Digital crop models for enhanced breeding performance and climate resilience

Overview

Crop breeding programs face a critical challenge: developing varieties that are both more productive and more resilient to climate change. While genomic selection is the industry standard for optimizing breeding programs—an area with approximately $3 billion in annual R&D investment—it has limitations in predicting complex traits such as yield due to constraints in available learning datasets. iMEAN's approach addresses this gap by injecting mechanistic, systems biology information into the prediction process, enabling more accurate forecasting of crop performance.

This technology is particularly valuable for breeding companies seeking to improve prediction accuracy for traits influenced by multiple genes, epistatic interactions, and environmental variations. By combining proprietary databases, advanced algorithms, and expert curation, iMEAN reconstructs genome-scale mathematical models that simulate complex biological systems from DNA sequences and omics datasets. The result is a predictive platform that can help accelerate breeding decisions and reduce the time and cost of developing improved crop varieties.

Technical specifications

Core approach:

• Builds fine-grained mathematical models that reproduce plant behavior in silico based on causal mechanisms
• Integrates systems biology approaches with genomic selection to predict complex traits
• Leverages proprietary databases, algorithms, and expert curation to generate high-quality predictive models

Validated performance:

• Proof of concept demonstrated on Arabidopsis thaliana as a model organism
• Achieved up to 50% improvement in prediction accuracy compared to state-of-the-art genomic selection methods
• Validated in both greenhouse and field conditions
• Capable of modeling emergent interactions between genetic network components, including epistatic effects

Target applications:

• Predicting crop performance for traits such as yield and pathogen resistance
• Applicable to multiple crop species including tomato, rapeseed, sunflower, maize, and soybean
• Supporting breeding program decisions through retrospective validation and prospective prediction

Technology readiness level

The technology has been validated at TRL 6 through a successful proof of concept on Arabidopsis thaliana, demonstrating significant improvements in prediction accuracy under both greenhouse and field conditions. The next development phase involves reconstructing digital models for crop species such as tomato, rapeseed, and sunflower, followed by demonstration on commercially relevant crops. iMEAN is actively seeking partnerships with breeding companies to validate the technology using historical breeding data and to test new predictions for traits of commercial interest. Successful demonstration on a crop species would advance the technology toward broader commercial deployment in agricultural breeding programs.