UvA

Non-equilibrium steady states are of great interest for the study of photosynthesis, molecular motors, biological switches or driven systems in statistical physics. The aim of this thesis is to broaden the understanding of non-equilibrium steady states (NESS) in soft-matter systems and provide a framework for reweighting dynamical information between NESS. The method is applied to phenomenological single particle systems described in full configurational coordinates and a tetraalanine peptide described in system-specific collective variables. We avoid the combinatorial explosion of microtrajectories by systematically constructing pathways through Markovian transitions. The reweighting is based on the information theoretic approach of Jaynes' Maximum Caliber, connecting data drawn from simulation or experiment to system information in the form of constraints. It is shown that local entropy production constraints define a NESS by controlling dissipation of a system on a local level. Non-dissipative contributions to the dynamics are drawn from the reference data and are shown to define an invariant quantity under reweighting. The presented reweighting method demonstrates the potential of the Maximum Caliber to understand and analyse systems off-equilibrium.