Bayesian Binary Mixture Models as a Flexible Alternative to Cut-Off Analysis of ELISA Results, a Case Study of Seoul Orthohantavirus

Abstract

Serological assays, such as the enzyme-linked immunosorbent assay (ELISA), are popular tools for establishing the seroprevalence of various infectious diseases in humans and animals. In the ELISA, the optical density is measured and gives an indication of the antibody level. However, there is variability in optical density values for individuals that have been exposed to the pathogen of interest, as well as individuals that have not been exposed. In general, the distribution of values that can be expected for these two categories partly overlap. Often, a cut-off value is determined to decide which individuals should be considered seropositive or seronegative. However, the classical cut-off approach based on a putative threshold ignores heterogeneity in immune response in the population and is thus not the optimal solution for the analysis of serological data. A binary mixture model does include this heterogeneity, offers measures of uncertainty and the direct estimation of seroprevalence without the need for correction based on sensitivity and specificity. Furthermore, the probability of being seropositive can be estimated for individual samples, and both continuous and categorical covariates (risk-factors) can be included in the analysis. Using ELISA results from rats tested for the Seoul orthohantavirus, we compared the classical cut-off method with a binary mixture model set in a Bayesian framework. We show that it performs similarly or better than cut-off methods, by comparing with real-time quantitative polymerase chain reaction (RT-qPCR) results. We therefore recommend binary mixture models as an analysis tool over classical cut-off methods. An example code is included to facilitate the practical use of binary mixture models in everyday practice.

Retrospective serological and molecular survey of myxoma or antigenically related virus in the Iberian hare, Lepus granatensis

“…The cut-off was selected by finite mixture models, which are independent of a reference test or reference samples (Swart et al, 2021).…”

“…Particularly, we aimed to maximize the diagnostic specificity to be able to detect antibodies to MYXV, or an antigenically related virus, minimizing the proportion of false positives. The cut‐off was estimated by finite mixture models (Swart et al., 2021 ). Finite mixture models allow to characterize the distributions of the subgroups (seropositive and seronegative in this study) within bimodal datasets (Benaglia et al., 2009 ), thus being an alternative tool to estimate the cut‐off of serological tests in the absence of reference tests (Migchelsen et al., 2017 ; Peel et al., 2014 ; Swart et al., 2021 ).…”

“…The cut‐off was estimated by finite mixture models (Swart et al., 2021 ). Finite mixture models allow to characterize the distributions of the subgroups (seropositive and seronegative in this study) within bimodal datasets (Benaglia et al., 2009 ), thus being an alternative tool to estimate the cut‐off of serological tests in the absence of reference tests (Migchelsen et al., 2017 ; Peel et al., 2014 ; Swart et al., 2021 ). Finite mixture models were implemented using the expectation‐maximization algorithm for mixtures of univariate semiparametric distributions, from the package ‘mixtools’ (Benaglia et al., 2009 ) in R. Sera were also analyzed using the cut‐off of the test defined by the manufacture for the detection of anti‐MYXV rabbit IgGs (IR10 ≥ 2.0 positive; IR10 < 2.0 negative).…”

A Novel Vaccine Strategy to Prevent Cytauxzoonosis in Domestic Cats

“…Results were recorded as the average absorbance values. Positive cut-off values for each antibody type were set by calculating the value of the negative control + 3 standard deviations as previously described [ 43 ].…”

“…Upper dashed horizontal line in ( b ) indicates strong positive cutoff, while the lower dashed horizontal line indicates weak positive cutoff, as reported in [ 42 ]. Dashed horizontal lines in ( c , d ) indicate positive cut-off values established as previously described [ 43 ]. **** p < 0.0001.…”