Formal thermodynamics of combinatorial molecular networks

Statistical Physics and Complexity Group meeting

Formal thermodynamics of combinatorial molecular networks

  • Event time: 2:00pm
  • Event date: 29th January 2010
  • Speaker: Vincent Danos (University of Edinburgh)
  • Location: Room 2511,

Event details

Modern molecular biology uncovers truely amazing combinational structures in biological networks (eg multi-component ubiquitination systems, multi-unit phosphatase networks). One burning issue is which tools and concepts are needed to capture this mixture of algebraic combination and physical constraints, and computationally analyse biological systems of combinatorial complexity. Rule- and agent-based approaches for the circumnavigation and transparent representation of combinatorial molecular interactions. These methods have shown particularly useful, as in the modeling of combinatorial promoters, the assembly of macro-molecular machines, and the fine-grained modelling of DNA-related processes with explicit 1D diffusion mechanisms such as epi- genetic maintenance.

Here we wish to take a step further and introduce a class of rule-based energy functionals, that will guarantee the thermodynamical correctness of models by construction.

This is related to the energy-based description of macro-molecular assemblies in Ref. 'stochastic dynamics of macromolecular-assembly networks', Mol Syst Biol - 2006.

Specifically, we talk about energy functions to guarantee the existence of a detailed balance in a rule-based (Kappa) continuous time Markov Chain; we consider a class of local energies for which we derive a criterion for such an energy to (prove the existence and) determine the equilibrium of a given rule set; we show that Julien Ollivier's ANC -a general allosteric language for protein networks- is a particular case; we discuss how to add non local energy terms, and why (they are connected to positional entropy); we discuss the reverse problem of finding an energy given rules with rates (undecidable in general).