Choosing Organisms for Studying Protein Interactions
Author Information
Author(s): Herman Dorota, Ochoa David, Juan David, Lopez Daniel, Valencia Alfonso, Pazos Florencio
Primary Institution: National Centre for Biotechnology (CNB-CSIC)
Hypothesis
The performance of methodologies for predicting protein interactions depends on the set of organisms used to build phylogenetic trees.
Conclusion
Using different sets of organisms based on computational resources and the type of interactions can optimize the prediction of protein interactions.
Supporting Evidence
- The performance of methodologies depends on the organism set used for building phylogenetic trees.
- Certain subsets of organisms are more suitable for predicting specific types of interactions.
- Using too many organisms can introduce redundancy that negatively affects predictions.
Takeaway
This study shows that picking the right group of organisms is important for figuring out how proteins interact with each other.
Methodology
The study evaluated the performance of three mirrortree-related methodologies using different subsets of organisms based on taxonomic criteria.
Potential Biases
The methodologies may be affected by phylogenetic redundancy and biases in organism selection.
Limitations
The study did not explore the effects of other taxonomic criteria combinations or the specific interactions detected.
Statistical Information
P-Value
p ≤ 10E-5
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website