“…However, there are some challenges in streamlining the transition from the discovery stage of an enzyme through its metagenomic analysis, and ultimately towards its end‐user applications (Jemli et al ., ). The major technological bottlenecks include (i) a low proportion of coding metagenomic DNA accessible for expression (Guazzaroni et al ., ), (ii) a low proportion of enzymes selected from screens perform well in industrial settings (Martínez‐Martínez et al ., ), (iii) a lack of relevant substrates for screening (Fernández‐Arrojo et al ., ), (iv) insufficient screening methods for rare enzymatic activities (Singh, ), (v) a poor performance of enzymes under non‐natural conditions (Fernández‐Arrojo et al ., ), (vi) the existence of enzymes that are inactive after expression in the widely used host Escherichia coli (Loeschcke et al ., ), (vii) the lack of reliable bioinformatics pipelines for analysis of next‐generation sequencing data generated from positive hits or direct sequencing (Nyyssönen et al ., ), and (viii) the lack of reliable functional prediction of hypothetical proteins (Mende et al ., ; Anton et al ., ; Bastard et al ., ; Chistoserdova, ). In addition, the minimization of amplification of annotation mistakes (sequence/activity incoherence) in databases (Fernández‐Arrojo et al ., ) is among the more challenging issues to be solved.…”