wpLogicNet: logic gate and structure inference in gene regulatory networks

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Gene regulation functions as a logic system, where target genes are controlled by logic gates formed by their regulators.
• Existing computational methods for gene regulatory network (GRN) reconstruction have largely overlooked the analysis of these logical relationships.

Purpose of the Study:

• To introduce wpLogicNet, a novel computational tool designed to simultaneously infer directed GRN structures and the logic gates governing gene regulation.
• To address the limitations of current logic-based models by enabling the inference of logic gates involving multiple regulators.

Main Methods:

• wpLogicNet employs a Bayesian mixture model to estimate the probability of target gene expression profiles and infer logic gates.
• The tool offers a structure-aware mode for reconstructing logic gates within known TF-gene or gene-gene interaction networks.
• It features low time-complexity, distinguishing it from previous models limited to pairwise interactions.

Main Results:

• Validation on simulated and real-world datasets (E. coli, DREAM project) demonstrates wpLogicNet's precision in reconstructing both network structures and logical relationships.
• wpLogicNet shows superior performance compared to established methods in inferring directed GRNs and gene logic gates.
• The tool successfully infers logic gates involving multiple regulators, a capability lacking in many existing models.

Conclusions:

• wpLogicNet provides a powerful and precise framework for deciphering the logic of gene regulation and reconstructing complex GRNs.
• The tool's ability to infer multi-regulator logic gates advances the field of systems biology and network inference.
• The availability of the wpLogicNet R package and datasets facilitates further research in gene regulatory network analysis.