MGIDI

MGIDI calculates a multi-trait genotype-ideotype distance index to analyze multivariate trait data and rank treatments or genotypes across simple and complex agronomic experimental designs.

Key Features:

• Multi-trait genotype-ideotype distance index: Computes a composite distance metric that summarizes multiple traits into a single index.
• Multivariate analysis: Addresses limitations of traditional univariate analyses and post-hoc tests by integrating multiple traits concurrently.
• Experimental design support: Extends theoretical foundations to one-way layouts with few treatments and traits as well as factorial treatment structures.
• Optional weighting process: Implements an optional trait-weighting mechanism to prioritize specific traits and modify treatment rankings accordingly.
• Treatment ranking: Produces ranked lists of treatments or factor combinations based on overall multivariate performance.
• Demonstrated on diverse data: Applicable to simulated datasets and real-world data, exemplified by a strawberry experiment with 22 phenological, productive, physiological, and qualitative traits.
• Balancing strengths and weaknesses: Integrates positive and negative trait contributions to identify overall favorable and unfavorable treatment profiles.

Scientific Applications:

• Plant breeding and agronomic experiments: Optimizes cultivar selection and factor combinations in experiments assessing multiple traits simultaneously.
• Strawberry factor optimization: Applied to optimize combinations of cultivar, transplant origin, and substrate mixture across 22 traits in a strawberry experiment.
• Identification of influential factors and outcomes: Identified cultivar, transplant origin, and substrate composition effects on strawberry traits, including Albion (imported transplants) and Camarosa (national transplants) as superior factor combinations and substrates with 70% burned rice husk as improving physical properties and water use efficiency.
• Method validation: Used with simulated data for demonstration and validation of the approach.

Methodology:

Computes the multi-trait genotype-ideotype distance index and applies an optional trait-weighting process to rank treatments from multivariate trait data.