Genomic Prediction of Green Fraction Dynamics in Soybean Using Unmanned Aerial Vehicles Observations

Area of Science:

• Agricultural Science
• Plant Breeding
• Genomics

Background:

• High-throughput phenotyping systems are increasing the availability of plant growth data.
• Genomic prediction applied to growth data can predict the performance of untested genotypes.
• Integrating growth modeling with genomic prediction is crucial for advanced crop breeding but remains understudied.

Purpose of the Study:

• To develop and evaluate a novel growth prediction scheme by combining dynamic growth modeling with genomic prediction.
• To assess the contribution of a dynamic model in predicting the growth process of soybean genotypes.
• To enhance prediction accuracy for early selection in crop breeding programs.

Main Methods:

• Acquired phenotype data for 198 soybean genotypes over 3 years.
• Utilized Unmanned Aerial Vehicle (UAV) remote sensing to measure longitudinal changes in green fractions.
• Fitted a dynamic model to green fraction data to extract five growth parameters and developed genomic prediction models using these parameters.

Main Results:

• The dynamic model effectively captured genetic diversity in soybean growth characteristics.
• The proposed two-step prediction model (genomic prediction of dynamic parameters) demonstrated higher accuracy than conventional genomic prediction.
• Improved prediction accuracy was particularly notable when predicting future growth using early-stage observed data.

Conclusions:

• The novel integrated approach successfully combines early growth information with genomic data for accurate prediction.
• This method offers a significant advancement for early-stage selection in crop breeding, potentially reducing costs and time.
• The findings highlight the utility of dynamic growth modeling within genomic prediction frameworks for plant breeding applications.