Predicting Metabolites of Marine Picoeukaryotes using In Silico Metabolomics

Metabolites are small organic molecules that are the product of gene activity. Metabolite profiles are used to understand physiology, organism interactions, and elemental cycling. However, interpretation of metabolomes is currently limited because they are difficult to annotate, authentic standards are required to confidently identify and quantify each metabolite, contamination can be hard to eliminate, and organisms do not necessarily produce all potential metabolites under a given growth condition. Using the Kyoto Encyclopedia of Genes and Genomics (KEGG), we predicted metabolomes from annotated pathways of marine photosynthetic picoeukaryotes, an understudied but abundant group of diverse phytoplankton. Using Python, we created a pipeline to collect, categorize, and visualize the predicted metabolites of picoeukaryotes within KEGG. Compound classification using Classyfire ontology allows for compounds to be grouped by structure for thorough exploration. With predicted metabolomes, standard mixes can be adapted to maximize metabolome annotations. In silico predictions of metabolomes have the potential to reveal compounds that have underappreciated significance in the ocean.