Mining Rules for Enzyme Classification
Author Information
Author(s): Chiu Shih-Hau, Chen Chien-Chi, Yuan Gwo-Fang, Lin Thy-Hou
Primary Institution: Food Industry Research and Development Institute
Hypothesis
Can an association algorithm effectively classify enzyme classes based on protein domain composition?
Conclusion
The method can classify enzyme classes based on enzyme domain rules, aiding both manual and automatic annotation.
Supporting Evidence
- A correct enzyme classification rate of 70% was obtained for prokaryote datasets.
- An 80% classification rate was achieved for eukaryote datasets.
- 88 out of 5085 test entries matched with the fungus rule set.
Takeaway
This study shows a way to automatically sort proteins into enzyme classes using rules from their building blocks, which helps scientists understand them better.
Methodology
The study used an association algorithm to mine rules from protein databases for enzyme classification.
Limitations
The coverage of the method is low, depending on the quality of the training dataset.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website