Detection of pre-transition phases during biological development using single-sample network entropy (SNE)

Area of Science:

• Computational Biology
• Systems Biology
• Biomedical Data Science

Background:

• Complex biological systems exhibit pre-transition phases before critical shifts.
• Identifying these critical states at the single-sample or single-cell level is crucial for timely interventions.
• Traditional statistical methods struggle with small sample sizes common in clinical settings.

Purpose of the Study:

• To develop a novel algorithm for identifying pre-transition phases at the individual sample level.
• To quantify sample-specific disturbances relative to reference data.
• To enable early detection of critical states in complex biological processes.

Main Methods:

• Introduction of single-sample network entropy (SNE), a novel algorithm for critical state analysis.
• Application of SNE to numerical simulations and diverse real-world datasets (influenza, single-cell, tumor data).
• Identification of optimistic SNE (O-SNE) and pessimistic SNE (P-SNE) biomarkers.

Main Results:

• SNE successfully identified pre-transition phases in simulations and eight real-world biological datasets.
• The SNE method demonstrated superior performance compared to existing single-sample approaches.
• Novel prognostic biomarkers (O-SNE and P-SNE) were identified, supported by functional signaling biomarkers.

Conclusions:

• The SNE algorithm provides a powerful computational approach for uncovering pre-transition phases and biomarkers at the single-sample/single-cell level.
• This method offers new insights for early personalized biological analysis, including disease diagnosis and prognosis.
• The findings facilitate timely interventions by revealing critical states in complex biological systems.