Per-Event Uncertainty Quantification for Flow Cytometry Using Calibration Beads

Area of Science:

• Biomedical Engineering
• Analytical Chemistry
• Quantitative Biology

Background:

• Flow cytometry is crucial for diagnostics but faces challenges in distinguishing signals from noise due to measurement uncertainty.
• Existing models often fail to account for both population variability and instrument effects, hindering accurate analysis, especially for small particles like extracellular vesicles.

Purpose of the Study:

• To develop an explicit probabilistic measurement model for flow cytometry.
• To accurately separate sources of uncertainty, including background and instrument-induced effects.
• To improve the identification of signals from small biological entities.

Main Methods:

• Formulated a probabilistic model incorporating volume and labeling variation, background signals, and fluorescence shot noise.
• Utilized raw data from per-event calibration measurements.
• Applied the model to separate distinct sources of measurement uncertainty.

Main Results:

• Successfully separated sources of uncertainty in flow cytometry measurements.
• Demonstrated the model's capability to account for inherent population variability and instrument effects.
• Provided a framework for improved decision-making and instrument characterization.

Conclusions:

• The developed probabilistic model offers a more accurate approach to analyzing flow cytometry data.
• This method enhances the ability to detect and characterize small particles, such as extracellular vesicles.
• Improved understanding of measurement uncertainty facilitates more reliable diagnostics and instrument performance evaluation.