Related Experiment Videos
Probabilistic Boolean Networks: a rule-based uncertainty model for gene regulatory networks.
Ilya Shmulevich1, Edward R Dougherty, Seungchan Kim
1Cancer Genomics Laboratory, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 85, Houston, TX 77030, USA. is@ieee.org
Bioinformatics (Oxford, England)
|February 16, 2002
Summary
Probabilistic Boolean Networks (PBNs) model gene regulation with uncertainty. This approach allows for quantifying gene influence and studying network dynamics using Markov chains and Bayesian networks.
Area of Science:
- Computational Biology
- Systems Biology
- Genetics
Background:
- Developing models for genetic regulatory networks is crucial for understanding gene interactions.
- Existing models often struggle with data uncertainty and quantifying gene influence.
Purpose of the Study:
- To construct a model for genetic regulatory networks that handles uncertainty.
- To enable systematic study of global network dynamics.
- To quantify the relative influence and sensitivity of genes.
Main Methods:
- Introduction of Probabilistic Boolean Networks (PBNs) as an extension of Boolean networks.
- Utilizing Markov chains for studying network dynamics.
- Establishing relationships between PBNs and Bayesian networks.
Main Results:
- PBNs offer robustness against uncertainty while retaining rule-based properties.
- Network dynamics can be analyzed probabilistically, with standard Boolean networks as a special case.
- Methods for quantifying gene influence within PBNs are presented.
Conclusions:
- PBNs provide a robust framework for modeling genetic regulatory networks.
- The model facilitates the analysis of network dynamics and gene interactions.
- PBNs integrate well with Bayesian network concepts for probabilistic dependency analysis.