Elucidation of Genetic Interactions in the Yeast GATA-Factor Network Using Bayesian Model Selection

Area of Science:

• Systems Biology
• Computational Biology
• Molecular Genetics

Background:

• Classical genetic assays for gene regulatory networks are often error-prone and yield ambiguous results.
• Even well-characterized networks may contain unproven interactions, hindering a complete understanding.
• Mechanistic mathematical modeling offers a powerful approach to gain detailed insights from experimental data.

Purpose of the Study:

• To apply mechanistic mathematical modeling to the GATA-factor transcriptional network in Saccharomyces cerevisiae.
• To resolve ambiguities in the network's organization and identify the most plausible interactions.
• To enhance understanding of yeast transcriptional responses to dynamic nitrogen source availability.

Main Methods:

• Development of 32 mechanistic mathematical models representing hypothesized interactions within the GATA network.
• Utilized Bayesian model selection to identify the most plausible network structure based on dynamic gene expression data.
• Validated the top-ranking model using newly generated GFP reporter dynamic data.

Main Results:

• Identified a top-ranking mechanistic model that accurately represents the yeast GATA transcriptional network.
• The validated model provides a clearer understanding of network organization and its role in nitrogen metabolism.
• Demonstrated the utility of combining Monte Carlo techniques for large-scale dynamical system inference.

Conclusions:

• Mechanistic modeling, coupled with Bayesian selection, effectively resolves ambiguities in complex gene regulatory networks.
• This study provides a crucial step towards a holistic view of yeast nitrogen regulation.
• The findings highlight the power of computational approaches in deciphering intricate biological systems.