ATEN: And/Or tree ensemble for inferring accurate Boolean network topology and dynamics

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Gene regulatory networks (GRNs) are crucial for understanding cellular processes.
• Inferring GRNs from gene expression time-series data is vital for biological insights.
• Boolean networks are a common approach for GRN inference, but accuracy is challenged by short and noisy data.

Purpose of the Study:

• To develop an accurate Boolean network inference algorithm for short and noisy gene expression time-series data.
• To improve the inference of both topology and dynamics of gene regulatory networks.
• To provide a tool for gaining deeper insights into complex regulatory mechanisms.

Main Methods:

• Proposed a novel Boolean network inference algorithm.
• Utilized an And/Or tree ensemble to select prime implicants for feature selection.
• Inferred Boolean functions for target genes using selected features.
• Combined individual gene Boolean functions to form the complete Boolean network.

Main Results:

• The algorithm accurately infers Boolean network topology and dynamics from short and noisy time-series data.
• Demonstrated superior performance compared to existing algorithms on both artificial and real-world gene regulatory network data.
• The proposed method enhances the ability to gain insights into complex regulatory mechanisms.

Conclusions:

• The developed algorithm effectively addresses the challenge of inferring accurate Boolean networks from limited and noisy experimental data.
• This approach offers a significant advancement in understanding gene regulatory mechanisms.
• The ATEN package is available for broader research application.