OPUS-DOSP: A Distance- and Orientation-Dependent All-Atom Potential Derived from Side-Chain Packing

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein structure modeling is crucial for understanding biological function.
• Existing statistical potentials have limitations in accurately recognizing native protein structures.
• Accurate protein structure prediction remains a significant challenge in molecular biology.

Purpose of the Study:

• To introduce OPUS-DOSP, a novel all-atom statistical potential for protein structure modeling.
• To enhance the accuracy of protein structure prediction through improved decoy recognition.
• To refine existing protein modeling frameworks with distance- and orientation-dependent features.

Main Methods:

• Developed OPUS-DOSP based on the OPUS-PSP framework.
• Incorporated distance- and orientation-dependent contributions based on contact distance ranges.
• Introduced a new auxiliary function to the energy term for adjusting extreme cases.
• Utilized 11 standard decoy sets for benchmarking the potential's performance.

Main Results:

• OPUS-DOSP demonstrated improved decoy recognition capabilities compared to existing potentials.
• Native structures were recognized in 239 cases without the auxiliary function and 249 cases with it.
• The new potential successfully identified a higher percentage of native structures within tested decoy sets.
• The inclusion of the auxiliary function further boosted the recognition accuracy.

Conclusions:

• OPUS-DOSP represents a significant advancement in all-atom statistical potentials for protein modeling.
• The distance- and orientation-dependent features, along with the auxiliary function, enhance decoy recognition.
• OPUS-DOSP shows superior performance in identifying native protein structures, offering a valuable tool for researchers.