BindCraft: one-shot design of functional protein binders.

Area of Science:

• Biochemistry and Structural Biology
• Computational Biology and Bioinformatics
• Molecular and Chemical Engineering

Background:

• Protein-protein interactions (PPIs) are fundamental to biological processes but challenging to design due to complex structural determinants.
• Current methods for designing protein binders often require extensive experimental screening and optimization.

Purpose of the Study:

• To introduce BindCraft, an open-source, automated pipeline for *de novo* protein binder design.
• To demonstrate the efficacy of BindCraft in generating binders with high affinity and experimental success rates.

Main Methods:

• Leveraging AlphaFold2 weights for *de novo* protein binder generation.
• Designing binders against diverse and challenging targets, including cell-surface receptors, allergens, and CRISPR-Cas9.
• Utilizing computational design without high-throughput screening or experimental optimization.

Main Results:

• Achieved experimental success rates of 10-100% for designed binders.
• Generated binders with nanomolar affinity against various targets.
• Demonstrated functional applications, including allergen neutralization, gene editing modulation, and targeted gene delivery.

Conclusions:

• BindCraft offers a significant advancement in computational protein design, enabling a "one design-one binder" approach.
• The pipeline shows immense potential for applications in therapeutics, diagnostics, and biotechnology.
• Automated *de novo* design with high experimental validation accelerates the development of novel protein-based solutions.