Comparison of frequency distributions in flow cytometry.

Area of Science:

• Biostatistics
• Flow Cytometry
• Data Analysis

Background:

• Statistical comparison of flow cytometric frequency distributions is crucial.
• The Kolmogorov-Smirnov (K-S) test is a common method but often criticized for excessive sensitivity.
• Existing methods may not be optimal for discrete flow cytometry data.

Purpose of the Study:

• To explore alternative statistical methods for comparing flow cytometric frequency distributions.
• To propose methods based on the Poisson distribution as a more appropriate alternative to the K-S test.
• To provide a framework for understanding data variability in statistical analysis.

Main Methods:

• Utilized Poisson distribution assumptions for statistical analysis.
• Developed channel-by-channel confidence intervals and chi-square tests.
• Presented graphical displays of statistical techniques.

Main Results:

• Poisson-based methods are simpler and more appropriate for discrete flow cytometry data compared to the K-S test.
• Channel-by-channel confidence intervals and chi-square tests are effective.
• Graphical displays aid in understanding and comparing distributions.

Conclusions:

• A proper understanding of data variability and an appropriate probability model are essential for sound statistical analysis.
• Poisson distribution-based methods offer a robust alternative for flow cytometry data comparison.
• The proposed techniques provide a more suitable approach for discrete flow cytometry data.