Abstact

Designing antibodies is complex and resource intensive. While deep learning and generative approaches have shown promise in the design of protein binders, achieving high affinity and stability remains challenging. We introduce EvolveX, a structure-based antibody design pipeline leveraging the empirical force field FoldX to design complementarity-determining regions (CDRs) of single-domain antibodies (VHHs). We demonstrate the ability of EvolveX to redesign a VHH targeting mouse Vsig4 (mVsig4) to address two challenges: enhancing stability and affinity for mVsig4 and redesigning it for high affinity to the human ortholog. Notably, EvolveX improved the binding affinity of VHHs to human Vsig4 by over 1,000-fold. Structural analyses by X-ray crystallography confirmed design accuracy. Next-generation sequencing (NGS) analysis further demonstrated the efficiency of FoldX-based design pipeline. Collectively, our study highlights EvolveX's potential to overcome current limitations in antibody design, offering a powerful tool for the development of therapeutics with enhanced specificity, stability, and efficacy.