timeseriesNEM

timeseriesNEM infers signalling dynamics by integrating interventional (steady-state perturbation and RNA interference) and non-interventional time-series data to reconstruct hierarchical relationships within biological signalling networks.

Key Features:

• Integration of Data Types: Combines steady-state perturbation data with non-interventional time series, incorporating RNA interference perturbation data and RNA sequencing measurements.
• Nested Effects Model Mapping: Maps integrated datasets onto a static nested effects model to capture hierarchical and interdependent signalling relationships.
• Efficient Computational Approach: Formulates the inference problem as an integer linear programme and employs heuristic algorithms to solve it for large-scale datasets.

Scientific Applications:

• Epithelial–Mesenchymal Transition (EMT) Analysis: Applied to murine mammary gland cells by integrating RNA interference perturbation data with time-resolved non-interventional time series to infer transcription factor and microRNA signalling networks involved in EMT, with experimental validation by luciferase reporter assays and supporting RNA sequencing and microscopy data.
• Signal Progression Inference: Assesses signal progression over time to identify critical regulatory moments required for biological processes such as EMT to advance.

Methodology:

Integrate RNA sequencing data from perturbation experiments with non-interventional time series; map the integrated data onto a static nested effects model; formulate inference as an integer linear programme and solve it using heuristic algorithms.