Bayesian Modeling and Analysis in Modern Biophysics
Modern experimental techniques provide measurements with high spatial or temporal localization. The acquired datasets monitor the evolution of the underlying systems with resolution that may reach the molecular level and commonly large datasets probing single proteins or other biomolecules are available. Excessive noise, caused by the measuring hardware, experimental procedures or unaccounted processes, necessitates the formulation of statistical methods for the analysis and interpretation of these datasets. Nevertheless, physical limitations and the inherent uncertainties in the underlying systems, such as unknown components, states, dynamics, etc, common in applications throughout Biology pose unique conceptual and computational challenges that most often require non trivial model selection. In this talk, I will present an overview of the modeling challenges encountered and highlight recent advances including Bayesian non-parametric approaches.