Single-sample landscape entropy reveals the imminent phase transition during disease progression

Area of Science:

• Systems Biology
• Computational Biology
• Genomics

Background:

• Biological systems exhibit abrupt state changes during disease progression, marked by tipping points.
• Predicting these disease transitions using omics data is challenging, especially with single-sample availability.

Purpose of the Study:

• To develop a novel method for identifying disease progression tipping points from single omics samples.
• To enable early disease prediction and characterization at an individual level.

Main Methods:

• Introduced the single-sample landscape entropy (SLE) method.
• Evaluated network disorder and criticality using network entropy from single-sample data.
• Identified dynamic network biomarkers associated with disease transitions.

Main Results:

• The SLE method successfully identified tipping points preceding severe disease symptoms in four real-world datasets.
• Validated applications include influenza virus infection, lung cancer metastasis, prostate cancer, and acute lung injury.
• Demonstrated the capability to characterize sample-specific disease states and predict individual disease outcomes.

Conclusions:

• The SLE method provides a robust approach for disease progression analysis using single omics samples.
• This method facilitates early detection and personalized disease state characterization.
• SLE offers a powerful tool for understanding dynamic biological changes during disease onset.