MGIDI: a powerful tool to analyze plant multivariate data

Area of Science:

• Agronomy
• Plant Breeding
• Quantitative Genetics

Background:

• Agronomic experiments commonly assess multiple traits, yet univariate analyses are often preferred over multivariate approaches.
• This preference may limit the full exploitation of complex datasets in agricultural research.

Purpose of the Study:

• To extend the multi-trait genotype-ideotype distance index (MGIDI) for analyzing multivariate data in simple and complex experimental designs.
• To introduce an optional weighting process to refine treatment rankings based on trait importance.

Main Methods:

• The study applied the extended MGIDI framework to simulated and real-world strawberry cultivation data.
• A factorial treatment structure involving cultivar, transplant origin, and substrate mixtures was analyzed.
• Twenty-two phenological, productive, physiological, and qualitative traits were evaluated.

Main Results:

• Most strawberry traits were significantly influenced by cultivar, transplant origin, substrate, and their interactions.
• The MGIDI identified specific cultivar and origin combinations (Albion/Imported and Camarosa/National) as superior.
• Optimal substrate formulations, such as 70% burned rice husk, improved water use efficiency.

Conclusions:

• The MGIDI offers a practical, robust, and user-friendly multi-trait framework applicable beyond plant breeding.
• It simplifies data presentation by reducing the number of tables and figures required.
• The MGIDI serves as a powerful tool for guiding researchers toward optimal treatment recommendations in multivariate studies.