Physics-informed lead optimization of m102.4, the only human anti-Nipah antibody to complete Phase 1 clinical trials. Starting from the published VH/VL sequences, we introduced 12-20 mutations per variant using:

1. Interface-focused mutagenesis targeting NiV-G binding hotspots
2. Solubility optimization (low cysteine, balanced charge)
3. Glycan-clearance filtering to avoid steric clashes with NiV-G N-glycans
4. Helix propensity scoring for structural stability

All variants maintain the core antibody fold while exploring sequence space for improved binding affinity.