Constructing and analyzing a large-scale gene-to-gene regulatory network--lasso-constrained inference and biological...

Area of Science:

• Systems Biology
• Genomics
• Computational Biology

Background:

• Gene regulatory networks (GRNs) are crucial for understanding cellular processes.
• Inferring GRNs from time-series expression data is challenging, especially with limited measurements.
• Previous linear modeling approaches for GRN inference have faced validation challenges.

Purpose of the Study:

• To construct a gene-to-gene regulatory network for Saccharomyces cerevisiae using time-series gene expression data.
• To validate the constructed network against existing biological knowledge from the Gene Ontology database.
• To assess the adequacy of linear modeling for inferring large-scale GRNs with limited data.

Main Methods:

• Construction of a gene-to-gene regulatory network using time-series whole-genome expression data from Saccharomyces cerevisiae.
• Analysis of the network's large-scale properties against established biological knowledge in the Gene Ontology database.
• Application and validation of a linear modeling approach for network inference.

Main Results:

• A gene-to-gene regulatory network for yeast was successfully constructed.
• The inferred network's large-scale properties align with known biological facts about Saccharomyces cerevisiae.
• The study provides validation for the linear modeling approach in inferring biological networks.

Conclusions:

• Linear modeling is an adequate approach for constructing gene regulatory networks from limited time-series expression data.
• The validated network and methodology support further investigations into yeast gene regulation.
• This study demonstrates the utility of computational approaches for understanding complex biological systems.