Fast and accurate detection of metal resistance genes using MetHMMDb

Heavy metal pollution, driven by industrialization, is a growing environmental challenge with severe ecological impacts. One of key factors in fighting this issue is microbial resistance to heavy metals, and its potential for use in bioremediation and environmental monitoring. Unfortunately, existing detection methods often fail to identify distant homologs or to provide insight into the function of microbial metal resistance genes (MMRGs). To overcome these limitations, we developed MetHMMDB - a novel database containing 254 profile Hidden Markov Models (HMMs) representing 121 microbial MMRGs and functions. Unlike traditional sequence similarity-based tools, MetHMMDB leverages HMMs to enhance detection sensitivity and functional specificity across diverse microbial communities. MetHMMDB was created through iterative database searches, combined with sequence clustering and structural prediction, combining both recent advances in bioinformatics and manal curation of data. In its core, the database prioritises functional annotation, ensuring accurate detection and classification of genes responsible for metal resistance. Our results indicate that MetHMMDB outperforms sequence-based approaches, identifying more than twice as many MMRGs in metagenomic datasets, including extreme environments. Additionally, application to agricultural soil revealed distinct resistance profiles linked to soil quality. By improving the identification and functional characterization of metal resistance genes, MetHMMDB facilitates interdisciplinary research in microbial ecology, environmental monitoring, and bioremediation. This robust and versatile resource represents a significant advancement in understanding microbial adaptation to heavy metal contamination and supports the development of effective environmental management strategies. The database is available at https://github.com/Haelmorn/MetHMMDB.