Hyppo-X

Hyppo-X performs unsupervised structure discovery on high-dimensional phenomics data to reveal higher-order relationships among genotype, phenotype, and environmental variables using algebraic topology and graph theory.

Key Features:

• Unsupervised Structure Discovery: Hyppo-X identifies higher-order structures and latent relationships in phenomics datasets without predefined labels.
• Environmental Influence Analysis: It delineates how environmental factors influence phenotypic traits across genotypes and temporal scales.
• Scalability and Flexibility: Implemented as a header-only library, it scales to large phenomics datasets and supports category definitions using single or multiple dataset columns.

Scientific Applications:

• Maize phenomics analysis: Applied to real-world maize datasets to characterize environmental impacts on phenotypic traits.
• Hypothesis extraction: Formalizes systematic hypothesis generation and extraction from complex phenomics data.

Methodology:

Hyppo-X applies unsupervised learning grounded in algebraic topology and graph theory to uncover higher-order structures in phenomics data.