Content area
Full Text
J. Math. Biol. (2009) 58:657687
DOI 10.1007/s00285-008-0212-0
Received: 28 June 2007 / Revised: 12 May 2008 / Published online: 8 October 2008 Springer-Verlag 2008
Abstract In this paper we compare two alternative theoretical approaches for simulating the growth of cell aggregates in vitro: individual cell (agent)-based models and continuum models. We show by a quantitative analysis of both a biophysical agent-based and a continuum mechanical model that for densely packed aggregates the expansion of the cell population is dominated by cell proliferation controlled by mechanical stress. The biophysical agent-based model introduced earlier (Drasdo and Hoehme in Phys Biol 2:133147, 2005) approximates each cell as an isotropic, homogeneous, elastic, spherical object parameterised by measurable biophysical and cell-biological quantities and has been shown by comparison to experimental ndings to explain the growth patterns of dense monolayers and multicellular spheroids. Both models exhibit the same growth kinetics, with initial exponential growth of the population size and aggregate diameter followed by linear growth of the diameter and power-law growth of the cell population size. Very sparse monolayers can be explained by a very small or absent cellcell adhesion and large random cell migration. In this case the expansion speed is not controlled by mechanical stress but by random cell migration and can be modelled by the FisherKolmogorovPetrovskiiPiskounov (FKPP) reactiondiffusion equation. The growth kinetics differs from that of densely packed aggregates in that the initial spread, as quantied by the radius of gyration, is
H. Byrne
School of Mathematical Sciences, University of Nottingham, Nottingham NG7 2RD, UKe-mail: [email protected]
D. Drasdo (B)
INRIA, Domaine Rocquencourt, B.P. 105, 78153 Le Chesnay Cedex, France e-mail: [email protected]
D. Drasdo
IZBI, University of Leipzig, Leipzig, Germany
Mathematical Biology
Individual-based and continuum models of growing cell populations: a comparison
Helen Byrne Dirk Drasdo
123
658 H. Byrne, D. Drasdo
diffusive. Since simulations of the lattice-free agent-based model in the case of very large random migration are too long to be practical, lattice-based cellular automaton (CA) models have to be used for a quantitative analysis of sparse monolayers. Analysis of these dense monolayers leads to the identication of a critical parameter of the CA model so that eventually a hierarchy of three model types (a detailed biophysical lattice-free model, a rule-based cellular automaton and a continuum approach)...