Programmable Virtual Humans Toward Human Physiologically-Based Drug Discovery
🤖 AI Summary
Current drug discovery relies heavily on traditional in vitro and animal experiments, which poorly predict human pharmacological efficacy and toxicity—leading to high clinical failure rates and a pronounced translational gap. To address this, we propose the “programmable virtual human” paradigm: an AI-driven, dynamic, multiscale physiological model spanning molecular, cellular, tissue, systemic, and phenotypic levels. It integrates cross-species single-cell and spatial multi-omics data with high-throughput perturbation experiments. Unlike static digital twins or conventional high-throughput screening, this model enables whole-body, otherwise infeasible virtual intervention studies. Its core contribution is the first end-to-end, physiology-anchored simulation of dynamic drug action in humans, providing quantitative early-stage predictions of both therapeutic efficacy and safety. This framework significantly improves drug development success rates and advances the field from empirical trial-and-error toward mechanism-driven discovery.
Application Category
📝 Abstract
Artificial intelligence (AI) has sparked immense interest in drug discovery, but most current approaches only digitize existing high-throughput experiments. They remain constrained by conventional pipelines. As a result, they do not address the fundamental challenges of predicting drug effects in humans. Similarly, biomedical digital twins, largely grounded in real-world data and mechanistic models, are tailored for late-phase drug development and lack the resolution to model molecular interactions or their systemic consequences, limiting their impact in early-stage discovery. This disconnect between early discovery and late development is one of the main drivers of high failure rates in drug discovery. The true promise of AI lies not in augmenting current experiments but in enabling virtual experiments that are impossible in the real world: testing novel compounds directly in silico in the human body. Recent advances in AI, high-throughput perturbation assays, and single-cell and spatial omics across species now make it possible to construct programmable virtual humans: dynamic, multiscale models that simulate drug actions from molecular to phenotypic levels. By bridging the translational gap, programmable virtual humans offer a transformative path to optimize therapeutic efficacy and safety earlier than ever before. This perspective introduces the concept of programmable virtual humans, explores their roles in a new paradigm of drug discovery centered on human physiology, and outlines key opportunities, challenges, and roadmaps for their realization.
Problem
Research questions and friction points this paper is trying to address.
Predicting drug effects in humans remains a fundamental challenge
Existing models lack resolution for molecular interactions and systemic consequences
Bridging the gap between early discovery and late development is critical
Innovation
Methods, ideas, or system contributions that make the work stand out.
AI-driven virtual humans for drug testing
Multiscale models simulate molecular interactions
Bridges gap between early and late discovery
🔎 Similar Papers
No similar papers found.
