As part of the ARPA-H APECx initiative, and led by PI Byung-Jun Yoon of Texas A&M, SPHERICAL (Scientific Platform for High Efficacy antigen design via Robust Integration of Computational experiments, AI, and protein modeLing) aims to accelerate the prediction of novel antigens with broad viral efficacy (ARPA-H).

The MADBio Lab is leading the development of a multiscale computational workflow for mapping molecular activity to functional outcomes in immunogen design.  Our Multiscale Modeling of Immune Responses Initiated by Antigen Dosing (MMIRIAD) model is a knowledge based, mechanistic model to predict immunogenicity, as it progresses over time from initial antigen dose to immune memory. It aims to bridge the gap between potential antigen-receptor interaction and functional antigenicity likely to produce a protective immune response in humans.  It does so by including critical dynamic features of the biological events that drive physiological responses from timescales of seconds to years. MMIRIAD enhances the SPHERICAL pipeline by integrating prior biological knowledge, predicting variability, and allowing causal interrogation of how an antigen candidate will perform within the body.