Metagenomic approaches have proven effective for the discovery of previously unknown virus species. While the usefulness of viral metagenomics for diagnostics and viral surveillance is still debated, these approaches have vastly extended our understanding of the ecological roles of viral communities, indicating that these are likely essential components of ecosystems as diverse as the human gut and the oceans. Almost a decade ago, a novel geo-referenced metagenomics approach was devised that could precisely link individual sequence reads to both the plant hosts from which they were obtained, and the spatial arrangements of these hosts1. Besides illuminating the diversity and the distribution of plant viruses at the ecosystem scale1, application of this spatial metagenomics approach has enabled the direct testing of hypotheses relating to the impacts of environmental conditions on plant virus diversity and prevalence. Specifically, a recent study2 has revealed that the detected proportion of virus-infected plant samples (a proxy for virus prevalence) was higher for cultivated plants than for uncultivated ones. Likewise, it was found that overall virus prevalence was generally higher in cultivated crop areas than in areas under natural vegetation. This indicates that agriculture substantially influences plant virus distributions and highlights the extent of current ignorance about the diversity and roles of viruses in nature. However, before the full potential of plant virus metagenomics can be achieved, several technical issues need to be resolved. The first is that, with current next generation sequencing technologies (e.g. Illumina), the full virus genomes and large viral sequence contigs that are routinely assembled during metagenomics studies, may in fact be chimeras of short reads derived from different viral genomes and, as such, may not actually exist. This problem undermines the effective analysis and taxonomic assignment of the sequences that are generated u