Image
Providence, R.I. – September 08, 2026 -- EpiVax, Inc. has launched artificial intelligence-driven upgrades to its proprietary ISPRI platform, adding a new JanusMatrix 2.1 model, a tolerance-adjusted immunogenicity score, and an enhanced anti-drug antibody (ADA) prediction model for biotherapeutic developers.
JanusMatrix 2.1 applies AI to weight cross-conservation of human epitopes using expression and prevalence data.
The update is designed to produce more precise characterization of therapeutic epitopes and their tolerance profiles, building on EpiVax's existing EpiMatrix and JanusMatrix algorithms that assess T-cell epitope content and human cross-conservation.A new JanusMatrix-Adjusted EpiMatrix (JAX) score integrates tolerance potential directly into immunogenicity scoring.
The JAX score is built to align more closely with clinical immunogenicity outcomes by incorporating JanusMatrix tolerance data into EpiMatrix scores, giving developers a biologically grounded read on effector epitope density and sequence-derived risk across therapeutic modalities.ADA 2.2 combines epitope-based measures with biophysical attributes and mode-of-action criteria.
Trained on a diverse set of clinical monoclonal antibodies, the updated model shows a stronger correlation between predicted and clinically observed immunogenicity, adding confidence to risk interpretation when paired with the JAX score.EpiVax said the enhancements align with recent FDA initiatives promoting adoption of human-relevant New Approach Methodologies (NAMs), helping biologics sponsors identify immunological risk, protect asset value, and reduce development program risk across the product lifecycle.
Published by
fairsonline_team
Industries
Company
Products
News Type