Statistical models of protein structure
Start with the big picture
The lesson surveys statistical approaches to protein structure and folding, beginning with the Ramachandran plot as a map of allowed backbone φ and ψ angles. It then considers helix–coil models, random-coil and lattice representations, Boltzmann weighting, and database-derived statistical potentials. Coarse-grained models and free-energy landscapes offer simplified views of global structure and folding states, while Markov state models describe transitions among conformational microstates. The topic also introduces principal component analysis, Monte Carlo sampling, and molecular dynamics as tools for exploring motions and conformational space. These approaches differ in what they represent and how they sample; their value lies in matching the model to the structural or dynamical question being asked.
What you'll learn
- Explain how the Ramachandran plot relates backbone angles to structural plausibility.
- Distinguish helix–coil, random-coil, and lattice models by their assumptions and uses.
- Describe how Boltzmann weighting and statistical potentials represent conformational preferences.
- Summarize how free-energy landscapes and Markov state models represent folding states and transitions.
- Identify the roles of coarse-graining, principal component analysis, and sampling methods in studying protein conformations.
Continue your study
Work through the complete notes and reinforce the topic with the study tools available in the full lesson.
