Edoardo Pasolli*
University of Naples Federico II
edoardo.pasolli [at] unina.it
Abstract
Strain-level metagenomics is transforming our understanding of microbial diversity across human-associated and food-related ecosystems, with major implications for health, food safety, and microbial ecology. This talk highlights recent advances in computational and analytical approaches for large-scale strain-resolved metagenomics in both human and food microbiomes.
Focus is given to machine learning strategies developed for metagenomic applications, including emerging foundation models for microbiome analysis. The presentation will also introduce key resources supporting cross-study integration, such as harmonized reference catalogs, standardized metadata frameworks, and scalable pipelines for both read-based strain profiling and metagenome-assembled genome reconstruction. Efforts aimed at discovering and systematically characterizing underexplored microbial species and strains will also be discussed, with emphasis on improving reproducibility and comparability across datasets and environments.
Examples of integrative analyses connecting human and food microbiomes will illustrate microbial transmission routes, shared reservoirs, strain tracking, and diet-associated microbial exchange. The session will conclude by addressing current limitations and future priorities toward the development of robust, standardized, and generalizable frameworks for strain-resolved microbiome research.
Keywords: Metagenomics; large-scale analysis; human-food microbiome