Predicting gene function in a hierarchical context with an ensemble of classifiers
2008

Predicting Gene Function Using Machine Learning

Sample size: 1726 publication 10 minutes Evidence: high

Author Information

Author(s): Guan Yuanfang, Myers Chad L, Hess David C, Barutcuoglu Zafer, Caudy Amy A, Troyanskaya Olga G

Primary Institution: Princeton University

Hypothesis

Can an ensemble of classifiers improve gene function prediction in multicellular organisms?

Conclusion

The ensemble method consistently performs among the top methods in the MouseFunc evaluation, indicating its effectiveness in predicting gene functions.

Supporting Evidence

  • The ensemble method achieved an average AUC of 0.72 across various GO terms.
  • The method performed in the top three of the nine MouseFunc submissions for average AUC.
  • It demonstrated good classification performance across diverse cellular processes.

Takeaway

The researchers created a smart computer program that helps figure out what genes do by looking at lots of data, and it works really well for both simple and complex organisms.

Methodology

An ensemble framework based on support vector machines was used to integrate diverse datasets in the context of the Gene Ontology hierarchy.

Potential Biases

Potential bias due to reliance on existing annotations and the quality of input data.

Limitations

The method's performance may vary based on the size of the GO terms and the availability of positive examples.

Participant Demographics

The study focused on gene functions in laboratory mice and Saccharomyces cerevisiae.

Statistical Information

P-Value

0.003

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/gb-2008-9-s1-s3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication