dcanr

dcanr implements methods and an evaluation framework to infer differential co-expression/association networks.

Key Features:

• Differential network inference: Implements various methods for inferring differential co-expression/association networks altered between two conditions.
• Simulation-based evaluation: Provides an evaluation framework using simulated datasets to benchmark and characterise inference method performance.
• Dynamical systems simulator: Includes a simulator based on dynamical systems modelling capable of simulating differential co-expression patterns.
• Performance benchmarking: Enables comparison of inference methods across three different levels of "true" networks for each simulation.

Scientific Applications:

• Condition-specific network identification: Identifies networks that are altered between two conditions such as health and disease states.
• Regulatory network elucidation: Supports identification of regulatory mechanisms specific to a given biological context in systems biology.
• Transcriptional data analysis: Enables differential co-expression-based detection of conditional relationships in transcriptional data, with demonstrated application to breast cancer.
• Causal network reconstruction: Supports interpretation of inferred differential networks to reconstruct putative causal networks based on differentially regulated targets.

Methodology:

The package implements a z-score-based method and other inference approaches, with evaluation using a dynamical systems simulator that generates simulated datasets with defined network and simulation properties.