EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GateMeClass: Gate Mining and Classification of cytometry data

Simone Caligola iD, Luca Giacobazzi, Stefania Canè, Antonio Vella, Annalisa Adamo, Stefano Ugel, Rosalba Giugno iD, Vincenzo Bronte

Bioinformatics · 2024

DOI: 10.1093/bioinformatics/btae322 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics, 2024. Verfasst von Simone Caligola und 7 weiteren Autor:innen. Beteiligte Einrichtung: Veneto Institute of Oncology IOV-IRCCS, Padova, Italy. Hauptthemen laut MEDLINE: Flow Cytometry, Data Mining. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Associazione Italiana per la Ricerca sul Cancro, NHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Cytometry comprises powerful techniques for analyzing the cell heterogeneity of a biological sample by examining the expression of protein markers. These technologies impact especially the field of oncoimmunology, where cell identification is essential to analyze the tumor microenvironment. Several classification tools have been developed for the annotation of cytometry datasets, which include supervised tools that require a training set as a reference (i.e. reference-based) and semisupervised tools based on the manual definition of a marker table. The latter is closer to the traditional annotation of cytometry data based on manual gating. However, they require the manual definition of a marker table that cannot be extracted automatically in a reference-based fashion. Therefore, we are lacking methods that allow both classification approaches while maintaining the high biological interpretability given by the marker table. Results We present a new tool called GateMeClass (Gate Mining and Classification) which overcomes the limitation of the current methods of classification of cytometry data allowing both semisupervised and supervised annotation based on a marker table that can be defined manually or extracted from an external annotated dataset. We measured the accuracy of GateMeClass for annotating three well-established benchmark mass cytometry datasets and one flow cytometry dataset. The performance of GateMeClass is comparable to reference-based methods and marker table-based techniques, offering greater flexibility and rapid execution times. Availability and implementation GateMeClass is implemented in R language and is publicly available at https://github.com/simo1c/GateMeClass

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Flow CytometryData MiningHumansSoftwareAlgorithmsTumor Microenvironment

Autor:innen und Institutionen

  • Veneto Institute of Oncology IOV-IRCCS, Padova, Italy.
  • Section of Immunology, Department of Medicine, University of Verona, Verona, Italy.
  • Section of Immunology, Azienda Ospedaliera Universitaria Integrata (AOUI), Verona, Italy.
  • Department of Computer Science, University of Verona, Verona, Italy.

ORCID-Profile: Simone Caligola, Rosalba Giugno

Förderung

  • Associazione Italiana per la Ricerca sul Cancro
  • NHS

Zitieren

APA

Caligola, S., Giacobazzi, L., Canè, S., Vella, A., Adamo, A., Ugel, S., Giugno, R., & Bronte, V. (2024). GateMeClass: Gate Mining and Classification of cytometry data. Bioinformatics. https://doi.org/10.1093/bioinformatics/btae322

Vancouver

Caligola S, Giacobazzi L, Canè S, Vella A, Adamo A, Ugel S, et al. GateMeClass: Gate Mining and Classification of cytometry data. Bioinformatics. 2024. doi:10.1093/bioinformatics/btae322

BibTeX
@article{Caligola2024,
  author = {Caligola, Simone and Giacobazzi, Luca and Canè, Stefania and Vella, Antonio and Adamo, Annalisa and Ugel, Stefano and Giugno, Rosalba and Bronte, Vincenzo},
  title = {GateMeClass: Gate Mining and Classification of cytometry data},
  journal = {Bioinformatics},
  year = {2024},
  doi = {10.1093/bioinformatics/btae322}
}
Kostenlos starten →