SLE

SLE quantifies landscape entropy from single-sample omics data to detect network criticality and predict tipping points in disease progression.

Key Features:

• Single-Sample Analysis: Operates on individual samples of omics data to assess network disorder without requiring longitudinal or cohort data.
• Network Entropy Evaluation: Projects omics measurements onto a network model and computes landscape entropy to capture network disorder and criticality.
• Dynamic Network Biomarkers Identification: Identifies dynamic network biomarkers associated with impending transitions (tipping points) in disease progression.
• Sample-Specific Disease Prediction: Characterizes sample-specific states to support personalized inference of risk or disease stage.

Scientific Applications:

• Influenza virus infection: Validated on influenza virus infection datasets, where it pinpointed tipping points preceding severe symptoms.
• Lung cancer metastasis: Applied to lung cancer metastasis datasets to identify critical transition points during metastatic progression.
• Prostate cancer: Validated on prostate cancer datasets to detect critical transition states in disease progression.
• Acute lung injury: Validated on acute lung injury datasets to identify early tipping points preceding severe injury.

Methodology:

SLE projects single-sample omics data onto a network model, constructs a network representation, and computes landscape entropy as a quantitative measure of disorder to assess proximity to tipping points.