Long-term dynamic simulation of cellular systems with inhomogeneous mass distribution.

Abstract

This paper presents a high-speed, long-term approach to simulating the mechanobiology of cells with an inhomogeneous mass distribution ranging from femtograms to picograms. An accurate representation of cellular processes necessitates the inclusion of subcellular structures characterized by minute masses and dimensions. The minute objects yield multiscale dynamic models with disproportionate terms, which require inordinate amounts of computational time to simulate. The computational requirements limit the time span of the simulation to time histories shorter than one second, even when employing supercomputers. This paper examines adipogenesis, the transformation of human bone marrow-derived mesenchymal stem cells (hMSC) into adipocytes, a process that spans two weeks. The proposed simulation techniques are based on a novel scaling technique that addresses differently sized masses. This work addresses unique challenges beyond the authors' earlier work, which addressed disproportionality between large stiffness and damping forces in relation to small masses in the dynamic model. This new approach reduces computational time to less than 1 hour and 45 minutes on a standard desktop computer for the two-week duration of adipogenesis.