Morning: a general introduction to intrinsically disordered proteins through a historical perspective to the most recent developments.

• brief history and basic concepts
• biophysical methods for the characterization of structural disorder
• functions and functional classification of disordered proteins
• structural disorder in diseases: can we target IDPs?
• ...

Afternoon: recent progress in database development

[Pietrek, Stelzl, & Hummer, J. Chem. Theory Comput. 2020]

Description of the practical session

1. Hierarchical Chain Growth
• Working with complete code of a functional HCG implementation for a simplified case
2. Bayesian Ensemble Refinement
   • Reweighting the HGC ensemble according to SAXS data via Bayesian Inference of Enesembles (BioEn)
3. Coarse-Grained Simulation
   • Homopolymer model calculations
   • MARTINI simulation of an IDP
4. Comparison of Experimental Observables between Models

• Statistical coil sampling of unfolded states
• Simulation of measurables and comparison with experimental values. Importance of averaging. This will include PREs, CSs, SAXS, RDCs.
• Simulation of NMR relaxation - importance of correct treatment of angular correlation functions/sampling of the conformational ensemble.
• Interactions between IDPs and partners - exchange methods and how these can be used to enhance our understanding of disorder in biology.
• Examples of the above, and biological insight that can be derived

The morning lecture and the afternoon practical will cover physics-based computational models that we have been developing in close collaboration with experimental collaborators for the last ten years.

The lecture will provide a broad overview of available atomistic and coarse-grained models and their relative benefits for specific research problems. A particular emphasis will be a discussion of models suitable for studying the liquid-liquid phase separation of proteins and nucleic acids.

The practical will cover the software and scripts necessary to use our coarse-grained modeling framework capable of providing molecular-level details on the sequence-determinants of protein phase separation due to specific disease mutations and the inclusion of a folded domain, and variation of chain length.

Part I:

• Modeling proteins in internal coordinates: a robotics-inspired approach
• Tripeptides as basic elements to model highly-flexible proteins and regions
• Structural propensity prediction from IDP sequences based on a structural database of tripeptides
• Generation of conformational ensemble models of IDPs/IDRs

Part II:

• Distance based structural measures: least RMSD and combined RMSD
• Nearest neighbor graphs of conformations and energy landscape models
• Persistence based analysis of energy landscapes
• Applications to IDPs

References:

1. Predicting secondary structure propensities in IDPs using simple statistics from three-residue fragments; A. Estaña, A. Barozet, A. Mouhand, M. Vaisset, C. Zanon, P. Fauret, N. Sibille, P. Bernadó, J. Cortés; Submitted.
2. Realistic ensemble models of intrinsically disordered proteins using a structure-encoding coil database; A. Estaña, N. Sibille, E. Delaforge, M. Vaisset, J. Cortés, P. Bernadó; Structure, 27(5): 381-391.e2, 2019
3. A reinforcement-learning-based approach to enhance exhaustive protein loop sampling; A. Barozet, K. Molloy, M. Vaisset, T. Siméon, J. Cortés; Bioinformatics, 36(4): 1099-1106, 2020
4. Characterizing molecular flexibility by combining lRMSD measures; F. Cazals, and R. Tetley; Proteins, 87 (5), 2019.
5. Energy landscapes and persistent minima; J. Carr, and D. Mazauric, and F. Cazals, and D. Wales; The Journal of Chemical Physics, 144 (5), 2016
6. Conformational Ensembles and Sampled Energy Landscapes: Analysis and Comparison; F. Cazals, and T. Dreyfus, and D. Mazauric, and A. Roth, and C.H. Robert; J. Comp. Chem., 36 (16), 2015.