Bayesian Methods in Structural Bioinformatics [Paperback]

This book is an edited volume, the goal of which is to provide an overview of the current state-of-the-art in statistical methods applied to problems in structural bioinformatics (and in particular protein structure prediction, simulation, experimental structure determination and analysis). It focuses on statistical methods that have a clear interpretation in the framework of statistical physics, rather than ad hoc, black box methods based on neural networks or support vector machines. In addition, the emphasis is on methods that deal with biomolecular structure in atomic detail. The book is highly accessible, and only assumes background knowledge on protein structure, with a minimum of mathematical knowledge. Therefore, the book includes introductory chapters that contain a solid introduction to key topics such as Bayesian statistics and concepts in machine learning and statistical physics.

This edited volume is a high-profile overview of the current state of play in statistical methods applied to structural bioinformatics. With almost 100 pages of introductory material, it covers topics including protein structure prediction and simulation.

Part I Foundations: An Overview of Bayesian Inference and Graphical Models.- Monte Carlo Methods for Inferences in High-dimensional Systems.- Part II Energy Functions for Protein Structure Prediction: On the Physical Relevance and Statistical Interpretation of Knowledge based Potentials.- Statistical Machine Learning of Protein Energetics from Experimentally Observed Structures.- A Statistical View on the Reference Ratio Method.- Part III Directional Statistics and Shape Theory: Statistical Modelling and Simulation Using the Fisher-Bingham Distribution.- Statistics of Bivariate von Mises Distributions.- Bayesian Hierarchical Alignment Methods.- Likelihood and Empirical Bayes Superpositions of Multiple Macromolecular Structures.lÛ