Evaluation of sensitivity and specificity in RNA-Seq-based detection of grapevine viral pathogens

J Virol Methods. 2021 Nov 26:114383. doi: 10.1016/j.jviromet.2021.114383. Online ahead of print.ABSTRACTVirus detection is a crucial step for the implementation of clean stock programs that preserve healthy crop species. Viral infections in grapevine, a vegetatively propagated perennial crop, cannot be eradicated from the vineyards by the application of agrochemicals and must be curtailed at the stage of nursery production during the propagation of planting material. Viral detection is routinely performed using enzyme-linked immunosorbent assays (ELISA) or Reverse Transcription-quantitative Polymerase Chain Reactions (RT-qPCR). High throughput sequencing (HTS) approaches have the potential to detect all viral pathogens in a plant specimen. However, to date, no published HTS-based study has used threshold selection based on ROC curves for discriminating positive from negative samples. To fill this gap, we assessed the specificity and sensitivity of different sequencing and bioinformatics approaches for nine common viruses, which were tested in the same specimens using ELISA and/or RT-qPCR. The normalized detection thresholds giving the best results were 19.28 Fragments Per Kilobase of transcript per Million mapped reads (FPKM) for alignment-based total RNA-Seq approaches, 386 Reads Per Million mapped reads (RPM) for metagenomics-based total RNA-Seq, 1,572 FPKM for alignment-based small RNA-Seq analysis and 0.97% of contigs for de novo analysis of small RNA-Seq data. Validation...

Source: Journal of Virological Methods - November 29, 2021 Category: Virology Authors: Gabriele Di Gaspero Slobodanka Radovic Elisa De Luca Alessandro Spadotto Gabriele Magris Luigi Falginella Federica Cattonaro Fabio Marroni Source Type: research