamics of cellular regulation can help to identify control points that may be used as targets for drug development.
Multicellular systems biology
Systems biology, however, mainly focuses on the dynamics inside the cell, although many regulatory circuits go beyond the cellular level and involve crosstalk between cells, the extracellular matrix and tissue biomechanics. Moreover, key events in embryonic development as well as disease progression manifest themselves at the tissue level and cannot be studied purely at the intracellular level. It is therefore important to address the multicellular context in which cells exist and perform their function.
Multicellular systems biology is the emerging field that places the intracellular networks back in their natural multicellular context. It extends systems biology by including the spatial interactions between cells as well as the biomechanics within tissues. This enables the exploration of morphogenesis, pattern formation and e.g. tumor progression that cannot addressed by current systems biological approaches.
To describe the physical interactions of cells, different mathematical concepts and models are required. Cells are typically represented as discrete interacting entities, described in various formalisms such as cellular automata or agents-based models. The description of their mechanical properties is often derived from physical models for materials such as foam.
Mathematical and computational models in multicellular systems biology describe interactions within a cell as well as interactions between cells. By definition, these descriptions are multi-scale models as they represent both the molecular, cellular and tissue scales. As such, they require the integration of disparate mathematical formalisms used to describe these different levels, from ordinary and partial differential equations to a variety of agent-based model formalisms.
Models in multicellular systems biology can be used to test hypotheses about the effect of molecular perturbations (e.g. mutations) on tissue-level processes (e.g. morphogenesis). Or, reversely, the effect of tissue disruption (e.g. enzymatic dissociation) on intracellular dynamics and cell fates (e.g. dedifferentiation). Such models have already found a wide range of applications, mainly in developmental biology (e.g. see vascular morphogenesis) and cancer biology, and their use of emerging in stem cell biology and regenerative biology (e.g. see cell fates and patterning) as well as tissue engineering.
Due to the complexity of multiscale models in multicellular systems biology, simulation is indispensable. However, constructing multiscale computational simulations is a complicated and error-prone task that requires considerable software engineering expertise. Software tools are therefore needed to support the development the models in a reliable and reusable fashion.
Existing software tools for systems biology are not equipped for the simulation of spatial, cell-based, multiscale models. Therefore, new software tools and computational platforms are required, dedicated to assisting researchers in the development of multiscale multicellular models.
Recently, several platforms for multiscale modeling of multicellular systems have been developed and made available. Projects such as Chaste, CompuCell3D, EPISIM, VirtualLeaf and Morpheus offer a diverse range of reliable and reusable implementations in a variety of forms of software, from libraries and frameworks to off-the-shelf applications.
With Jörn Starruß, I have developed Morpheus, a user-friendly modeling and simulation environment of multicellular systems biology. Morpheus separates modeling from implementation and automates the error-prone process of model integration. It features a domain-specific modeling language and intuitive graphical user interface and does not require any programming. It enables everyone with an interest in mathematical biology to create and simulate complex computational models of multicellular systems.
Bioinformatics, 30 (9), pp. 1331–1332, 2014.
To foster and strengthen the visibility and development of this type of scientific software, I have developed a number of initiatives with the Tissue Imaging and Analysis Center (TIGA) at BioQuant in Heidelberg. Recently, we have organized an ECMTB symposium which featured invited talks by developers of all the major software platforms. This type of meetings is necessary to identify common challenges and approaches and outline the path towards SBML-like standardization for multicellular systems biology. Similar meetings are planned to discuss the current opportunities for multicellular modeling targeted at a biomedical audience.