NARO historical phenotype dataset from rice breeding.

Area of Science:

• Agricultural Science
• Plant Breeding
• Genetics

Background:

• Phenotypic data and historical breeding trial records are crucial for enhancing breeding program efficiency.
• The National Agriculture and Food Research Organization (NARO) has accumulated extensive data from its rice breeding program.

Purpose of the Study:

• To organize historical data and develop a comprehensive phenotype dataset for the NARO rice breeding program.
• To analyze trends in 15 rice traits, including yield, and their relationship with cultivar release year.

Main Methods:

• Developed a historical phenotype dataset with 6052 records for 667 varieties (1991-2018).
• Utilized Best Linear Unbiased Predictions (BLUPs) and Principal Component Analysis (PCA) for data analysis.
• Examined relationships between traits and factors like cultivar release year and breeding location.

Main Results:

• Significant increases observed in yield-related traits (grains per panicle, plant weight, grain yield, thousand-grain weight) over time.
• Significant decrease in the number of panicles, increase in ripening time, and decrease in lodging degree and brown rice protein content.
• Principal Component Analysis (PCA) indicated greater trait diversity in more recent rice varieties.

Conclusions:

• Selection has favored panicle-weight-type, high-yielding rice varieties with improved lodging resistance.
• Observed trends in traits suggest distinct breeding objectives across different NARO research stations.
• The developed historical dataset provides valuable insights into long-term breeding progress and trait evolution in rice.