Evaluation of a Viable-Cell Detection Assay for Xanthomonas fragariae with Latent Class Analysis

Abstract

Most molecular diagnostic assays are designed to detect the simple presence of a protein or nucleic acid without regard to whether that protein or nucleic acid originated from a viable and presumably pathogenic organism at the time of isolation. We recently developed a viable-cell detection assay (propidium monoazide [PMA]-qPCR) for specific detection of viable (living) cells of Xanthomonas fragariae, the pathogen causing angular leaf spot of strawberry, and herein describe a unique set of statistical analyses for validation of this assay. For any detection assay or test, calculation of its sensitivity and specificity is essential for determining its diagnostic capabilities, particularly when evaluating competing assays or tests. This is often achieved by running competing tests concurrently on a set of samples with known pathogen density or disease status and cross-tabulating results. However, the PMA-qPCR assay is unique among PCR-based diagnostics because it is designed specifically to detect DNA from living cells only, whereas most traditional PCR assays are designed to detect DNA from the target organism regardless of its state of viability. Thus, one cannot directly compare the results from a cell-viability assay with those from a general assay to gauge the performance of either assay because the assays target two different but overlapping populations (i.e., the viable-cell population and the viable- plus nonviable-cell population). To address this challenge, two standard statistical approaches to diagnostic test evaluation were used jointly to estimate the test performance of the PMA-qPCR assay relative to two common assays for detection of X. fragariae. In both analyses, the PMA-qPCR outperformed the qPCR for detection of viable cells under a range of conditions. Viability testing is extremely useful in certification and disease management applications, and with the information on test performance generated here, the test can be put to practical use to design sampling strategies to account for the errors in testing. [Formula: see text] The author(s) have dedicated the work to the public domain under the Creative Commons CC0 “No Rights Reserved” license by waiving all of his or her rights to the work worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law, 2023.