Perturbing the energy landscape for improved packing during computational protein design

Area of Science:

• Computational Biology
• Protein Engineering
• Biochemistry

Background:

• The FastDesign protocol in Rosetta aids in de novo protein design by iterating sequence optimization and structure refinement.
• This method has been instrumental in creating novel protein structures and complexes for research and medicine.
• Current FastDesign incorporates energy landscape smoothing by reducing repulsive forces to enhance conformational and sequence sampling.

Purpose of the Study:

• To address the observed bias against larger amino acids in the protein core during FastDesign refinement.
• To develop an improved ramping strategy for repulsive forces to enhance protein design outcomes.
• To validate the optimized protocol through experimental characterization of newly designed proteins.

Main Methods:

• Investigated the effect of protein compression during refinement on amino acid selection in the core.
• Tested alternative strategies for adjusting repulsive forces during the energy landscape smoothing stages.
• Designed and experimentally characterized over 4000 proteins using the modified FastDesign protocol.

Main Results:

• Discovered that early-stage protein compression in FastDesign inadvertently disfavors larger amino acids in the core.
• Identified an optimized repulsive weight ramping strategy that yields lower energy designs.
• The improved protocol resulted in designs with more native-like sequence composition in the protein core.
• Experimentally validated proteins designed with the new protocol demonstrated enhanced stability.

Conclusions:

• The refined FastDesign protocol overcomes limitations in core amino acid selection.
• The optimized protocol generates more stable and energetically favorable de novo protein designs.
• This advancement has significant implications for protein engineering and therapeutic development.