Open AccessA robust and interpretable end-to-end deep learning model for cytometry data
Author(s) -
Zicheng Hu,
Alice Tang,
Jaiveer Singh,
Sanchita Bhattacharya,
Atul J. Butte
Publication year - 2020
Publication title -
proceedings of the national academy of sciences of the united states of america
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 5.011
H-Index - 771
eISSN - 1091-6490
pISSN - 0027-8424
DOI - 10.1073/pnas.2003026117
Subject(s) - mass cytometry , cytometry , convolutional neural network , computer science , deep learning , flow cytometry , artificial intelligence , population , end to end principle , machine learning , computational biology , immunology , biology , medicine , biochemistry , environmental health , gene , phenotype
Significance Cytometry technologies are able to profile immune cells at single-cell resolution. They are widely used for both clinical diagnosis and biological research. We developed a deep learning model for analyzing cytometry data. We demonstrated that the deep learning model accurately diagnoses the latent cytomegalovirus (CMV) in healthy individuals. In addition, we developed a method for interpreting the deep learning model, allowing us to identify biomarkers associated with latent CMV infection. The deep learning model is widely applicable to other cytometry data related to human diseases.