Assessing Performance of Orthology Detection Strategies Applied to Eukaryotic Genomes
2007

Evaluating Orthology Detection Methods

Sample size: 27562 publication 10 minutes Evidence: moderate

Author Information

Author(s): Chen Feng, Mackey Aaron J., Vermunt Jeroen K., Roos David S.

Primary Institution: University of Pennsylvania

Hypothesis

How do different orthology detection methods compare in terms of performance?

Conclusion

The study provides a comprehensive evaluation of various orthology detection methods, highlighting trade-offs between sensitivity and specificity.

Supporting Evidence

  • Two algorithms, INPARANOID and OrthoMCL, showed the best balance of sensitivity and specificity.
  • BLAST-based methods were characterized by high sensitivity, while tree-based methods showed high specificity.
  • Latent Class Analysis allows for the estimation of false positive and false negative rates in the absence of a gold standard.

Takeaway

This study looks at different ways to find similar genes in different species and shows which methods work best.

Methodology

Latent Class Analysis was used to evaluate the performance of various orthology detection methods on a dataset of eukaryotic genomes.

Potential Biases

Conditional dependencies between methods may affect performance estimates.

Limitations

The lack of a genomic-scale 'gold standard' dataset limits the ability to comprehensively assess performance.

Participant Demographics

The study analyzed protein sequences from six eukaryotic genomes.

Statistical Information

P-Value

0.48

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1371/journal.pone.0000383

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication