ClustENM: ENM-Based Sampling of Essential Conformational Space at Full Atomic Resolution

Area of Science:

• Computational Biology
• Biophysics
• Structural Biology

Background:

• Sampling conformational space in large biomolecular systems is challenging for molecular dynamics (MD) simulations.
• Accurately capturing transitions between functional substates requires efficient computational methods.

Purpose of the Study:

• To develop an unbiased computational method for sampling large conformational changes in biomolecules of various sizes and oligomerization states.
• To introduce ClustENM, an Elastic Network Model (ENM)-based approach for enhanced conformational sampling.

Main Methods:

• Developed ClustENM, an iterative method combining ENM with energy minimization and clustering.
• Applied ClustENM to six diverse biomolecular systems, including adenylate kinase, calmodulin, and the 70S ribosomal complex.
• Validated generated conformer ensembles against experimental data (X-ray, NMR) and independent MD simulations.

Main Results:

• ClustENM successfully sampled large conformational changes for various biomolecular systems.
• Generated ensembles showed good agreement with 979 experimental structures.
• The method encompassed conformational subspaces explored by independent MD simulations for specific systems.

Conclusions:

• ClustENM is a computationally efficient tool for characterizing biomolecular conformational space at atomic detail.
• The method generates representative conformer ensembles beneficial for simulating substrate/ligand-binding events.
• ClustENM provides an unbiased approach requiring only an initial structure.