In-Silico Validation of AI-Designed Conformer-Specific Binders Targeting PrPSc

This dataset documents the in-silico validation and design pipeline for conformer-specific therapeutic binders targeting the pathogenic prion protein isoform (PrP^Sc) in prion diseases. The project integrates cryo-EM structural data, AI-based binder design (AlphaFold-Multimer, RFdiffusion, ProteinMPNN), Rosetta energy evaluation, and translational delivery feasibility using lipid nanoparticles (LNPs) and focused ultrasound (FUS). Using conserved structural motifs of PrP^Sc (residues 90–230), multiple synthetic binder scaffolds were generated and optimized for binding affinity, structural stability, and diagnostic potential. Docking and energy analysis with Rosetta InterfaceAnalyzer predicted nanomolar-range affinities (ΔG ≈ –12 to –15 kcal mol⁻¹). Diagnostic fusion modeling demonstrated that luciferase or fluorescent tags can be incorporated without interfering with binding geometry, enabling a unified “diagnose–image–target–treat” paradigm for neuroproteinopathies. The dataset includes AlphaFold-predicted complexes (.pdb, .json), Rosetta scoring outputs, design scripts, SHA-256 hash verification files, and an authorship record establishing computational provenance. All computations were performed between April–October 2025 on publicly available prion structures (PDB 6LNI, 6UUR, 7DWV). This framework provides a reproducible, AI-driven blueprint for developing high-affinity, conformation-specific protein binders with translational potential across prion, tau, α-synuclein, and Huntington’s-like proteinopathies.