Boreal forests represent a large carbon sink, accounting for 32% of the global carbon store. As intensive human activity (e.g. forestry, land use change, fertilization) in a vast majority of boreal forest rapidly alters the ecosystems, it is essential to advance our understanding of soil processes, to enable informed future policy decisions about forest management and ensure a maintained, and preferably increased, carbon sink. Soil fungi play a central role in regulation of soil organic matter dynamics in forest ecosystems and are therefore subject to particular attention. In boreal forest soils, fungi are the main microbial group in term of biomass, and mycorrhizal fungi, living in symbiosis with plant roots, are particularly important in regulating carbon sequestration. New innovative molecular techniques (metatranscriptomics) allow us to open the black box of microbial process in soils, by obtaining massive data of expressed genes from entire microbial communities, and investigate how microorganisms regulate soil organic matter dynamics, directly in the ecosystem. The objective of this project is to develop an approach to highlight fungal functional traits related to soil fungal community ecology. Specifically, to evaluate the mechanisms relative to carbon storage and how they are affected by different forms of rotation forestry. In this purpose, the role of fungal community in carbon transformation, during decomposition and microbial metabolism, is assessing by analyzing expression of genes involved in the production of enzymes implicated in organic matter degradation, stress tolerance and intracellular CO2 release, at the ecosystem level. The project have focused on several different study areas situated in boreal forests of different ages and with different degree of forestry impact. RNAs have been extracted from the soil and mRNAs have been subjected to massive Illumina HiSeq sequencing (more than 100 million reads per sample). To decrease the calculation time during bioinformatic analysis and focus on relevant ecological questions, we develop a targeted markers assembly using the HMMER software, based on hidden Markov models. Publically available reference sequence databases (e.g. CAZy or JGI Mycocosm) are used to guide transcript filtering and identification of expressing organisms. To conclude, the purpose of this pioneer approach is to provide novel insights about the interplay between fungal traits, such as decomposer capacity and metabolic efficiency, and ecosystem level processes, such as carbon sequestration, in order to increase the predictive capacity of ecosystem models.