Predicting Protein Phenotypes from Interaction Networks
Author Information
Author(s): Hu Lele, Huang Tao, Liu Xiao-Jun, Cai Yu-Dong
Primary Institution: Institute of Systems Biology, Shanghai University, Shanghai, China
Hypothesis
Can protein phenotypes be predicted using protein-protein interaction networks?
Conclusion
The proposed method successfully predicts protein phenotypes in yeast with a first-order accuracy of 65.4%.
Supporting Evidence
- The first-order prediction accuracy achieved was 65.4%.
- The likelihood of the first three predicted phenotypes including all real phenotypes was 70.6%.
- The method can be applied to predict protein phenotypes in other organisms.
Takeaway
Scientists can guess what a protein does by looking at how it interacts with other proteins, and this study shows a way to do that better.
Methodology
A network-based method was developed to predict protein phenotypes by analyzing protein-protein interactions.
Potential Biases
Potential biases may arise from false positive interactions in the protein-protein interaction network.
Limitations
The method's performance depends on the quality of the protein-protein interaction network and the number of seed proteins.
Participant Demographics
The study focused on budding yeast (Saccharomyces cerevisiae) proteins.
Statistical Information
P-Value
0.0001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website