Biological Systems

Biological systems are highly complex, and understanding how they respond and adapt to a dynamic environment is particularly challenging. Agent-based simulation can provide a convenient medium for domain experts to express and communicate complex ideas during hypothesis formulation. The following paper provides a high-level introduction to Bioscience Computing:

  • BioScience Computing and the role of computational simulation in biology and medicine (Clack, in Intelligent Algorithms in Ambient and Biomedical Computing, eds. W.Verhaegh, E. Aarts and J.Horst, Philips Research Book Series, Vol. 7, pp 3-19, ISBN 1-4020-4953-6, Springer 2006).

Multi-agent simulation (MAS) techniques support exploration of non-evolutionary adaptations within a single lifetime. The following papers explore the computational processes that underlie specific biological mechanisms and for example demonstrate how a reasonably simple MAS can exhibit lifetime adaptation with multiple goal-oriented behaviours as emergent behaviour:

  • The Artificial Cytoskeleton for Lifetime Adaptation of Morphology (Bentley and Clack, in Proceedings SODANS workshop proceedings of the Ninth International Conference on the Simulation and Synthesis of Living Systems (ALIFE IX), pp 13-16, 2004).
  • Morphological Plasticity: Environmentally Driven Morphogenesis (Bentley and Clack, LNCS 3630:118-127, ISSN 0302-9743, Springer 2005)
  • Multi-level behaviours in agent-based simulation: colonic crypt cell populations (Chen, Nagl and Clack, in Proceedings Seventh International conference on Complex Systems, paper 22, New England Complex Systems Institute (NECSI) and Interjournal, 2008)

Clack collaborated with Eileen Cox (Natural History Museum) to supervise Katie Bentley's work in developing a novel agent-based technique to simulate morphological dynamics in living tissue; the following paper addresses diatom morphology and the technique has since been applied to cancer tumour morphology. This work has led to the founding of a morphogenesis modelling laboratory at Harvard. Unlike previous models, this simulation recreates detailed mechanisms and chemical interactions and the following paper (JCR impact factor 2.071) provides unexpected evidence that a single mechanism produces two different types of morphogenesis. Rib-domination phenomena are faithfully reproduced and the new model, experimental methodology, and results, were assessed against state of the art techniques and SEM data by the Natural History Museum.

  • Diatom colony formation: A computational study predicts a single mechanism can produce both linkage and separation valves due to an environmental switch (Bentley, Clack and Cox, Journal of Phycology 48(3):716-728, Phycological Society of America, 2012)

Christopher D. Clack
Department of Computer Science
Gower Street