Stable Gene Regulatory Network Modeling From Steady-State Data.

Area of Science:

• Systems Biology
• Computational Biology
• Genomics

Background:

• Gene regulatory networks (GRNs) map gene dependencies in cells.
• GRN structure is often unknown and requires reverse engineering from expression data.
• Microarray experiments measure gene expression levels, typically mRNA concentrations.

Purpose of the Study:

• To adapt the LASSO-VAR method for GRN recovery from noisy perturbation data.
• To develop a method for inferring sparse and stable regulatory networks.
• To apply the method to real biological systems without prior topological knowledge.

Main Methods:

• Adaptation of the least absolute shrinkage and selection operator-vector autoregressive (LASSO-VAR) model.
• Inclusion of a stability constraint for network recovery.
• Application to steady-state gene expression data from perturbation experiments.

Main Results:

• Successful recovery of GRNs from noisy experimental data.
• Demonstrated application to the SOS pathway in *Escherichia coli* and the cell cycle in *Saccharomyces cerevisiae*.
• The method infers networks without requiring prior topology information.

Conclusions:

• The adapted LASSO-VAR method effectively recovers sparse and stable gene regulatory networks.
• The approach is efficient for large-scale networks due to its convex nature.
• Enables GRN inference without prior biological network knowledge.