Raising awareness among plant virologists on the richness of their high-throughput sequencing data

High-throughput sequencing (HTS), more specifically metatranscriptomics (RNA-seq) of plant tissue, has become an indispensable tool for plant virologists to detect and identify plant viruses. In other plant pathology disciplines, HTS is not often used for this purpose since other techniques are more suitable. Plant virologists typically compare the obtained sequences to reference virus databases only, which lead to our hypothesis that we might be missing possible traces of other pathogens or pests in our data.
In this study, we set up a community effort to re-analyze existing RNA-seq datasets used for virus detection to check for the potential presence of non-viral pathogens or pests. The participants first mapped their sequence data against a given rRNA database (based on SILVA) and based on these mapping results (more specifically the estimated fraction of non-plant, non-viral rRNA), some datasets were selected to explore in more detail. Two analyses were conducted on these selected samples: 1) meta-assembly followed by best-hit taxonomic classification using diamond blastx against Uniprot and 2) direct taxonomic classification of the reads using a k-mer based tool (Kraken2) against the complete GenBank non-redundant Nucleotide BLAST database. The resulting taxonomic classifications and taxa assignments were checked for the presence of non-viral plant pests and pathogens and the plausibility of their presence was evaluated.
In total 101 RNA-seq datasets from 15 participants derived from 51 different plant species were re-analyzed, of which 37 were selected for subsequent in-depth analyses. In 26 of the 37 selected samples (70%), we found convincing traces of non-viral plant pathogens or pests. Plant pathogenic fungi were the most common organism group observed (16/37 datasets), while plant-pathogenic bacteria were much less noticed (3/37 datasets). Surprisingly, many traces were found from mites (9/37 datasets) and insect pests (13/37 datasets). Oomycetes were found in 6 datasets, while 1 dataset was clearly positive for a phytoplasma.
In conclusion, we were able to show that it is possible to detect non-viral pathogens or pests in these metatranscriptomics datasets, in this case primarily fungi, insects and mites. With this study, we hope to raise awareness among plant virologists that their data might be useful for fellow plant pathologists in other disciplines (bacteriology, mycology, entomology) as well.