A two-decade microbial time series from a freshwater lake, introducing the limony and TYMEFLIES datasets

How microbial communities change over time is key to understanding how ecosystems will respond to global change, and to answering fundamental questions about how microbial ecology and evolution unfold. However, our understanding of microbial change is limited by a lack of long-term observations. Using archived filters from freshwater Lake Mendota (WI, USA), we created a 20-year microbial time series that begins in year 2000, a decade before Illumina DNA sequencing hit the market. We characterized 1,023 samples and controls with 16S rRNA gene amplicon sequencing in the "limony" dataset, and 471 samples and controls with shotgun metagenome sequencing in the "TYMEFLIES" dataset. In addition to the raw sequencing data, we point users to associated data products: paired environmental data collected by the North Temperate Lakes Long-Term Ecological Research program (NTL-LTER), our R package to work with curated limony data, and metagenomic assemblies and metagenome-assembled genomes (MAGs) created from the TYMEFLIES dataset. The limony and TYMEFLIES datasets, along with their paired environmental data and reference genomes, are a unique and ready-to-use resource for diverse scientists including limnologists, ecologists, evolutionary biologists, bioinformaticians, modellers, and educators.