Biohub Launches AI Protein Biology Platform That Outperforms AlphaFold 3 and Maps 6.8 Billion Sequences
Summary
Biohub launches a groundbreaking open-source AI protein biology platform featuring three powerful tools that outperform AlphaFold 3 in antibody-antigen prediction, design cancer-targeting protein binders with up to 88% lab-validated hit rates, and make 6.8 billion protein sequences searchable to accelerate global scientific discovery.
Key Points
- Biohub releases a world model of protein biology featuring three open tools — ESMC, ESMFold2, and ESM Atlas — designed to predict protein structures, design functional binders, and map the protein universe across billions of sequences.
- ESMFold2 outperforms AlphaFold 3 in antibody-antigen structure prediction and successfully designs protein binders against five cancer and immunology targets in days, achieving lab-validated hit rates of up to 88% with therapeutic-level affinity and specificity.
- ESM Atlas makes 6.8 billion protein sequences and 1.1 billion predicted structures searchable using AI-learned relationships, uncovering previously unannotated biology and evolutionary connections invisible to existing databases — all freely available to the global scientific community.