Axia Discovery is a preclinical-stage biotech using a hybrid platform of physics-based molecular simulation, generative AI, and multi-omics data to design novel cyclic peptide therapeutics in silico. The company's ten patent-pending programs span oncology radioligand therapy, CNS, cardiometabolic, and renal indications, with wet-lab validation currently underway to advance candidates toward clinical development.
Funding
Funding not disclosed
Founders
Product
Problem
Traditional drug discovery relies heavily on large-scale library screening and iterative medicinal chemistry, which is time-consuming, expensive, and often fails to explore the full chemical space of potential therapeutics. This approach particularly limits the development of complex modalities like cyclic peptides, which are difficult to design rationally and optimize for specific targets.
Solution
Axia Discovery provides an end-to-end computational drug discovery platform that integrates physics-based molecular simulation, generative AI, and multi-omics data to design novel cyclic peptide therapeutics entirely in silico, before any synthesis occurs. The platform moves from target identification through hit-to-lead and lead optimization, generating patent-pending candidates that are then validated through wet-lab experimentation. Axia's approach enables de novo design of peptides—created from scratch rather than library-screened—targeting undisclosed receptors and antigens across multiple therapeutic areas. The company's pipeline currently comprises ten programs led by oncology radioligand therapy, with additional candidates in CNS, cardiometabolic, and renal indications, all supported by 11 U.S. provisional patent applications filed in 2026.
Target Audience
Pharmaceutical and biotechnology companies seeking computational co-development partnerships, licensing opportunities for preclinical-stage assets, or advancement of difficult therapeutic targets that resist conventional drug discovery approaches.
Features
- Hybrid computational platform combining physics-based molecular simulation with generative AI for candidate design
- De novo cyclic peptide design from scratch, eliminating reliance on existing compound libraries
- Multi-omics data integration to inform target selection and candidate optimization
- Pipeline of ten patent-pending programs spanning oncology (radioligand therapy and composition IP), CNS/neuropsychiatry, cardiometabolic, and renal indications
- Wet-lab validation underway for in-silico-designed candidates to confirm experimental performance
- Unbiased approach to difficult targets, including undisclosed GPCRs, cell-surface antigens, and clinically-validated receptors