Network Analysis for Uncovering the Relationship between Host Response and Clinical Factors to Virus Pathogen:...

Area of Science:

• Immunology
• Data Science
• Infectious Diseases

Background:

• Analyzing complex patient datasets, especially in viral infections, is challenging due to heterogeneous data types.
• Univariate comparisons are insufficient for understanding multifactorial patient characteristics and host responses.

Purpose of the Study:

• To demonstrate the utility of multivariate network approaches for analyzing complex patient data.
• To identify key host immune and clinical factors associated with the severity of SARS-CoV-2 infection.

Main Methods:

• Employed a patient similarity network (PSN) approach.
• Analyzed approximately 85 immunological and 70 clinical factors in 250 patients with COVID-19.

Main Results:

• Identified a minimal immune signature associated with disease severity.
• Uncovered clinical and laboratory factors strongly correlated with COVID-19 severity.
• Successfully visualized relationships between host factors and clinical outcomes.

Conclusions:

• Multivariate network approaches, like PSN, are beneficial for identifying relevant factors and visualizing complex relationships in infectious disease datasets.
• The PSN model is broadly applicable to various complex datasets beyond viral infections.