Classifying transcription factor targets and discovering relevant biological features
2008

Classifying Transcription Factor Targets in Yeast

Sample size: 163 publication Evidence: moderate

Author Information

Author(s): Dustin T Holloway, Kon Mark, Charles DeLisi

Primary Institution: Boston University

Hypothesis

Can a supervised-learning approach improve the identification of transcription factor targets and reveal biological features contributing to regulation?

Conclusion

The study successfully predicts new transcription factor targets and provides insights into their regulatory functions.

Supporting Evidence

  • The method predicted targets for 163 transcription factors, revealing new insights into their regulatory roles.
  • Predictions were made available on a web server for further analysis.
  • Statistical enrichment was observed in biological processes related to carbon metabolism and energy generation.

Takeaway

The researchers figured out how to find new targets for proteins that control gene activity in yeast, helping us understand how genes work together.

Methodology

The study used a supervised learning approach with support vector machines to classify transcription factor targets based on various genomic datasets.

Potential Biases

The choice of negative training sets may introduce bias, as it is difficult to determine which genes are truly not bound by transcription factors.

Limitations

The predictions may contain false positives, and the biological significance of some predicted interactions needs further validation.

Statistical Information

P-Value

p ≤ 0.05

Statistical Significance

p ≤ 0.05

Digital Object Identifier (DOI)

10.1186/1745-6150-3-22

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication