Overview

• A team from Indiana University, Johns Hopkins Medicine, the University of Maryland School of Medicine and Oregon Health & Science University developed the human-interpretable hypothesis grammar.
• The framework converts spreadsheet-based biological rules into mathematical equations for agent-based modeling of multicellular systems.
• Validation experiments replicated breast tumor growth dynamics and predicted patient-specific immunotherapy responses in pancreatic cancer using genomics and spatial transcriptomics data.
• In neuroscience applications, the software simulated cortical layer formation by leveraging data from the Allen Brain Atlas.
• Ongoing work integrates artificial intelligence to automate model generation and link simulations to new data, enabling scalable in silico experimentation.