Related Experiment Videos
Pedigree analysis package (PAP) vs. MORGAN: model selection and hypothesis testing on a large pedigree
1Department of Biostatistics, University of Washington, Seattle 98195-7720, USA.
Abstract:
The MORGAN package of programs is compared to a commonly used package, PAP, with respect to model selection in segregation analysis of a quantitative trait. MORGAN uses Monte Carlo Markov chain (MCMC) methods to estimate the likelihood, whereas both versions of PAP used employ an approximation to the likelihood for the mixed model. Comparisons are done by using results obtained from simulated data. All simulations were done on the same 232-member pedigree using data generated under each of several variations of models, which included different combinations of environmental, polygenic, and major gene components. PAP, version 4.0, and MORGAN gave similar results with respect to model selection for the majority of situations, suggesting that MCMC methods provide a computationally tractable approach for analysis of more complex models that cannot be analyzed by more direct computational methods. PAP, version 3.0, gave somewhat more disparate results compared with either PAP version 4.0 or MORGAN. Both MORGAN and the two versions of PAP confirmed that the major gene component is much easier to detect in the presence of some dominance. All three packages frequently falsely accepted the polygenic model when there was high residual heritability.
Insights
The MORGAN package, using Monte Carlo Markov chain (MCMC) methods, shows comparable model selection to PAP for quantitative trait segregation analysis. MCMC offers a viable approach for complex genetic models.
Area of Science:
- Biostatistics
- Statistical Genetics
- Computational Biology
Background:
- Segregation analysis is crucial for understanding the genetic architecture of quantitative traits.
- Accurate model selection is essential for identifying major genes and polygenic effects.
- Existing software packages like PAP use approximations, potentially limiting analysis of complex models.
Purpose of the Study:
- To compare the MORGAN package, utilizing Monte Carlo Markov chain (MCMC) methods, with the PAP package for model selection in quantitative trait segregation analysis.
- To evaluate the performance of MCMC-based likelihood estimation against approximated likelihoods in mixed models.
- To assess the utility of MORGAN for analyzing complex genetic models intractable by direct computational methods.
Main Methods:
- Simulated data from a 232-member pedigree were analyzed.
- Models included various combinations of environmental, polygenic, and major gene effects.
- MORGAN's MCMC approach was compared against two versions of the PAP package (versions 3.0 and 4.0) employing approximated likelihoods.
Main Results:
- MORGAN and PAP version 4.0 yielded similar model selection results across most simulated scenarios.
- PAP version 3.0 produced more divergent results compared to MORGAN and PAP version 4.0.
- All tested packages (MORGAN, PAP v3.0, PAP v4.0) detected major gene components more readily with dominance.
- A frequent finding was the false acceptance of the polygenic model when residual heritability was high.
Conclusions:
- Monte Carlo Markov chain (MCMC) methods, as implemented in MORGAN, provide a computationally tractable and reliable approach for complex segregation analysis.
- MORGAN offers a robust alternative for genetic model selection, especially for scenarios beyond the scope of approximated likelihood methods.
- The presence of dominance significantly enhances the detectability of major gene effects in segregation analyses.