We designed de novo protein binders to neutralize the Nipah virus Receptor Binding Protein (PDB: 2VSM). Our strategy targets conserved hotspots (specifically the Trp Gate, Salt Bridge, and Hydrophobic Floor) to achieve high-affinity binding with favorable developability.

Computational Pipeline

• Generation (Boltzgen): We produced 1,500 candidates (130–160 residues) with explicit binding constraints on critical residues (317, 346, 393, 401) and their neighbors to force direct engagement with the receptor.

• Filtering: Designs were initially ranked by design_to_target_iptm, min_interaction_pae, and developability metrics.

• Validation: The top 30 candidates underwent rigorous ensemble-based validation using Boltz-2 (diffusion sampling). Final selection was driven by IPSAE and pDockQ scores to identify designs with robust, high-confidence interface geometry across multiple structural predictions.