Using a custom Python script to predict metabolites in red algae using KEGG functional annotations

Red algae are an important target for medical innovation, biodiscovery, and metabolite extraction. As part of our research into red algal metabolic pathways, we have functionally annotated and provided metabolic reconstructions for over 70 publicly available red algal genome and transcriptome assemblies. However, determining the potential metabolic profiles for these assemblies was not viable at scale. Thus, we developed a custom Python program to automatically retrieve compound data based on metabolic pathway annotations sourced from the KEGG database, including gene orthologues, modules, and reactions. Our program uses regular expression matching to search entries retrieved from the KEGG API to identify other entry codes, such as reaction or module codes, as well as associated data such as formulas, weights, and classes, and can retrieve compounds either directly from modules or reactions, or from KEGG orthologues via those entries. Compounds identified through this process were largely conserved between algae of the same genera, but showed much more variation between higher taxa, which correlated with the conservation of orthologues and completed modules between algae. Identifying compounds via modules returned the smallest range of compounds, while searching by orthologue returned more, especially when searching via reactions. Red seaweeds belonging to the class Florideophyceae showed the most unique compounds. This research is part of our greater research aim of exploring red algal metabolism and metabolic pathways. By using our program, we were able to compare the metabolites predicted to be produced by various red algae. We envision that it may be used to refine in vitro analyses of organisms where specific compounds are being searched for, such as in the development of compound databases for spectral libraries, or could be integrated into other research fields, such as phylogenomics and gene duplication analysis.