Advertisement

Mass Cytometry pp 309-332 | Cite as

Supervised Machine Learning with CITRUS for Single Cell Biomarker Discovery

  • Hannah G. Polikowsky
  • Katherine A. Drake
Protocol
Part of the Methods in Molecular Biology book series (MIMB, volume 1989)

Abstract

CITRUS is a supervised machine learning algorithm designed to analyze single cell data, identify cell populations, and identify changes in the frequencies or functional marker expression patterns of those populations that are significantly associated with an outcome. The algorithm is a black box that includes steps to cluster cell populations, characterize these populations, and identify the significant characteristics. This chapter describes how to optimize the use of CITRUS by combining it with upstream and downstream data analysis and visualization tools.

Key words

CITRUS Biomarker discovery Supervised machine learning viSNE 

References

  1. 1.
    Kvistborg P, Gouttefangeas C, Aghaeepour N et al (2015) Thinking outside the gate: single-cell assessments in multiple dimensions. Immunity 42(4):591–592.  https://doi.org/10.1016/j.immuni.2015.04.006CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Newell EW, Cheng Y (2016) Mass cytometry: blessed with the curse of dimensionality. Nat Immunol 17(8):890–895.  https://doi.org/10.1038/ni.3485CrossRefPubMedGoogle Scholar
  3. 3.
    Bruggner RV, Bodenmiller B, Dill DL et al (2014) Automated identification of stratifying signatures in cellular subpopulations. Proc Natl Acad Sci U S A 111(26):E2770–E2777.  https://doi.org/10.1073/pnas.1408792111CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Fraietta JA, Lacey SF, Orlando EJ et al (2018) Determinants of response and resistance to CD19 chimeric antigen receptor (CAR) T cell therapy of chronic lymphocytic leukemia. Nat Med 24(5):563–571.  https://doi.org/10.1038/s41591-018-0010-1CrossRefPubMedPubMedCentralGoogle Scholar
  5. 5.
    Pelissier Vatter FA, Schapiro D, Chang H et al (2018) High-dimensional phenotyping identifies age-emergent cells in human mammary epithelia. Cell Rep 23(4):1205–1219.  https://doi.org/10.1016/j.celrep.2018.03.114CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Subrahmanyam PB, Dong Z, Gusenleitner D et al (2018) Distinct predictive biomarker candidates for response to anti-CTLA-4 and anti-PD-1 immunotherapy in melanoma patients. J Immunother Cancer 6(1):18.  https://doi.org/10.1186/s40425-018-0328-8CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Ben-Shaanan TL, Azulay-Debby H, Dubovik T et al (2016) Activation of the reward system boosts innate and adaptive immunity. Nat Med 22(8):940–944.  https://doi.org/10.1038/nm.4133CrossRefPubMedGoogle Scholar
  8. 8.
    Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc 57(1):289–300Google Scholar
  9. 9.
    van der Maaten LJP, Hinton GE (2008) Visualizing high-dimensional data using t-SNE. J Mach Learn Res 9:2579–2605Google Scholar
  10. 10.
    Amir el AD, Davis KL, Tadmor MD et al (2013) viSNE enables visualization of high dimensional single-cell data and reveals phenotypic heterogeneity of leukemia. Nat Biotechnol 31(6):545–552.  https://doi.org/10.1038/nbt.2594CrossRefGoogle Scholar
  11. 11.
    Knapp D, Kannan N, Pellacani D et al (2017) Mass cytometric analysis reveals viable activated caspase-3(+) luminal progenitors in the normal adult human mammary gland. Cell Rep 21(4):1116–1126.  https://doi.org/10.1016/j.celrep.2017.09.096CrossRefPubMedGoogle Scholar
  12. 12.
    Hahne F, Khodabakhshi AH, Bashashati A et al (2010) Per-channel basis normalization methods for flow cytometry data. Cytometry A 77(2):121–131.  https://doi.org/10.1002/cyto.a.20823CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Cytobank (2018) How to configure and run a viSNE analysis. https://support.cytobank.org/hc/en-us/articles/206439707-How-to-Configure-and-Run-a-viSNE-Analysis. Accessed 27 July 2018
  14. 14.
    Tusher VG, Tibshirani R, Chu G (2001) Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A 98(9):5116–5121.  https://doi.org/10.1073/pnas.091062498CrossRefPubMedPubMedCentralGoogle Scholar
  15. 15.
    Tibshirani R, Hastie T, Narasimhan B et al (2002) Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proc Natl Acad Sci U S A 99(10):6567–6572.  https://doi.org/10.1073/pnas.082099299CrossRefPubMedPubMedCentralGoogle Scholar
  16. 16.
    Tibshirani R (1996) Regression shrinkage and selection via the lasso. J R Stat Soc Ser B 58:267–288Google Scholar
  17. 17.
    Finak G, Jiang W, Krouse K et al (2014) High-throughput flow cytometry data normalization for clinical trials. Cytometry A 85(3):277–286.  https://doi.org/10.1002/cyto.a.22433CrossRefPubMedGoogle Scholar
  18. 18.
    Van Gassen S, Gaudiliere B, Dhaene T, et al (2017) A cross-sample cell-type specific normalization algorithm for clinical mass cytometry datasets. Paper presented at the 32nd congress of the International Society for Advancement of cytometry, Boston, MAGoogle Scholar
  19. 19.
    Hoy T (2006) Rare-event detection. In: Wulff S (ed) Guide to flow cytometry. Dako, Carpinteria, CA, pp 55–58Google Scholar
  20. 20.
    Baniyash M (2004) TCR zeta-chain downregulation: curtailing an excessive inflammatory immune response. Nat Rev Immunol 4(9):675–687.  https://doi.org/10.1038/nri1434CrossRefPubMedGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  • Hannah G. Polikowsky
    • 1
  • Katherine A. Drake
    • 1
  1. 1.Cytobank, IncSanta ClaraUSA

Personalised recommendations