1. Methodology

1. Sampling effort:

1. We collected major agricultural pests (notably Helicoverpa armigera, Nezara viridula and Tuta absoluta) and their host-plants from agricultural ecosystems located in France, Italy and Spain between 2014 and 2021.
2. We processed ~1000 samples, of pooled or unpooled individuals, by viral metagenomics (i.e. VANA) coupled to an automated pipeline (i.e. NearVANA) for virus identification.

• VANA: Virion-associated nucleic acids metagenomic. The VANA protocol (available on HAL) allows high resolution screening of multiplexed arthropod and plant samples.

General workflow of VANA virome generation and analysis. This figure was produced by Sarah François, see this review paper.

• NearVANA: an automated pipeline for virus identification from metagenomic data. The NearVANA pipeline was developed by Aymeric Antoine-Lorquin (available on GitHub).

NearVANA Workflow. Raw reads can be paired or unpaired, multiplexed or not, resulting from VANA or RNAseq processing. This figure was produced by Aymeric Antoine-Lorquin.

2. Results

1. We revealed a high diversity of viruses in insect pests:
   1. We detected 54 families of viruses.
   2. We could classify some of those viruses into 236 virus species. We reconstructed the complete coding sequence of 58 novel species, including 17 insect viruses, 26 plant viruses and 12 bacteriophages.
2. We also showed disparities in insect viruses prevalence and abundance, which are potentially linked to their host range.

This figure was produced by Sarah François