Predicting the Network of Substrate-Enzyme-Product Triads
Author Information
Author(s): Chen Lei, Feng Kai-Yan, Cai Yu-Dong, Chou Kuo-Chen, Li Hai-Peng
Primary Institution: Institute of Systems Biology, Shanghai University
Hypothesis
Can a computational approach effectively predict the network of substrate-enzyme-product triads?
Conclusion
The developed computational method shows a high success rate in predicting the substrate-enzyme-product network.
Supporting Evidence
- The model achieved a 98.71% overall success rate in predicting the network for testing samples.
- 95.41% success rate was observed for the testing networking triads.
- 729,600 non-networking triads were included in the benchmark dataset.
Takeaway
This study created a computer program that can guess how different substances work together in living things, which is usually hard to figure out.
Methodology
A mathematical model was developed using a nearest neighbour algorithm to predict triads based on compound similarity and functional domain composition.
Limitations
The method may perform poorly for query triads without any similarity to the training datasets.
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