On Identifying the Optimal Number of Population Clusters via the Deviance Information Criterion
2011

Identifying Optimal Number of Population Clusters

Sample size: 1056 publication Evidence: high

Author Information

Author(s): Gao Hong, Bryc Katarzyna, Bustamante Carlos D.

Primary Institution: Stanford University

Hypothesis

Can the Deviance Information Criterion (DIC) effectively estimate the number of population clusters from genetic data?

Conclusion

The Deviance Information Criterion (DIC) outperforms other methods in estimating the number of population clusters in various genetic contexts.

Supporting Evidence

  • DIC consistently outperformed other methods in estimating the number of clusters in various demographic scenarios.
  • The study found that DIC is robust to small sample sizes but less effective with fewer genetic markers.
  • DIC was applied to real data from the Human Genome Diversity Panel, estimating five major population clusters.

Takeaway

This study shows a way to figure out how many groups of related people there are by looking at their genes, and it finds a good method to do this.

Methodology

The study used coalescent simulations and applied the Deviance Information Criterion to estimate the number of population clusters from genetic data.

Potential Biases

Potential biases may arise from the assumptions of the underlying probabilistic model used in InStruct.

Limitations

The accuracy of DIC can fluctuate based on the quality of individual classifications and may not perform well under high migration rates.

Participant Demographics

The study analyzed data from 1056 individuals across 52 populations.

Digital Object Identifier (DOI)

10.1371/journal.pone.0021014

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication