Tuning Materials-Binding Peptide Sequences toward Gold- and Silver-Binding Selectivity with Bayesian Optimization.

Area of Science:

• Biomaterials science
• Nanotechnology
• Computational chemistry

Background:

• Materials-selective peptides are crucial for nanomaterial applications but are difficult to discover.
• Existing machine learning approaches are limited by small datasets of known selective peptides.
• Expanding this knowledge base is experimentally challenging and costly.

Purpose of the Study:

• To develop a novel approach for discovering materials-selective peptide sequences.
• To significantly expand the library of gold (Au)-selective peptides.
• To identify new silver (Ag)-selective peptides and their binding motifs.

Main Methods:

• Integration of experimental techniques with computational modeling.
• Introduction of the Bayesian Effective Search for Optimal Sequences (BESOS) algorithm.
• Systematic screening and analysis of peptide sequences for selective binding.

Main Results:

• A significant expansion of the dataset for Au-selective peptide sequences.
• Identification of a novel Ag-selective peptide sequence.
• Characterization of binding motifs for Ag-binding peptides, providing a basis for future machine learning predictions.

Conclusions:

• The BESOS approach effectively accelerates the discovery of materials-selective peptides.
• The identified Ag-selective peptides and motifs will enable broader integration of Ag nanoparticles in biological systems.
• This work provides a roadmap for future AI-driven discovery of selective peptide sequences.