LBF-MI: A New Method for Inferring Gene Regulatory Networks
Author Information
Author(s): Barman Shohag, Farid Fahmid Al, Gope Hira Lal, Hafiz Md. Ferdous Bin, Khan Niaz Ashraf, Ahmad Sabbir, Mansor Sarina, Lonardi Stefano
Primary Institution: Multimedia University
Hypothesis
Can a novel method combining limited Boolean functions and multivariate mutual information improve the inference of gene regulatory networks from time-series gene expression data?
Conclusion
The LBF-MI method outperformed existing techniques in inferring gene regulatory networks from both artificial and real datasets.
Supporting Evidence
- LBF-MI significantly outperformed mutual information-based Boolean network inference, context likelihood relatedness, and relevance network on artificial datasets.
- The method successfully reconstructed all nine interactions in the E. coli gene regulatory network.
- LBF-MI achieved higher F-scores, precision, and recall compared to other methods in both artificial and real datasets.
Takeaway
This study created a new way to understand how genes control each other by looking at their activity over time, and it works better than older methods.
Methodology
The LBF-MI method uses limited Boolean functions and multivariate mutual information in a two-phase approach to infer gene regulatory networks.
Limitations
The method has slower running times compared to other techniques due to its comprehensive search process.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website