False positive reduction in protein-protein interaction predictions using gene ontology annotations
2007

Reducing False Positives in Protein Interaction Predictions

Sample size: 1042 publication Evidence: high

Author Information

Author(s): Mahdavi Mahmoud A, Lin Yen-Han

Primary Institution: Department of Chemical Engineering, University of Saskatchewan

Hypothesis

Can Gene Ontology annotations help reduce false positive predictions in protein-protein interaction datasets?

Conclusion

Using Gene Ontology annotations and specific rules can help remove false predicted protein interactions, improving the accuracy of computational predictions.

Supporting Evidence

  • The sensitivity of the top eight keywords was 64.21% in yeast and 80.83% in worm datasets.
  • The filtered datasets showed higher true positive fractions than non-filtered datasets.
  • The proposed algorithm can significantly reduce false positive predictions in protein-protein interaction datasets.

Takeaway

This study shows that using specific keywords from Gene Ontology can help scientists find the right protein pairs and avoid mistakes in predictions.

Methodology

The study used experimentally obtained protein pairs to extract keywords from Gene Ontology annotations and applied heuristic rules to filter predicted protein-protein interaction datasets.

Potential Biases

Potential bias due to reliance on existing annotations which may not be consistent or complete.

Limitations

The accuracy of predictions is limited by the quality and completeness of Gene Ontology annotations.

Participant Demographics

The study focused on protein interactions in two model organisms: Saccharomyces cerevisiae (yeast) and Caenorhabditis elegans (worm).

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-8-262

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication