The Tree of Life eDNA metabarcoding reveals a similar taxonomic richness but dissimilar evolutionary lineages between seaports and marine reserves.

Coastal areas host a major part of marine biodiversity but are seriously threatened by ever-increasing human pressures. Transforming natural coastlines into urban seascapes through habitat artificialization may result in loss of biodiversity and key ecosystem functions. Yet, the extent to which seaports differ from nearby natural habitats and marine reserves across the whole Tree of Life is still unknown. This study aimed to assess the level of α and ß-diversity between seaports and reserves, and whether these biodiversity patterns are conserved across taxa and evolutionary lineages. For that, we used environmental DNA (eDNA) metabarcoding to survey six seaports on the French Mediterranean coast and four strictly no-take marine reserves nearby. By targeting four different groups-prokaryotes, eukaryotes, metazoans and fish-with appropriate markers, we provide a holistic view of biodiversity on contrasted habitats. In the absence of comprehensive reference databases, we used bioinformatic pipelines to gather similar sequences into molecular operational taxonomic units (MOTUs). In contrast to our expectations, we obtained no difference in MOTU richness (α-diversity) between habitats except for prokaryotes and threatened fishes with higher diversity in reserves than in seaports. However, we observed a marked dissimilarity (ß-diversity) between seaports and reserves for all taxa. Surprisingly, this biodiversity signature of seaports was preserved across the Tree of Life, up to the order. This result reveals that seaports and nearby marine reserves share few taxa and evolutionary lineages along urbanized coasts and suggests major differences in terms of ecosystem functioning between both habitats.

New deep learning-based methods for visualizing ecosystem properties using environmental DNA metabarcoding data.

Environmental DNA (eDNA) metabarcoding provides an efficient approach for documenting biodiversity patterns in marine and terrestrial ecosystems. The complexity of these data prevents current methods from extracting and analyzing all the relevant ecological information they contain, and new methods may provide better dimensionality reduction and clustering. Here we present two new deep learning-based methods that combine different types of neural networks (NNs) to ordinate eDNA samples and visualize ecosystem properties in a two-dimensional space: the first is based on variational autoencoders and the second on deep metric learning. The strength of our new methods lies in the combination of two inputs: the number of sequences found for each molecular operational taxonomic unit (MOTU) detected and their corresponding nucleotide sequence. Using three different datasets, we show that our methods accurately represent several biodiversity indicators in a two-dimensional latent space: MOTU richness per sample, sequence α-diversity per sample, Jaccard's and sequence ß-diversity between samples. We show that our nonlinear methods are better at extracting features from eDNA datasets while avoiding the major biases associated with eDNA. Our methods outperform traditional dimension reduction methods such as Principal Component Analysis, t-distributed Stochastic Neighbour Embedding, Nonmetric Multidimensional Scaling and Uniform Manifold Approximation and Projection for dimension reduction. Our results suggest that NNs provide a more efficient way of extracting structure from eDNA metabarcoding data, thereby improving their ecological interpretation and thus biodiversity monitoring.