Predicting Functions of Proteins in Mouse Based on Weighted Protein-Protein Interaction Network and Protein Hybrid Properties
2011

Predicting Protein Functions in Mice Using Interaction Networks

Sample size: 14732 publication Evidence: moderate

Author Information

Author(s): Hu Lele, Huang Tao, Shi Xiaohe, Lu Wen-Cong, Cai Yu-Dong, Chou Kuo-Chen

Primary Institution: Institute of Systems Biology, Shanghai University, Shanghai, China

Hypothesis

Can a new computational method effectively predict the functions of uncharacterized proteins in mice?

Conclusion

The new approach shows promising results in predicting protein functions based on interaction networks and hybrid properties.

Supporting Evidence

  • The new predictor achieved a first-order success rate of 69.1% on the training set.
  • The success rate for the test set was 70.2%.
  • The method combines protein interaction data with biochemical properties for better predictions.

Takeaway

Scientists created a computer program to guess what proteins do in mice, and it worked pretty well!

Methodology

The study used a combination of protein-protein interaction data and biochemical properties to predict protein functions.

Limitations

The method may not work well for proteins without interaction data.

Participant Demographics

Proteins from mouse species were analyzed.

Digital Object Identifier (DOI)

10.1371/journal.pone.0014556

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication