Measuring Gene Interactions in Time Series Data
Author Information
Author(s): Xin Gao, Daniel Q Pu, Peter X-K Song
Primary Institution: York University
Hypothesis
Can a new measure of gene-gene interaction based on transition probabilities reveal nonlinear dependencies in time series microarray data?
Conclusion
The proposed method effectively identifies significant gene-gene interactions that traditional correlation methods fail to detect.
Supporting Evidence
- The method identified 302 significant genes interacting with CD44.
- Traditional correlation methods failed to detect many of these interactions.
- The proposed method revealed new insights into gene regulatory networks.
Takeaway
This study shows how genes talk to each other over time, helping scientists understand their relationships better.
Methodology
The study uses hidden Markov models to analyze time series microarray data and assess gene-gene interactions through transition probabilities.
Limitations
The method assumes stationarity in the hidden processes, which may not hold true in all biological contexts.
Statistical Information
P-Value
0.00110891
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website