In recent years, metabarcoding has become a key tool to describe microbial communities from natural and artificial environments. Thanks to its high throughput nature, metabarcoding efficiently explores microbial biodiversity under different conditions. It can be performed on environmental (e)DNA to describe so-called total microbial community, or from environmental (e)RNA to describe active microbial community. As opposed to total microbial communities, active ones exclude dead or dormant organisms. For what concerns Fungi, which are mostly filamentous microorganisms, the relationship between DNA-based (total) and RNA-based (active) communities is unclear. In the present study, we evaluated the consequences of performing metabarcoding on both soil and wood-extracted eDNA and eRNA to delineate molecular operational taxonomic units (MOTUs) and differentiate fungal communities according to the environment they originate from. DNA and RNA-based communities differed not only in their taxonomic composition, but also in the relative abundances of several functional guilds. From a taxonomic perspective, we showed that several higher taxa are globally more represented in either "active" or "total" microbial communities. We also observed that delineation of MOTUs based on their co-occurrence among DNA and RNA sequences highlighted differences between the studied habitats that were overlooked when all MOTUs were considered, including those identified exclusively by eDNA sequences. We conclude that metabarcoding on eRNA provides original functional information on the specific roles of several taxonomic or functional groups that would not have been revealed using eDNA alone.