BGC expression correlation analysis.

To assess the relationships between expression and (i) abiotic environmental conditions and (ii) abundance of known competitors and predators of Microcystis, simple linear regressions were generated. Correlations were generated for all abiotic variables and for all organisms identified in each metagenome that had at least 200 genes assigned to their taxonomic identification. Organism abundance and analysis were completed at the genus level. Pearson’s correlation coefficient, R, as well as P values were determined to assess the strength of correlation and the significance of individual variables, respectively. Abiotic environmental condition data were provided by CIGLR and can be found in Table S6. The relative abundance of organisms was determined by aligning QC reads to the UniRef100 database using DIAMOND (111). Relative abundance of each organism was then determined by calculating reads per kilobase per million (RPKM) of all genes assigned to the taxa using the AnnotateContigs.pl script found at https://github.com/TealFurnholm/Strain-Level_Metagenome_Analysis/wiki/Step-2.-Gene-Annotation-and-Community-Analysis, which summarizes the complete community analysis workflow and outputs.