• DOI:10.48550/arXiv.2208.05745
• Corpus ID: 251492978

Interpretable cytometry cell-type annotation with flow-based deep generative models

@article{Blampey2022InterpretableCC,
  title={Interpretable cytometry cell-type annotation with flow-based deep generative models},
  author={Quentin Blampey and Nad{\`e}ge Bercovici and Charles-Antoine Dutertre and Isabelle Pic and Fabrice Andr'e and Joana Mourato Ribeiro and Paul-Henry Courn{\`e}de},
  journal={ArXiv},
  year={2022},
  volume={abs/2208.05745}
}

• Quentin Blampey, N. Bercovici, +4 authors P. Cournède
• Published 11 August 2022
• Biology, Computer Science
• ArXiv

Cytometry enables precise single-cell phenotyping within heterogeneous populations. These cell types are traditionally annotated via manual gating, but this method suffers from a lack of reproducibility and sensitivity to batch-effect. Also, the most recent cytometers — spectral flow or mass cytometers — create rich and high-dimensional data whose analysis via manual gating becomes challenging and time-consuming. To tackle these limitations, we introduce Scyan 1 , a Single-cell Cytometry Annotation Network that automatically annotates cell types using only prior expert knowledge about the cytometry panel. We demonstrate that Scyan significantly outperforms the related state-of-the-art models on multiple public datasets while being faster and interpretable. In ad-dition, Scyan overcomes several complementary tasks such as batch-effect removal, debarcoding, and population discovery. Overall, this model accelerates and eases cell population characterisation, quantification, and discovery in cytometry.