Testing eDNA sampling approaches in tropical waterways

Brief description

In this study, we optimised methodology for a citizen science project targeting fish taxa, ensuring simple, safe, and effective eDNA sampling. We compared species richness and composition of fish eDNA data obtained from a single type of passive sampler with that obtained from Longmire's preserved water samples at three waterways in north Queensland, Australia.
Lineage: DNA samples were amplified in triplicate using the Mitochondrial 16S (m16S) universal fish primer described in Berry et al. (2017) and metazoan universal primers Mitochondrial COI described in (Leray et al., 2013). Three no-template-control (NTC) samples and three positive DNA reference samples were included as amplification controls for both assays. Amplification success was interrogated through qPCR amplification plots and the NTC samples did not amplify. Final libraries were sent to the Australian Genome Research Facility (AGRF) for Illumina paired-end sequencing. Sequenced data were processed using the Greenfield Hybrid Amplicon Pipeline (GHAP) (Greenfield, 2017), which applies USEARCH tools V11 (Edgar, 2010, 2013). All data across the dataset were handled consistently. Paired reads were merged, trimmed, de-replicated, and clustered at 97% similarity, then further denoised (removing the sequencing tail and OTUs that were present in less than 3 samples) to generate a set of representative Operational Taxonomic Units (OTUs) sequences. in addition to clustering zero radius OTUs (zotus) were also obtained using the GHAP. The classified, denoised OTUs and the read counts for each sample were used to generate OTU and zotu tables.
The acceptable pairwise identity match to assign OTUs from the COI libraries to the taxonomic level was: Species: >98%; Genus: < 98% >95%; Family: 85% to 95%; and Order: 85-80%. This acceptable level cut-off was based on literature and consistent with the work from Leray et al. (2013). The acceptable pairwise identity match to assign OTUs from the m16S libraries to taxonomic level was: Species: >98%; Genus: >94.5%; and Family: >86.5% and was applied based on acceptable criteria applied by Wang et al. (2023) and Valsecchi et al. (2020). The OTUs were aggregated by taxon based on pairwise matches and matching accession-number taxonomic assignments.