adverSCarial: assessing the vulnerability of single-cell RNA-sequencing classifiers to adversarial attacks.

Journal: Bioinformatics (Oxford, England)
PMID:

Abstract

MOTIVATION: Several machine learning (ML) algorithms dedicated to the detection of healthy and diseased cell types from single-cell RNA sequencing (scRNA-seq) data have been proposed for biomedical purposes. This raises concerns about their vulnerability to adversarial attacks, exploiting threats causing malicious alterations of the classifiers' output with defective and well-crafted input.

Authors

  • Ghislain Fievet
    INSERM U1256, Nutrition, Genetics, and Environmental Risk Exposure (NGERE), University of Lorraine, Nancy, 54500, France.
  • Julien Broséus
    INSERM U1256, Nutrition, Genetics, and Environmental Risk Exposure (NGERE), University of Lorraine, Nancy, 54500, France.
  • David Meyre
    INSERM U1256, Nutrition, Genetics, and Environmental Risk Exposure (NGERE), University of Lorraine, Nancy, 54500, France.
  • Sébastien Hergalant
    INSERM U1256, Nutrition, Genetics, and Environmental Risk Exposure (NGERE), University of Lorraine, Nancy, 54500, France.