Selection of nitrogen responsive root architectural traits in spinach using machine learning and genetic correlations

Area of Science:

• Plant Science
• Agronomy
• Genetics

Background:

• Root architecture is crucial for plant nutrient acquisition, especially nitrogen (N).
• Optimizing nitrogen use efficiency (NUE) requires understanding root trait contributions.
• Soilless cultivation highlights the genetic basis of root traits and nutrient response.

Purpose of the Study:

• Identify root traits that best predict vegetative biomass under varying nitrogen (N) conditions.
• Determine genetic and phenotypic correlations between root traits and shoot weight (SWt).
• Prioritize root traits for selection to enhance NUE in spinach.

Main Methods:

• Grew 202 spinach lines in high (HN) and low (LN) nitrogen conditions.
• Measured eight image-derived root architectural traits.
• Utilized random forest machine learning to predict trait importance for shoot weight (SWt).
• Calculated broad-sense heritability (H) and genetic (r_g) and phenotypic (r_p) correlations.

Main Results:

• Number of root tips (Tips) and root length (RLength) were most important under HN.
• Crossings (Xsings) and root average diameter (RAvdiam) were most relevant under LN.
• A strong negative genetic correlation (r_g = -0.83) existed between SWt and RAvdiam under LN.
• RAvdiam showed no significant correlation with SWt under HN.

Conclusions:

• Root tip number and length are important for biomass under high nitrogen.
• Selecting against larger root average diameter under low nitrogen may improve biomass and NUE.
• Root trait selection strategies should consider nitrogen availability for optimal plant growth and yield.