Blijdorp & Biesbosch eDNA Acipenser sturio ONT metabarcoding & metagenomics

Molecular techniques involving environmental DNA (eDNA) are being increasingly used for aquatic species detection. Metabarcoding, one of the most widely adapted molecular techniques, suffers from primer bias; the uneven amplification of (target) species. The primer bias can be eliminated by omitting PCR, thereby sequencing all eDNA present in a water sample. This method, known as metagenomics, offers potential benefits for (relative) abundance estimates and epigenetic modifications. However, metagenomics is poorly understood for eukaryotic communities and environmental DNA. In this study, we compared fish species detection efficiency between multi-marker metabarcoding and metagenomics of two filter types (traditional 1.2 µm vs. low-cost coarse) within two environments (clean controlled setup vs. turbid field setting). In addition, we explore relative abundance and methylation patterns obtained from metagenomic data. Metabarcoding overestimated the number of species present in the controlled setup, while it likely underestimated the number of species present in the field setting. The low-cost filter recovered more species using metabarcoding in both environments, but did not improve species detection using metagenomics. The relative read counts of some species were severely overestimated using metabarcoding, likely due to primer bias. In all filters and locations, the vast majority of metagenomic sequences were identified as host species, Acipenser sturio. Three base modifications were called on the 18S region of A. sturio, but no significant differences were observed between filter types or locations. Our results indicate that although metabarcoding and metagenomics currently complement each other in terms of species detection, metagenomics could provide additional insights into relative abundance and life-history traits.