Related Experiment Video
Updated: Oct 3, 2025
Morphology-Based Distinction Between Healthy and Pathological Cells Utilizing Fourier Transforms and Self-Organizing Maps
Published on: October 28, 2018
Unsupervised cluster analysis and subset characterization of abnormal erythropoiesis using the bioinformatic
Anna Porwit1,2, Despoina Violidaki1,2, Olof Axler1,2
1Department of Clinical Sciences, Oncology and Pathology, Lund University, Faculty of Medicine, Lund, Sweden.
FlowSOM, an AI tool for multiparameter flow cytometry (MFC) data, identified novel erythropoietic precursor subsets in myelodysplastic syndrome (MDS) patients. This AI approach reveals subtle erythropoiesis abnormalities not seen with traditional methods.
Area of Science:
- * Computational Biology
- * Immunology
- * Hematology
Background:
- * FlowSelf Organizing Maps (FlowSOM) is an AI program for unsupervised analysis of multiparameter flow cytometry (MFC) data.
- * Flow cytometry is a crucial technique for analyzing cellular populations.
Purpose of the Study:
- * To investigate the utility of FlowSOM for characterizing erythropoiesis abnormalities in patients with anemia and myelodysplastic syndrome (MDS).
- * To identify novel erythropoietic precursor (EP) subsets and their characteristics in MDS using AI-driven analysis.
Main Methods:
- * Analysis of MFC data from 16 patients (5 non-clonal anemia, 11 MDS) using the FlowSOM algorithm.
- * Comparison of EP subsets identified in patient samples against a reference of normal bone marrow samples.
- * Examination of antigen expression (CD36, CD71) and side scatter characteristics of identified EP subsets.
Main Results:
- * FlowSOM identified 18 novel EP subsets in MDS patients, including subtle alterations in CD36/CD71 expression and side scatter.
- * Three distinct patterns of EP abnormalities were observed in MDS patients.
- * Treatment with azacytidine and allogeneic stem-cell transplantation led to a decrease in abnormal EP subsets in one MDS patient.
Conclusions:
- * Unsupervised FlowSOM analysis of MFC data can detect subtle erythropoiesis alterations missed by conventional methods.
- * This AI approach provides new insights into the pathophysiology of MDS and related conditions.
- * FlowSOM represents a powerful tool for the interpretation of complex MFC datasets in hematological research.
More Related Videos
06:01Visualization and Quantification of High-Dimensional Cytometry Data using Cytofast and the Upstream Clustering Methods FlowSOM and Cytosplore
Published on: December 12, 2019
15:32Identification and Analysis of Mouse Erythroid Progenitors using the CD71/TER119 Flow-cytometric Assay
Published on: August 5, 2011