Computational Methods for Peptide Discovery
Axia Discovery's platform pairs physics-based molecular simulation with generative AI to design novel, ownable targeting ligands for hard receptor targets — in silico, before synthesis.
De Novo Design
Generative AI designs novel cyclic-peptide sequences from scratch against a chosen target — ownable compositions of matter optimized for target engagement and drug-like properties.
Physics-Based Simulation
Physics-based molecular simulation predicts how candidates bind and behave, prioritizing the most selective, highest-quality designs before any synthesis.
Selectivity by Design
Candidates are engineered to discriminate the intended target from its closest receptor relatives — selectivity built in from the first design.
Developability
Computational assessment of PK, safety, and developability liabilities early in the campaign — surfacing risk before the wet lab.
Peptide Chemistry
Cyclization and stabilization strategies that improve metabolic stability and developability of the designed peptides.
Wet-Lab Validation
A closed loop between computational design and laboratory testing — candidates are designed in silico, with wet-lab validation underway.
Why Peptides?
↑ Target Specificity
Peptides offer exquisite selectivity for challenging targets like protein-protein interactions and conformational epitopes that small molecules struggle to address.
↑ Chemical Space
The peptide chemical space is vast and largely unexplored. Generative design enables rapid optimization of potency, selectivity, and developability.
↑ Scalability
Generative design explores far more of this space than synthesis-led screening — novel candidates are designed in silico, before committing a single synthesis.
Technology Foundation
Generative AI
Generative models design novel cyclic-peptide binders from scratch against a chosen target — proposing ownable compositions of matter rather than selecting from existing libraries.
Physics-Based Simulation
Physics-based molecular simulation models binding and conformational behavior, prioritizing selective, high-quality designs before synthesis.
Multi-Omics Data
A multi-modal data foundation spanning target genetics, expression, and structural biology informs target selection and design.
Novel peptides designed from scratch — not library-screened
Designed before synthesis — wet-lab validation underway
Patent-pending programs / U.S. provisionals filed 2026
Want to learn more about our technology?
Schedule a Discussion