Laccases belong to the superfamily of multicopper oxidases (MCO), a group of enzymes with the ability to reduce oxygen to water in a reaction without producing harmful byproducts. Laccase activity is influenced by many factors, such as structure; the number, location, and binding status of copper ions; and the substrate-binding status. A large number of sequences that have not been experimentally characterized yet have been annotated as laccases. However, the biological functions of the characterized MCOs are thought to vary, and the substrate spectrum overlaps with that of other MCOs. Here, we identified 34 putative bacterial laccase sequences from metagenome data of industrial wastewater. We used machine-learning tools to screen enzymes with laccase activity by combining the T1 copper-binding capacity, the overall copper-binding capacity, and the substrate-binding capacity. We also used the software comparisons to remove sequences with large discrepancies between different software applications. Three-dimensional structures of identified enzymes were predicted using AlphaFold, the positions of metal ions within the proteins were predicted by Metal3D and AutoDock-Vina, and their docking with 2,2'-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) as a substrate was predicted by Rosetta and AutoDock-Vina. Based on the docking results, we selected 10 high-scoring proteins, two low-scoring proteins, and one composite protein for expression using the pET-21d (+) vector. In line with our predictions, all selected high-scoring proteins exhibited activity towards ABTS. Overall, we describe a method for discovering and designing novel bacterial laccase-like multicopper oxidases, thereby offering increased possibilities for the degradation of various harmful components derived from environmental pollution.