Boolean network sketches: a unifying framework for logical model inference.

Area of Science:

• Systems Biology
• Computational Biology
• Bioinformatics

Background:

• Model inference is crucial for systems biology, with Boolean networks (BNs) offering a computationally efficient approach for large biological networks.
• Current BN inference methods often struggle to explore the full range of possible models, even with extensive data and prior knowledge.

Purpose of the Study:

• To develop a novel formal instrument, Boolean network sketches, for comprehensive Boolean network inference.
• To integrate diverse forms of knowledge, including topology, update logic, transition properties, and experimental data, into the inference process.

Main Methods:

• Boolean network sketches integrate partial knowledge of network topology and update logic.
• An algorithm extends an initial sketch with data restrictions to a data-informed sketch.
• The method employs symbolic representation and colored model-checking to compute all consistent BNs.

Main Results:

• The proposed algorithm efficiently computes a compact representation of all inferred BNs consistent with integrated knowledge and data.
• The approach successfully handles a broad spectrum of biological knowledge and data restrictions.
• Evaluation on real-world and simulated data demonstrates the method's effectiveness.

Conclusions:

• Boolean network sketches provide a powerful and flexible framework for systems biology model inference.
• This approach enhances the discovery of admissible biological network models by exploring a wider solution space.