“…BLAST (Basic local alignment search tool), which is based on sequence similarity, and HMM (Hidden Markov model), which is based on functional domains are important tools for gene or protein annotation in meta‐omics datasets [ 23 , 24 ]. With the increasingly updated of many large comprehensive databases (e.g., KEGG, eggNOG, and Uniprot), as well as the release of small specific functional gene databases (e.g., VFDB, BacMet, and NCcyDB), and the development of pollutant biodegradation resources databases (e.g., mibPOPdb, OxDBase, and EAWAG‐BBD) [ 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 ]. Function prediction of metagenomic genes has been accelerated and can be used to evaluate the metabolic potential of microbial communities.…”