Bias in Predicting Genetic Gain from Selection
Author Information
Author(s): José Marcelo Soriano Viana, Vinícius Ribeiro Silva, Admilson da Costa
Primary Institution: Universidade Federal de Viçosa
Hypothesis
The study investigates the extent of bias in predicting genetic gain from selection methods in breeding programs.
Conclusion
The predictions of genetic gain from selection methods are often biased, even when genetic parameters are estimated accurately.
Supporting Evidence
- The predicted gains were biased even with accurate genetic parameter estimates.
- Mass selection showed a tendency to overestimate genetic gains.
- Half-sib selection predictions were more biased than mass selection.
- Bias in predictions varied with the degree of dominance of the traits.
- Correlation between realized and predicted gains was generally high.
Takeaway
When scientists try to guess how much better plants will get from breeding, they often guess wrong, even if they have good information.
Methodology
The study simulated 10 cycles of a breeding program involving various traits, population classes, and selection methods to assess bias in genetic gain predictions.
Potential Biases
The predictions were found to be biased, with potential overestimation or underestimation of genetic gains.
Limitations
The study's predictions may not apply to all breeding scenarios due to the specific conditions and assumptions made during simulations.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website