Metatranscriptomics unmasks Mollusca virome with a remarkable presence of rhabdovirus in cephalopods

Molluscs play a significant role in marine ecosystems and have economic value for aquaculture. Unexpected and unexplained high mortalities among molluscs have occasionally been described. The role of potential pathogens such as viruses remains unknown due to the lack of molluscan cell cultures, which is one of the major drawbacks to determining the viral role in such mortalities. Several oceanographic studies have suggested a high abundance of viruses in the oceans. Virus identification and understanding of viral interaction with organisms in marine ecosystems are in their infancy. Metatranscriptomics could become a useful tool to identify viruses using a shotgun approach and the growing number of viral genomes and sequences deposited in public databases. In this work, several bioinformatics approaches were set up to screen Mollusca RNA sequences to find and confirm viral traces in their transcriptomes. An extensive search of SRA datasets available in the NCBI database belonging to molluscs was conducted, selecting a total of 55 SRA datasets that were further analysed searching for viral sequences. Specifically, 22 bivalves, 19 cephalopods and 16 gastropods from 16 geographical origins and 17 different tissues were considered. The domain search approach was the most productive method to find viral sequences. This virus search showed that Cephalopoda samples (Idiosepius notoides and Amphioctopus fangsiao) exhibited the highest number of virus identifications. Some of the detected viral sequences were similar or identical to others previously identified. However, 33 putative new viruses were identified and analysed phylogenetically when the RdRp domain was available. Specifically, Cephalopoda samples showed a considerable number of viruses belonging to the Rhabdoviridae family.