Using Artificial Neural Networks to Analyze Gene Interactions
Author Information
Author(s): Liu Ying, Duan Weimin, Paschall Justin, Saccone Nancy L
Primary Institution: Washington University School of Medicine
Hypothesis
Can artificial neural networks effectively identify gene interactions and linkage signals in quantitative traits?
Conclusion
Artificial neural networks can analyze quantitative traits and detect gene interactions, but achieving stable results is challenging.
Supporting Evidence
- The study identified linkage signals consistent with traditional methods.
- Artificial neural networks showed potential for analyzing complex gene interactions.
- Cross-validation did not significantly improve the stability of results.
Takeaway
This study used computers to find out how different genes work together, but it was hard to get clear answers.
Methodology
The study used SNP genotypes and gene expression phenotypes from CEPH pedigrees, applying artificial neural networks to analyze linkage and interactions.
Limitations
Some results were noisy, and the method struggled with stability in the presence of real data.
Participant Demographics
14 three-generation CEPH Utah families
Statistical Information
P-Value
8.8 × 10-12
Statistical Significance
p<0.0001
Want to read the original?
Access the complete publication on the publisher's website