Estimating time since influenza virus exposure using single-cell proteomic data.

Area of Science:

• Immunology
• Infectious Diseases
• Computational Biology

Background:

• Accurate determination of respiratory infection onset is crucial for clinical management and understanding transmission dynamics.
• Current diagnostic tools lack biomarkers to indicate time since exposure, limiting infection timing insights.

Purpose of the Study:

• To develop immune-based predictive models for estimating infection timing and shedding status.
• To address the gap in diagnostics by providing information on time since exposure.

Main Methods:

• Utilized data from a controlled human influenza A/California/2009 (H1N1) challenge study.
• Longitudinally profiled immune cell subsets using 42-marker mass cytometry.
• Trained random forest machine learning models to predict shedders/non-shedders and days post-infection challenge (DPC).
• Validated models using an independent controlled human influenza challenge study.

Main Results:

• Single-cell immune population dynamics reveal a robust temporal structure post-influenza infection.
• Immune profiles accurately predict virus exposure timing.
• Models demonstrated reliable performance in distinguishing shedders and estimating DPC.

Conclusions:

• Host immune responses during influenza infection possess predictable temporal characteristics.
• Findings support the development of immune-based diagnostics for infection timing and contagiousness assessment.
• This represents a step towards advanced diagnostics beyond simple pathogen detection.