Rxivist logo

AutoSpill: A method for calculating spillover coefficients to compensate or unmix high-parameter flow cytometry data

By Carlos P. Roca, Oliver T. Burton, Teresa Prezzemolo, Carly E. Whyte, Richard Halpert, Ɓukasz Kreft, James Collier, Alexander Botzki, Josef Spidlen, Stephanie Humblet-Baron, Adrian Liston

Posted 29 Jun 2020
bioRxiv DOI: 10.1101/2020.06.29.177196

Compensating in classical flow cytometry or unmixing in spectral systems is an unavoidable challenge in the data analysis of fluorescence-based flow cytometry. In both cases, spillover coefficients are estimated for each fluorophore using single-color controls. This approach has remained essentially unchanged since its inception, and is increasingly limited in its ability to deal with high-parameter flow cytometry. Here, we present AutoSpill, a novel approach for calculating spillover coefficients or spectral signatures of fluorophores. The approach combines automated gating of cells, calculation of an initial spillover matrix based on robust linear regression, and iterative refinement to reduce error. Moreover, autofluorescence can be compensated out, by processing it as an endogenous dye in an unstained control. AutoSpill uses single-color controls and is compatible with common flow cytometry software, but it differs in two key aspects from current methods: [(1)][1] it is much less demanding in the preparation of controls, as it does not require the presence of well-defined positive and negative populations, and [(2)][2] it does not require manual tuning of the spillover matrix, as the algorithm iteratively computes the tuning, producing an optimal compensation matrix. Another algorithm, AutoSpread, complements this approach, providing a robust estimate of the Spillover Spreading Matrix (SSM), while avoiding the need for well-defined positive and negative populations. Together, AutoSpill and AutoSpread provide a superior solution to the problem of fluorophore spillover, allowing simpler and more robust workflows in high-parameter flow cytometry. ### Competing Interest Statement The VIB and the Babraham Institute received funding from BD Bioscience in return for pre-publication access to and consultancy on the AutoSpill algorithm, in order to be incorporated into FlowJo v.10.7. RH and JS are affiliated with FlowJo, a wholly owned subsidiary of Becton, Dickinson and Company. The other authors declare no competing financial interests. [1]: #disp-formula-1 [2]: #disp-formula-2

Download data

  • Downloaded 3,538 times
  • Download rankings, all-time:
    • Site-wide: 3,039
    • In immunology: 81
  • Year to date:
    • Site-wide: 2,134
  • Since beginning of last month:
    • Site-wide: 2,134

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide


Sign up for the Rxivist weekly newsletter! (Click here for more details.)