Automatic detection of exonic splicing enhancers (ESEs) using SVMs
2008

Detecting Exonic Splicing Enhancers Using Machine Learning

Sample size: 4835 publication Evidence: high

Author Information

Author(s): Britta Mersch, Alexander Gepperth, Sándor Suhai, Agnes Hotz-Wagenblatt

Primary Institution: German Cancer Research Center (DKFZ)

Hypothesis

Can support vector machines (SVMs) effectively identify exonic splicing enhancers (ESEs) in human DNA sequences?

Conclusion

The study successfully developed a method using SVMs that can accurately classify potential ESE motifs with an accuracy of about 90%.

Supporting Evidence

  • The SVM with combined oligo kernel achieved a classification rate of 90.74%.
  • The method produced consistent training and test data leading to good classification rates.
  • Using SVMs with the combined oligo kernel yielded a high accuracy of about 90 percent.

Takeaway

The researchers created a computer program that helps find important parts of DNA that help make proteins by looking for special patterns called exonic splicing enhancers.

Methodology

The study used support vector machines with special sequence-based kernels to classify ESEs, training on both positive and negative examples derived from exon sequences.

Potential Biases

The reliance on unverified training data may introduce bias in the classification results.

Limitations

The method relies on heuristics for selecting training data, which may not always correspond to true ESEs.

Statistical Information

P-Value

p<0.001

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1186/1471-2105-9-369

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication