Selecting Thresholds in Gene Networks Using Graph Theory
Author Information
Author(s): Andy D Perkins, Michael A Langston
Primary Institution: Mississippi State University
Hypothesis
Can spectral graph theory methods provide a systematic approach for selecting thresholds in gene co-expression networks?
Conclusion
The proposed method offers a more systematic and conservative approach to threshold selection compared to traditional methods.
Supporting Evidence
- The method uses eigenvalues and eigenvectors to analyze gene relationships.
- It was applied to well-studied microarray datasets from humans and yeast.
- The spectral method identified potential threshold values of 0.78 for yeast and 0.83 for human networks.
Takeaway
This study shows a new way to pick important connections between genes by using math, which helps scientists understand how genes work together better.
Methodology
The study used spectral graph theory to analyze gene co-expression networks and identify appropriate thresholds based on community structure.
Limitations
The method may require significant computational resources and may not be applicable to all types of networks.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website