NAE

NAE quantifies regulatory network activity by measuring consistency between gene regulatory network architectures and time-series gene expression data.

Key Features:

• Dynamic Bayesian Network Model: Uses a dynamic Bayesian network to infer gene regulatory network structure from time-series profiling data.
• Regularized Constraint Programming: Applies an interpretable general loss function with regularization penalties that integrate prior knowledge about gene regulatory networks to compute consistency between network architecture and expression data.
• Optimization Algorithm (ADMM): Uses a fast, convergent alternating direction method of multipliers (ADMM) algorithm to optimize the regularized constraint programming objective.
• Statistical Significance Measurement: Quantifies network activity both as the degree of consistency between the regulatory network and gene expression data and as a statistical significance score.

Scientific Applications:

• Regulator–Target Interaction Mapping: Elucidates functional connections between regulators such as transcription factor proteins and miRNAs and their target genes.
• Regulatory Event Activity Evaluation: Evaluates activity of regulatory events within gene networks to assess how changes in molecular environments influence cellular responses.
• Gene Regulation and Cell State Dynamics: Supports research into gene regulation, cell state dynamics, and related biological processes by providing quantitative network activity measures.

Methodology:

Extracts molecular interactions from experimental data in knowledge-based repositories and aligns network architecture with gene expression profiles across specific states.