Project Details
Projekt Print View

A mechano-geometric framework to characterize macromolecular ensembles

Subject Area Mechanics
Term from 2019 to 2023
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 401512690
 
Final Report Year 2023

Final Report Abstract

Macromolecules such as proteins, RNA, and DNA dynamically change their three-dimensional structure to perform their cellular functions. Understanding a molecule’s structural ensemble is crucial for revealing its key roles. While Molecular Dynamics (MD) simulations provide atomically trajectories, their computational cost is often prohibitive. In contrast, kinematic methods inspired by robotics can efficiently and fast model molecular motion, providing useful tools for data interpretation and integration. The objective in this project was to extend our existing kinematic molecular modeling method into a unified, mechano-geometric framework to study conformational ensembles of complex macromolecules. We combined geometric constraints and rigidity theory to study molecular flexibility and imposed a hierarchy of motions, enabling efficient conformation space and energy landscape sampling in an ultra-high-dimensional environment. We also identified allosteric hotspots for drug targeting through changes in dihedral degrees of freedom, steric or hydrophobic contacts, and constraint violations. A further objective was the validation and direct guidance with experimental data to bridge the gap between experimental and computational analysis. First, our method was extended for multi-chain proteins and molecular complexes. It is applied to identify key residues involved in allosteric communication, predicting mutation effects, and finding potential inhibitors. We found that the activation loop rigidified the active state kinases conformation, which is valuable for developing potential inhibitors. Furthermore, our method yields information about rigidified dihedral angles which affect molecular vibrational modes. We applied our method to the SARS CoV-2 main protease (Mpro ) and its mutation to analyze potential drugs. By analyzing 47 mutation sites for more than 3,300 different structures, we found that mutations generally increased the flexibility of Mpro. We also optimized the conformational transition motion planning for ultra-highdimensional conformation. We proposed dynamic Clash Constraints and a randomized Poisson-disk motion planner to address the challenge. Our results agreed with those from MD simulations. Second, for the data integration, we use a combination of techniques to study how covalent catalysis modulates isocyanide hydratase (ICH) conformational dynamics. By using several methods such as X-ray free electron laser (XFEL), we demonstrate that Gly150 mutations that modify helical mobility decrease ICH catalytic. Additionally, we use site-specific cysteine photo-oxidation to investigate changes in protein structure. We analyzed the role of non-native contacts and identified energy barriers, helping us understand ligand. Our method bridges rigidity theory and elastic network models (ENM), revealing a protein-fold-specific, spatial hierarchy of motions encoded by the hydrogen bonding network and rationalizes experimental and simulated protein stiffness variations.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung