Efficient parameter search for qualitative models of regulatory networks using symbolic model checking

Area of Science:

• Systems Biology
• Computational Biology
• Bioinformatics

Background:

• Analyzing complex biological networks requires assessing the consistency between network structure and observed behavior.
• Key questions involve verifying if a hypothesized regulatory network structure aligns with experimental data or can generate specific system dynamics.

Purpose of the Study:

• To develop a computational method for parameter searching in qualitative models of biological regulatory networks.
• To address the challenge of determining if a given network structure is consistent with observed system behavior or can produce desired behaviors.

Main Methods:

• A novel method based on symbolic model checking is developed to avoid exhaustive enumeration of all possible parameter combinations.
• The approach is applied to qualitative models of regulatory networks, enabling efficient parameter space exploration.

Main Results:

• The symbolic model checking method demonstrates effective performance on real biological problems, including the IRMA synthetic network and benchmark datasets.
• The study validates the consistency between the IRMA model and time-series gene expression profiles.
• Parameter modifications were identified to enhance the robustness of external control over system behavior.

Conclusions:

• Symbolic model checking provides an efficient and scalable approach for analyzing biological regulatory network models.
• The method facilitates the validation of network structures against experimental data and aids in designing synthetic biological systems with desired functionalities.