Related Experiment Videos
Maximum likelihood analysis of rare binary traits under different modes of inheritance
G Thaller1, L Dempfle, I Hoeschele
1Institut für Tierwissenschaften, Technischa Universitat München-Weihenstephan, Germany.
Genetics
|August 1, 1996
Summary
Maximum likelihood methods accurately identified major genes influencing rare binary traits in swine, even with complex polygenic backgrounds. This approach reliably distinguished genetic models, crucial for effective breeding programs.
Area of Science:
- Quantitative genetics
- Animal breeding
- Statistical genetics
Background:
- Determining the mode of inheritance for rare binary traits is challenging in complex populations.
- Understanding genetic architecture is vital for optimizing livestock breeding strategies.
Purpose of the Study:
- To apply maximum likelihood methodology for identifying major genes in swine populations.
- To evaluate the performance of different genetic models, including polygenic and mixed major gene models.
Main Methods:
- Utilized maximum likelihood estimation with deterministic algorithms for model fitting.
- Conducted a simulation study with varying population structures (10/10, 30/30, 100/30 sire/dam designs).
- Employed the Average Information Criterion (AIC) for model selection.
Main Results:
- Successfully detected major gene effects with high power under mixed inheritance models, except in the smallest simulated design.
- Parameter estimates were largely unbiased with acceptable standard errors, enabling clear model discrimination.
- Empirically evaluated likelihood ratio statistics as asymptotic theory was insufficient.
Conclusions:
- Maximum likelihood is effective for detecting major genes influencing rare binary traits in swine.
- Larger population sizes improve the power and precision of genetic model detection.
- The Average Information Criterion reliably selected the correct underlying genetic model in simulations.