Predicting Co-Expression Patterns of Human Intronic microRNAs
Author Information
Author(s): Wang Dong, Lu Ming, Miao Jing, Li Tingting, Wang Edwin, Cui Qinghua
Primary Institution: Peking University Health Science Center
Hypothesis
Can a machine learning approach predict the co-expression of intronic microRNAs with their host genes?
Conclusion
The study developed a machine learning method that accurately predicts the co-expression patterns of human intronic microRNAs with their host genes.
Supporting Evidence
- The method achieved an accuracy of 79% in leave-one-out cross-validation.
- The classifier correctly predicted 20 out of 21 results in an independent testing dataset.
- The approach can be extended to other species.
Takeaway
This study created a computer program that helps scientists figure out where certain tiny RNA molecules are active in the body by looking at their relationship with nearby genes.
Methodology
A support vector machine (SVM) classifier was trained using feature vectors derived from a dataset of intronic microRNAs and their host genes.
Potential Biases
The method may miss detailed information that could improve prediction accuracy.
Limitations
The current training dataset is small, which may affect the reliability of the classifier.
Statistical Information
P-Value
0.02
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website