Optimized distance-dependent atom-pair-based potential DOOP for protein structure prediction.

Area of Science:

• Computational Biology
• Structural Biology
• Bioinformatics

Background:

• Protein structure prediction is crucial for understanding biological function.
• Existing energy functions often struggle with accurately ranking native structures.
• Empirical potentials based on atom-pair interactions are a promising approach.

Purpose of the Study:

• To develop and optimize a novel energy function, the DOcking decoy-based Optimized Potential (DOOP), for protein structure prediction.
• To improve the accuracy of native protein structure recognition.
• To model the hierarchical funnel-like energy landscape of protein interactions.

Main Methods:

• Decomposition of native protein structures into polypeptide chain segments forming hypothetical ligand-receptor systems.
• Generation of 8609 ligand-receptor systems with 1000 docking decoys each (0-10 Å iRMSD).
• Application of a neural network-based optimization method to derive energy parameters mimicking a funnel-like energy landscape.

Main Results:

• The DOOP energy function, combined with a torsion potential term, demonstrated superior performance in ranking native protein structures across various decoy sets.
• Outperformed other statistical potentials, particularly on the challenging ROSETTA decoy set.
• Effectively models the energy landscape of protein interactions.

Conclusions:

• The DOOP energy function represents a significant advancement in protein structure prediction accuracy.
• The method provides a robust approach for optimizing statistical potentials.
• The DOOP energy function and associated data are publicly available for further research.