Inferring transcriptional logic from multiple dynamic experiments.

Area of Science:

• Systems Biology
• Computational Biology
• Genomics

Background:

• Dynamic gene expression data under multiple conditions is crucial for understanding gene regulation.
• Identifying transcriptional regulators and their logical interactions is a key goal in systems biology.

Purpose of the Study:

• To develop a novel, interpretable computational method for inferring gene transcriptional regulation.
• To identify the logical structure of gene regulation by transcription factors (TFs).

Main Methods:

• A dynamic model linking target gene mRNA transcription rates to TF activation states.
• Utilizing a trans-dimensional Markov Chain Monte Carlo (MCMC) algorithm for efficient sampling of regulatory logic.
• Validating the method with simulations and applying it to Arabidopsis thaliana microarray time series data.

Main Results:

• The proposed method successfully infers gene regulatory logic with biological interpretability.
• The dynamic model captures TF interactions consistent across multiple experiments and time.
• The MCMC algorithm efficiently ranks potential regulatory models by posterior probabilities.

Conclusions:

• The novel method accurately detects complex regulatory interactions consistent across diverse experimental conditions.
• This approach advances the understanding of gene regulatory networks using dynamic expression data.
• The developed tool is available for use in further systems biology research.