Proteins of well-defined structures can be designed without backbone readjustment by a statistical model

Area of Science:

• Computational biology
• Protein engineering
• Biophysics

Background:

• Protein sequence design is crucial for creating novel proteins with specific functions.
• Accurate prediction of protein structure from sequence remains a significant challenge.
• Existing methods often require computationally intensive backbone relaxation or experimental evolution.

Purpose of the Study:

• To develop and validate a revised statistical energy model for high-confidence protein sequence design.
• To demonstrate the model's capability in designing de novo proteins for specific backbones.
• To assess the necessity of considering backbone relaxation in protein design.

Main Methods:

• Rational revision of a statistical energy function for backbone-based protein sequence design.
• Application of the revised model to design a de novo protein for a pre-defined target backbone.
• Experimental verification of the designed protein's structure and foldability.

Main Results:

• The revised statistical energy model significantly improved accuracy in protein sequence design.
• A de novo protein was successfully designed, exhibiting excellent agreement between its actual and target backbone structures.
• The design protocol proved effective across different backbone types, yielding well-defined protein structures.

Conclusions:

• High-confidence protein sequence design is achievable without explicit consideration of backbone relaxation using an appropriately resolved energy model.
• The developed protocol offers a more efficient approach to de novo protein design.
• This work advances the ability to design proteins with precise structures computationally.