AI-Enhanced eDNA Solutions for Comprehensive Marine Biodiversity Monitoring and Reporting – ADNeIA

Human-induced pressures and climate change are having profound and growing impacts on marine ecosystems, making the continuous monitoring of biodiversity and ecological status increasingly vital. Effective monitoring requires the development, modeling, interpretation, and mapping of ecological inventories and indicators, particularly for fish and crustaceans which are central to such assessments given their ecological roles but also vulnerability. Traditional monitoring techniques for these species—such as fishing surveys, diver observations, and baited underwater cameras—are often invasive, costly, and limited in scope. Over the past decade, the use of environmental DNA (eDNA) metabarcoding has emerged as a revolutionary non-invasive alternative. All marine organisms release DNA into their environment through fluids and cells, which can persist in water for hours or days. This DNA can be collected, filtered, and sequenced to identify species using universal primers targeting specific taxonomic groups. SPYGEN, a pioneer in the field of eDNA research since 2011, has played a central role in advancing these methods, especially for aquatic ecosystems. Their foundational research has been widely cited and has significantly influenced the field. However, effectively assessing ecological status and biodiversity patterns from eDNA surveys requires sophisticated modeling approaches that account for complex interactions between a myriad of factors. This is where artificial intelligence can unlock the potential of eDNA monitoring, reporting and mapping, with the key contribution of the Chair holder. To face this current scientific and market challenges and continue driving innovation, SPYGEN now aims to enter a new phase of development by proposing advanced tools for sampling biodiversity in hard-to-reach environments like offshore wind farms, underwater canyons, and deep-sea habitats. The company also plans to enhance its genetic reference databases and bioinformatic pipelines by integrating the latest advances in artificial intelligence, including large language models, to better analyze and interpret large volumes of eDNA data. Moreover, the Chair seeks to model and contextualize biodiversity indicators using deep learning algorithms coupled with satellite imagery, offering more accurate and scalable assessments of marine ecosystems. Finally, the Chair plans to develop a digital mapping platform to visualize biodiversity trends and species distribution over time and space. This system will support online access to data and produce standardized reports aligned with emerging frameworks such as Verifiable Nature Units (VNU), biodiversity credits, and corporate sustainability reporting (CSRD).