Figure 1.

Automated cytometry data analysis workflow. Pipeline of the automated analysis steps of cytometric data. Starting from Flow Cytometry Standard (FCS) files, data are pre-processed to ensure reproducible and reliable results. Pre-processing phases include compensation (spectral overlap correction), data transformation (improvement of cell population visualization and automated cell types identification), data cleaning (removal of dead cells, debris, doublets, etc.), and normalizations (removal of batch effect between samples or balancing the contribution of each marker to the analysis). Pre-processed samples are analyzed with automated tools here classified as “Automated sequential gates”, “Boolean combinations gates”, and “Multivariate analysis” which include “Clustering algorithms”, “Dimensionality reduction methods”, and “Trajectory inference” techniques. Finally, statistical tests, correlation analysis and supervised machine learning techniques, such as regression and classification, can be applied to detect differences between experimental groups or to discover biomarkers.