Profiling the baseline performance and limits of machine learning models for adaptive immune receptor repertoire classification.

Journal: GigaScience
PMID: 35639633

Abstract

BACKGROUND: Machine learning (ML) methodology development for the classification of immune states in adaptive immune receptor repertoires (AIRRs) has seen a recent surge of interest. However, so far, there does not exist a systematic evaluation of scenarios where classical ML methods (such as penalized logistic regression) already perform adequately for AIRR classification. This hinders investigative reorientation to those scenarios where method development of more sophisticated ML approaches may be required.

Authors

  • Chakravarthi Kanduri
    Centre for Bioinformatics, Department of Informatics, University of Oslo, Oslo 0373, Norway.
  • Milena Pavlovic
    UiO: RealArt Convergence Environment, University of Oslo, Oslo, Norway.
  • Lonneke Scheffer
    Department of Informatics, University of Oslo, Oslo, Norway.
  • Keshav Motwani
    Department of Pathology, Immunology and Laboratory Medicine, University of Florida, FL 32610, USA.
  • Maria Chernigovskaya
    Department of Immunology, Oslo University Hospital Rikshospitalet and University of Oslo, Norway.
  • Victor Greiff
    Department of Immunology, Oslo University Hospital, Oslo, Norway.
  • Geir K Sandve
    Centre for Bioinformatics, Department of Informatics, University of Oslo, Oslo 0373, Norway.

Keywords

External Resources

Popular Topics
Recent Journals