Multi-scale predictive model for plant stress response to drought and viral infection

Overview

This solution is a comprehensive multi-scale computational model that describes and predicts how plants react to both abiotic stress, specifically drought, and biotic stress, specifically viral infection. By directly incorporating cell-level and tissue-level models into the plant-level model structure, the framework improves the representation of plant behavior, clarifies central modeling assumptions, and guides the design of experiments to verify them. The approach moves crop yield optimization from qualitative observation toward quantitative, systems-based prediction, supporting precision agriculture applications where crop behavior must be anticipated and managed in response to environmental inputs.

Technical specifications

Key features:

• Multi-scale architecture linking cell-level, tissue-level, and plant-level models into a single integrated framework
• Cell-scale model capturing basic viral infection cycle events, validated in tomato
• Tissue-scale model representing viral replication and movement between cells and tissues
• Plant-scale model describing the evolution of viral population distribution across the entire plant
• Parameterization using literature results and experimental observations across all scales
• Uncertainty quantification of parameter values on model outcomes to identify critical knowledge gaps
• Planned fractal redesign of the plant model to more accurately portray plant growth and structure
• Integrated experimental design and model refinement loop comparing predictions to experimental findings at every scale
• Computational optimization to minimize runtime and enable practical deployment

Technology readiness level

The framework has been demonstrated through a proof-of-concept focused on tomato under viral infection stress, with cell-, tissue-, and plant-scale models constructed and linked. Parameter uncertainty effects on outcomes have been quantified, establishing a foundation for guiding targeted experiments. Next steps include updating the plant model to a fractal design, developing an integrated experiment-and-refinement workflow, and optimizing computational performance. Once validated for viral stress, the model is intended to be extended to other biotic and abiotic stresses, additional crop species, and ultimately integrated with real-time precision agriculture solutions.