Improved homology-driven computational validation of protein-protein interactions motivated by the evolutionary gene duplication and divergence hypothesis
2009

Improved Methods for Validating Protein-Protein Interactions

Sample size: 135276 publication Evidence: high

Author Information

Author(s): Christian Frech, Michael Kommenda, Viktoria Dorfer, Thomas Kern, Helmut Hintner, Johann W Bauer, Kamil Önder

Primary Institution: Upper Austria University of Applied Sciences

Hypothesis

The sensitive search for homologous protein-protein interactions (PPIs) should be a successful strategy for PPI validation.

Conclusion

New PPIs are primarily derived from preexisting PPIs and not invented de novo, making the search for homologous PPIs an efficient strategy to identify biologically relevant interactions.

Supporting Evidence

  • The study compiled the largest collection of known physical binary PPIs from multiple sources.
  • ROC curves demonstrated the high efficacy of the proposed validation method.
  • The method improves upon previous binary validation techniques by incorporating multiple homology detection methods.

Takeaway

Scientists found that most protein interactions come from older interactions, so looking for similar ones can help confirm if a new interaction is real.

Methodology

The study combined FASTA and PSI-BLAST to search for homologous protein pairs in a large PPI database and used a novel scoring scheme to validate interactions.

Limitations

The incompleteness of current PPI data sets and the elusive nature of common ancestry represent major practical obstacles.

Digital Object Identifier (DOI)

10.1186/1471-2105-10-21

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication