Characterization of a Bayesian genetic clustering algorithm based on a Dirichlet process prior and comparison among Bayesian clustering methods
2011

Bayesian Genetic Clustering Algorithm Characterization

publication Evidence: moderate

Author Information

Author(s): Onogi Akio, Nurimoto Masanobu, Morita Mitsuo

Primary Institution: Maebashi Institute of Animal Science, Livestock Improvement Association of Japan, Inc.

Hypothesis

Can a Bayesian approach based on a Dirichlet process prior improve the inference of genetic population structures?

Conclusion

The DP prior method is effective for inferring population structures, especially with unbalanced sample sizes.

Supporting Evidence

  • The SAMS sampler improved the accuracy of population structure analysis.
  • The DP prior method is suitable for data sets with unbalanced sample sizes.
  • Inferring the hyperparameter for allele frequencies can enhance the method's performance.

Takeaway

This study shows a new way to group animals based on their genes, which works better when there are different numbers of animals in each group.

Methodology

The study evaluated a new sampler for Bayesian clustering and compared it with existing methods using simulated and real data sets.

Limitations

The effectiveness of the method may vary depending on the uniformity of allele frequencies among loci.

Digital Object Identifier (DOI)

10.1186/1471-2105-12-263

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication