Distributions of autofluorescence after compensation: Be panglossian, fret not

In this technical note, I describe an in silico model of multiparameter fluorescence measurements that takes into account intrinsic cellular autofluorescence and stained dye fluorescence distributions, fluorescence spectrum spillovers, and photon counting statistics. Using this model, it is easy to manipulate spectral variables as well as error terms to understand the impact of each on the distributions of estimated cell-associated dyes (e.g., conjugated monoclonal antibodies in immunophenotyping). An application of this model was to understand the genesis of "unusual" autofluorescence distributions that occasionally happen in multi-color immunophenotyping. These unusual distributions show striking correlated patterns (diagonals) for graphs of certain pairs of parameters. Here I show that these arise from combinations of spillover-spreading from unviewed parameters. While disconcerting to researchers taught to look for diagonals in distributions as heralding improper compensation, these distributions are in fact appropriate. In general, one can ignore the characteristics of cell distributions within the same limits as background (e.g., as proscribed by a fluorescence-minus-one, or FMO, control), as there is essentially no information content in that region. Published 2016 International Society for Advancement of Cytometry. This article is a US Government work, and as such, is in the public domain in the USA.