Abstract
Marine protists form complex communities that are shaped by environmental and biological ecosystem properties, as well as ecological interactions between organisms. While all of these factors play a role in shaping protistan communities, the specific ways in which these properties and interactions influence protistan communities remain poorly understood. Fourteen years and 9 months of eukaryotic amplicon (18S-V4 rRNA gene) data collected monthly at the San Pedro Ocean Time-series (SPOT) station were used to evaluate the impacts that environmental and biological factors, and protist-protist interactions had on protistan community composition. Statistical analysis of the amplicon data revealed that seasonal patterns in protistan community composition were apparent, but that the environmental data collected through routine time-series sampling efforts could not explain most of the variability that was evident in the communities. To identify some of the protist-protist interactions that may have played a role in shaping protistan communities, ecological networks were constructed using the amplicon data and the network predictions were compared against a database of confirmed protist-protist interactions. The database comparisons revealed hundreds of established parasitic, predator-prey, photosymbiotic, and mutualistic relationships in the networks. Although many interactions were confirmed using the database, these confirmed interactions constituted only 2% of the interactions identified at the SPOT station, highlighting the need to better characterize protist-protist interactions in marine environments. Finally, the network-predicted interactions that were not found in the database were used to identify putative, novel protist-protist interactions that may have played a role in structuring the protistan communities at the SPOT station. Importance: Network analyses are commonly used to identify some of the ecological interactions that may be occurring between protists in the ocean; however, evaluating predictions obtained from these analyses remains difficult due to the large number of interactions that may be recovered and the limited amount of information available on protist-protist interactions in nature. In this study, ecological network analyses were conducted using data collected at the San Pedro Ocean Time-series (SPOT) station and the network predictions were compared against a database of established protist-protist interactions. These database comparisons revealed hundreds of confirmed protist-protist interactions, and thousands of putative, novel interactions that may be occurring at the SPOT station. The database comparisons carried out in this study provide a new way of evaluating network predictions and highlight the complex, yet critical role that ecological interactions play in shaping protistan community composition in marine ecosystems.