SiteSeek: A New Method for Predicting Protein Phosphorylation Sites
Author Information
Author(s): Yoo Paul D, Ho Yung Shwen, Zhou Bing Bing, Zomaya Albert Y
Primary Institution: The University of Sydney
Hypothesis
Can a new machine learning model improve the prediction of protein phosphorylation sites?
Conclusion
SiteSeek outperforms existing predictors in identifying protein phosphorylation sites.
Supporting Evidence
- SiteSeek achieved 86.6% accuracy in predicting phosphorylation sites.
- The model showed 83.8% sensitivity and 92.5% specificity.
- SiteSeek outperformed nine existing machine learning models and four known predictors.
Takeaway
This study created a new tool called SiteSeek that helps scientists find where proteins get modified, which is important for understanding how cells work.
Methodology
The study used a new machine learning model called Adaptive Locality-Effective Kernel Machine and a dataset named PS-Benchmark_1 to train and test the model.
Potential Biases
The dataset is biased towards a few kinase families, which may affect the model's generalizability.
Limitations
The training dataset may contain incorrect information, and the model's performance could be biased towards certain kinase families.
Participant Demographics
The dataset includes 1,668 polypeptide chains from various protein databases.
Statistical Information
P-Value
0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website