Introduction
Metatranscriptomics has emerged as a very useful technology for the study of microbiomes from RNA-seq reads. This method provides additional information compared to the sequencing of ribosomal genes such as the possible detection of viral sequences or the gene expression evaluation of both, host and associated microorganisms (dual-transcriptomics). Moreover, the targeted search for specific and active pathogens is a valuable strength.
Molluscs industry, and especially bivalves’ market is an economic sector of high potential in the Spanish coast, especially in Galicia. These species cope with highly variable environmental conditions in estuarine and intertidal regions. However, the current climate change scenario implies severe changes such as temperature increase, acidification, variations in water salinity or oxygen availability. This have a direct effect on the marine ecosystem and therefore on the species that are cultivated, which are fully adapted to specific conditions. Furthermore, there emerge additional risks such as the appearance of new pathogens, which may result in production problems.
In addition to the risk posed by the emergence of new diseases, there are a significant number of known bivalve pathogens that are already causing problems for cultures today. For instance, clams are susceptible to bacterial diseases (mainly vibriosis caused by V. splendidus, V. alginolyticus or Vibrio tapetis) and also multiple protozoan parasites such as Bonamia, Perkinsus, Haplosporidium or Marteilia, which invade tissues and have a negative impact on reproduction and growth. Another notable case is the high susceptibility of oysters to the OsHV-1 herpesvirus. Mollusc diseases have a common denominator, which is the absence of treatment, therefore the only possibility of control lies in prevention and monitoring of stocks.
Material and methods
Datasets available in public databases were used to perform a meta-analysis. Prior to the analysis, raw reads (RNA-seq reads) were trimmed to remove sequencing adapters and short and low-quality sequences (quality score limit 0.01 = PHRED 20). Moreover, datasets were randomly subsampled to equal the lowest number of reads to compare microbial abundance in a more appropriate manner. To analyze the different microbiome profiles, a working plan consisting of mapping the reads to several reference databases was implemented. The microorganism reference databases were built with all the virus, protozoan, and fungal genomes deposited in RefSeq. Specifically, these databases contained a total of 2,396 bacterial sequences, 10,907 viral sequences, 93 protozoan assemblies, and 388 fungal assemblies. Available genome assemblies were used to filter putative host-specific reads. Mapping parameters used to classify read packages in different taxonomic groups were as follows: length fraction = 0.5, similarity fraction = 0.8, and minimum seed length = 50. This minimum seed length parameter defines the minimum perfect match length for a position in the reference to be considered a valid candidate when matching the read.
Results
In this work, we used the metatranscriptomic approach to study a series of relevant mollusc pathogens in available transcriptomic data, to have a broad view of the prevalence of several pathogens in different geographical locations.
Moreover, we could also detect other associated microorganisms, being possible to stablish some relationships and interactions.