Predicting Disease-Related Amino Acid Changes
Author Information
Author(s): Li Yizhou, Wen Zhining, Xiao Jiamin, Yin Hui, Yu Lezheng, Yang Li, Li Menglong
Primary Institution: Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University
Hypothesis
Can network topological features help predict disease-associated single amino acid polymorphisms (SAPs)?
Conclusion
Network topological features can effectively identify disease-associated SAPs and complement existing prediction methods.
Supporting Evidence
- The model achieved an accuracy of 83.0% and an MCC of 0.64 using 5-fold cross-validation.
- An independent dataset yielded an accuracy of 80.8% and an MCC of 0.59.
- Higher centrality values in protein structure networks are associated with disease-related mutations.
Takeaway
Scientists used a special method to look at how parts of proteins interact to figure out which changes in proteins might cause diseases.
Methodology
The study used a random forest algorithm to analyze network topological features of amino acid substitutions.
Limitations
The model's performance may vary with different datasets and may not capture all types of interactions.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website