A viral metagenomics strategy on a non-metagenomics experiment: mining next generation sequencing datasets from pig DNA pools.

In this study, we mined Ion Torrent reads obtained by sequencing DNA isolated from archived blood samples collected from 100 performance tested Italian Large White pigs. Two reduced representation libraries were prepared from two DNA pools constructed each from 50 equimolar DNA samples. A specific bioinformatics pipeline was designed to mine unmapped reads on the reference pig genome that were obtained from the two NGS datasets. Our approach identified sequences matching several viruses of the Parvoviridae family. This study demonstrated the potential of mining NGS datasets non originally derived by metagenomics experiments for viral metagenomics analyses in a livestock species.