A bioinformatic ecological risk assessment framework for externally applied dsRNA-based biopesticides.

Double-stranded RNA (dsRNA)-based biopesticides are a promising new method of pest management. These biopesticides leverage the endogenous RNA interference (RNAi) pathway to selectively regulate expression of key genes involved in growth and development in pests, providing the potential to minimize harmful environmental effects by highly specific targeting. As dsRNA-based biopesticides are presented for regulatory review, evaluating potential off-target effects on Non-Target Organisms (NTOs) in a manner that may be unique to this novel sequence-specific mode of action (MoA) is crucial. To address this, we propose here a bioinformatics framework for consideration of sequence-specific off-target effects in NTOs. This framework includes the careful consideration of NTOs based on potential exposure and susceptibility and recommends standardizing analyses to search for 21 nucleotide (nt) long stretches of perfect identity and 80% overall identity between the dsRNA and off-target transcripts. We recommend a three-pronged approach to ensure a comprehensive risk assessment which includes 1) phylogenetic analysis of gene orthologs that defines the taxonomic scope of sequence similarity, 2) broad searches of large databases to identify potential unexpected similarity in distantly related species, and 3) full transcriptome analyses in NTO species of particular concern for a thorough understanding of all potential hazards. Finally, we recommend considering the results of bioinformatic analyses in the context of risk characterization, which means considering likely exposure to the dsRNA-based pesticide and potential susceptibility or barriers to dsRNA uptake. This approach enables a robust ecological risk assessment for dsRNA-based biopesticides and a regulatory path forward for this promising new pest management tool.