Detecting robust time-delayed regulation in Mycobacterium tuberculosis

Area of Science:

• Systems Biology
• Molecular Biology
• Genomics

Background:

• Gene regulatory networks often exhibit time delays, which are challenging for conventional network inference techniques.
• The DNA repair system in Mycobacterium tuberculosis displays unusually slow kinetics, making it a relevant model for studying delayed gene regulation.

Purpose of the Study:

• To develop and apply a novel computational method for inferring gene regulatory networks that explicitly accounts for time delays.
• To investigate the delayed regulatory mechanisms and robustness of the DNA damage response in Mycobacterium tuberculosis.

Main Methods:

• A skip-chain model utilizing Viterbi paths of Markov chains to identify time-delayed gene interactions.
• Application of the model to time-course gene expression data following DNA damage induced by Mitomycin C.
• Incorporation of Dirichlet and Gibbs priors to mitigate over-fitting in network generation.

Main Results:

• Identification of several time-delayed regulatory interactions within the Mycobacterium tuberculosis DNA repair system.
• Observation of network hubs, indicating that a few transcriptional factors control a significant portion of the system.
• Comparison of the derived gene networks with those obtained using dynamic Bayesian networks (DBNs).

Conclusions:

• Skip-chain models are effective in capturing long-distance and time-delayed gene regulations, crucial for understanding biological pathways.
• Different transcriptional networks are dynamically active at various stages of biological processes, maintaining constant feedback.
• The use of priors significantly enhances the robustness and accuracy of network inference by reducing over-fitting.