AlphaDesign: a de novo protein design framework based on AlphaFold

Area of Science:

• Synthetic biology
• Computational biology
• Protein engineering

Background:

• De novo protein design is crucial for synthetic biology.
• Numerous computational methods exist, but generality and usability remain challenges.

Purpose of the Study:

• To introduce AlphaDesign, a versatile computational framework for de novo protein design.
• To enable rapid generation and validation of proteins with specific interactions, conformations, and oligomeric states.

Main Methods:

• AlphaDesign combines AlphaFold with autoregressive diffusion models.
• The framework allows for controllable protein generation without class-dependent retraining.
• In vivo validation of designed proteins against bacterial retrons was performed.

Main Results:

• AlphaDesign successfully generated and validated novel proteins.
• Designed proteins function as active inhibitors of bacterial retron systems.
• The framework demonstrated generality and usability in protein design.

Conclusions:

• AlphaDesign offers an efficient and rational approach for de novo protein design.
• The developed framework paves the way for designing novel proteins as biologics.
• This work advances the field of protein engineering and synthetic biology.