Improving Inference of Genetic Networks Using Criticality
Author Information
Author(s): Wenbin Liu, Harri Lähdesmäki, Edward R Dougherty, Ilya Shmulevich
Primary Institution: Institute for Systems Biology
Hypothesis
Incorporating the assumption of criticality in the inference process will reduce the inference error of Boolean networks.
Conclusion
The proposed method improves the accuracy of predicting genetic networks, especially with small sample sizes.
Supporting Evidence
- The proposed method significantly reduces inference error in small sample situations.
- Taking criticality into account improves prediction accuracy for state transitions and network wiring.
- Performance of both methods becomes similar as sample size increases.
Takeaway
This study shows that using a special assumption about how genes interact can help scientists better understand genetic networks, especially when they have only a little data.
Methodology
The study used simulations of Boolean networks to analyze the impact of incorporating criticality into the inference process.
Limitations
The performance of the proposed method decreases as sample size increases, and the true sensitivity of networks is often unknown.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website