“…In the case of the CRISPR‐Cas system, we used the CrisprOpenDB database (comprises 11,767,782 spacer sequences from bacterial genomes) as a comparative reference (Dion et al, 2021). Also, we used the Mining CRISPRs in Environmental Datasets tool ( MinCED v0.4.2 ) (Bland et al, 2007) to retrieve 15,495 spacer sequences from 178 MAGs obtained from Microcoleus ‐dominated mats and non‐axenic isolates from different rivers in USA, New Zealand and Canada (Table S1) (Bouma‐Gregson et al, 2019, 2022; Conklin et al, 2020; Tee et al, 2021; Valadez‐Cano et al, 2023). All spacer sequences were compared against the putative phage sequences recovered from the Wolastoq and Eel River mats using the BLASTn‐short function ( E ‐value ≤1.0 × 10 −10 , sequence identity ≥95%, covering at least 90% of the query spacer sequence) of the BLAST+ package.…”