RESEARCH AND DEVELOPMENT
ERCIM News No.48, January 2002 [contents]

The Silicon Cell

by Joke Blom


The Dutch Silicon Cell Initiative aims at computing Life at the cellular level, on the basis of our rapidly expanding knowledge of a cell’s genetics, biochemistry, and molecular dynamics. Four Amsterdam-based research institutes have formed a Consortium (SiC) to realise this aim, which is expected to take decades. CWI’s involvement concerns the analytic and numerical study of models, and visualisation.

The Silicon Cell Consortium consists of the Swammerdam Institute for Life Sciences (SILS) and the Institute for Informatics, both of the University of Amsterdam, the Institute for Molecular Biological Sciences (IMBS, coordinator Westerhoff) of the Vrije Universiteit in Amsterdam, and CWI. A collaboration with the Humboldt University (GraduiertenKolleg) in Berlin started recently.

The long-term goal of the Consortium is ‘the computation of Life at the cellular level on the basis of the complete genomic, transcriptomic, proteomic, metabolomic, and cell-physiomic information that will become available in forthcoming years’. SiC combines theory and experiment: on the basis of real experimental data and mathematical models, the implications for cell functioning will be computed. In this way we may hope to achieve a fundamental and quantitative understanding of the cell. If the cell can be approached in a rational and integrated way, it can be utilised as a ‘factory’ to produce chemicals or pharmaceutical components. In biomedical research, once the molecular basis of diseases is known, a rational treatment can be developed which can even be personalised based on the patient’s gene information. Of course, these are dreams of the future. A well-known metaphor is that the genome is a parts list for Life, but that no one knows how to build the engine.

SiC’s current, more modest ambitions are: computational modelling of specific modules of cell metabolism, and development of specific tools and approaches for modelling living cell behaviour. Model cells are E-coli (prokaryote) and S. Cerevisiae (baker’s yeast, eukaryote), since for these two cells, extensive experimental knowledge and facilities exist at SILS and IMBS.

Some of the mathematical and computational requirements concern the application of already existing techniques, including the analysis and numerical solution of partial differential equations, parameter estimation in non-linear dynamical systems, and graph theory. New challenges concern the interaction between phenomena involving a wide range of scales (both in space and time) and organisational complexity, and the incorporation of uncertainty into the models. An important aim of SiC is the discovery of principles to reduce the size of the problem without losing essential information.

Current research at CWI concerns:

  • Numerical and analytic studies of the influence of spatial variations in concentrations of biochemical species. Here a new type of control coefficient summation theorem was found
  • Mesoscopic description of biological membranes by PDE’s with a computationally expensive non-local component (also numerical and analytic studies)
  • Reduction of models for very complex chemical networks in bacteria, involving about thousand different proteins. Study of the existence and stability of equilibria. Mathematical modelling of enzyme influence on cell processes
  • Interactive visualisation of protein dynamics by essential dynamics methods (cf. articles by Robert van Liere, ERCIM News 44, January 2001, p.35, and p.39)
  • Visualisation of time-dependent confocal microscopy data, eg, (de-)condensation of chromosomes during cell division with the interactive VR system proteus.

Future research:

  • Modelling of spatial structures: comparison between the microscopic (particle-based), mesoscopic, and macroscopic (continuum PDE-based) approach
  • Numerical solution of stiff biochemical reaction systems (many different time-scales) with ‘noisy’ kinetic parameters
  • Study of cellular processes with time-critical characteristics, eg, in metabolic or in cell-cycle oscillations. Issues include feedback and regulatory mechanisms
  • Mathematical modelling of ternary protein structure using dynamic graph theory.

Link:
The CWI Silicon Cell website: http://www.cwi.nl/~gollum/SiC/

Please contact:
Joke Blom, CWI
Tel: +31 20 592 4263
E-mail: Joke.Blom@cwi.nl