DisplayTitle

The need to screen increasing numbers of plant viruses in a given sample has resulted in the development of highly parallel methods of detection and identification such as DNA microarray and next generation sequencing. Based on our results with a prototype universal plant virus microarray (UPVM) here we present the development and validation of a full UPVM to detect and identify all known plant viruses, viroids and satellites. A total of 9600 60-mer oligonucleotide probes were selected from 302,230 candidates representing every taxon. After confirming the capacity of selected probes to identify and discriminate viral species by <i>in silico</i> analysis, probes were synthesised without any linker or modifications and spotted on poly-l-lysine coated glass slides. The UPVM was validated with non-amplified amino-allyl labelled cDNA from nine healthy plants and more than 30 plants infected with characterized viruses. Validation is in progress with more samples from ATCC and ‘unknown’ samples. High-titre viruses could easily be detected and identified. For low-titre virus samples several strategies including increasing total RNA for labelling, depleting the host rRNAs and random-amplification are being tested. Comparison of hybridization profiles of healthy and infected plant samples identified a few non-specific reactions. Microarray data was also analyzed by hierarchical clustering and a predictive algorithm, T-Predict, to identify one or more viruses in a mixed infection.<p><p>Keywords: