A step towards the validation of bacteria biotic indices using DNA metabarcoding for benthic monitoring

Environmental genomics is a promising field for monitoring biodiversity in a timely fashion. Efforts have increasingly been dedicated to the use of bacteria DNA derived data to develop biotic indices for benthic monitoring. However, a substantial debate exists about whether bacteria-derived data using DNA metabarcoding should follow, for example, a taxonomy-based or a taxonomy-free approach to marine bioassessments. Here, we showcase the value of DNA-based monitoring using the impact of fish farming as an example of anthropogenic disturbances in coastal areas and compare the performance of taxonomy-based and taxonomy-free approaches in detecting environmental alterations. We analysed samples collected near to the farm cages and along distance gradients from two aquaculture installations, and at control sites, to evaluate the effect of this activity on bacterial assemblages. Using the putative response of bacterial taxa to stress we calculated the taxonomy-based biotic index microgAMBI. The distribution of individual amplicon sequence variants (ASVs), as a function of a gradient in sediment acid volatile sulphides, was then used to derive a taxonomy-free bacterial biotic index specific for this data set using a de novo approach based on quantile regression splines. Our results show that microgAMBI revealed a organically enriched environment along the gradient. However, the de novo biotic index outperformed microgAMBI by providing a higher discriminatory power in detecting changes in abiotic factors directly related to fish production, whilst allowing the identification of new ASVs bioindicators. The de novo strategy applied here represents a robust method to define new bioindicators in regions or habitats where no previous information about the response of bacteria to environmental stressors exists.

Automated summary

The study demonstrated the value of DNA-based monitoring in assessing the impact of anthropogenic disturbances on bacterial communities, comparing the performance of taxonomy-based and taxonomy-free approaches in detecting environmental alterations.