Inferring transcriptional regulatory networks from high-throughput data

Area of Science:

• Systems Biology
• Computational Biology
• Genomics

Background:

• Understanding gene regulation requires identifying transcription factor (TF) target relationships.
• TF activity is complex, influenced by post-translational modifications and cooperative binding, making direct measurement difficult.
• Standard methods struggle to capture the nuances of combinatorial control in gene regulation.

Purpose of the Study:

• To develop a novel computational method for inferring transcriptional regulatory networks (TRNs).
• To account for protein cooperation and combinatorial control in gene regulation.
• To accurately predict regulatory relationships using gene expression data.

Main Methods:

• Formulated TRN inference as a linear programming (LP) problem using gene expression data and estimated TF activities.
• Incorporated protein transcription complex information and mass action law.
• Integrated ChIP-Chip data and multiple gene expression datasets.

Main Results:

• The proposed method effectively infers TRNs by considering protein cooperation.
• Demonstrated the method's ability to predict regulatory relationships using synthetic and yeast experimental data.
• Achieved globally optimal solutions for TRN inference based on L(1) norm error.

Conclusions:

• The novel method accurately predicts transcription factor-target gene interactions, including co-regulators.
• This approach enhances the understanding of complex regulatory mechanisms in cellular systems.
• The method provides a robust framework for TRN inference from diverse biological data.