Index selection with nonlinear profit function as a tool to achieve simultaneous genetic gain

Area of Science:

• Forestry
• Quantitative Genetics
• Plant Breeding

Background:

• Simultaneous improvement of multiple, often negatively correlated, traits is a key objective in forest tree breeding.
• Traditional breeding methods face challenges in balancing diverse trait improvements.

Purpose of the Study:

• To develop a novel linear index for optimizing multiple trait selection in forest tree breeding.
• To provide an efficient algorithm for calculating index coefficients, applicable to various breeding scenarios.

Main Methods:

• Development of a profit function incorporating linear and cross-product trait weights.
• Construction of a linear index based on the profit function.
• Detailed algorithm for determining index coefficients, avoiding direct search techniques.

Main Results:

• The proposed linear index effectively facilitates simultaneous improvement across multiple traits.
• The index is also effective for optimizing the selection based on trait ratios.
• The algorithm simplifies the calculation of index coefficients, enhancing practical application.

Conclusions:

• The developed linear index offers a robust tool for multi-trait selection in tree breeding.
• This approach provides a safety net against economic uncertainties by promoting balanced trait progress.
• The method offers statistical properties comparable to classical indices while simplifying the process.