ABSTRACTS – ORAL PRESENTATIONS - AMCA, spol. s r.o.

30. AN AUTOMATED ANALYSIS OF HIGHLY COMPLEX FLOW CYTOMETRY-BASED
PROTEOMIC DATA
Jan Stuchlý 1,2 , Veronika Kanderová 1,2 , Karel Fišer 1,2 , Daniela Černá 1,2 , Anders Holm 4 ,
WeiWei Wu 4 , Ondřej Hrušák 1,2 , Fridtjof Lund-Johansen 4 , Tomáš Kalina 1,2
1
Department of Pediatric Hematology and Oncology, 2 nd Faculty of Medicine, Charles
University Prague and University Hospital Motol, Prague, Czech Republic
2
CLIP - Childhood Leukemia Investigation Prague
3
Department of Numerical Mathematics, Faculty of Mathematics and Physics, Charles
University, Prague, Czech Republic
4
Department of Immunology, Rikshospitalet Medical Center and the University of Oslo,
Oslo, Norway
Correspondence to: jan.stuchly@lfmotol.cuni.cz
The combination of color-coded microspheres as carriers and flow cytometry as a
detection platform provides new opportunities for multiplexed measurement of biomolecules.
Here, we developed a software tool capable of automated gating of colorcoded
microspheres, automatic extraction of statistics from all subsets and validation,
normalization, and cross-sample analysis. The approach presented in this article
enabled us to harness the power of high-content cellular proteomics. In size exclusion
chromatography-resolved microsphere-based affinity proteomics (Size - MAP), antibodycoupled
microspheres are used to measure biotinylated proteins that have been
separated by size exclusion chromatography. The captured proteins are labeled with
streptavidin phycoerythrin and detected by multicolor flow cytometry. When the results
from multiple size exclusion chromatography fractions are combined, binding is detected
as discrete reactivity peaks (entities). The information obtained might be approximated
to a multiplexed western blot. We used a microsphere set with >1000 subsets, presenting
an approach to extract biologically relevant information. The R-project environment
was used to sequentially recognize subsets in two-dimensional space and gate them.
The aim was to extract the median streptavidin phycoerythrin fluorescence intensity
for all 1000+ microsphere subsets from a series of 96 measured samples. The resulting
text files were subjected to algorithms that identified entities across the 24 fractions.
Thus, the original 24 data points for each antibody were compressed to 1–4 integrated
values representing the areas of individual antibody reactivity peaks. Finally, we provide
experimental data on cellular protein changes induced by treatment of leukemia cells
with imatinib mesylate. The approach presented here exemplifies how large-scale flow
cytometry data analysis can be efficiently processed to employ flow cytometry as a highcontent
proteomics method.
Acknowledgements
This work was supported by IGA NT/13462 and GAUK 596912
Analytical Cytometry VII 55