Metabarcoding is the large scale taxonomic identification of complex environmental samples via analysis of DNA reads of one marker gene (16S, ITS, 18S, COI…). The aim of metabarcoding analysis is to provide a table of abundance of OTUs/ASVs, as close as possible to the species, per sample as well as a descriptive statistical analysis of the composition of the targeted microbial population. The FROGS software offers such results [1, 2]. The different tools developed within FROGS allow users to process their data in command lines or via a user friendly Galaxy interface and to obtain different graphical and descriptive outputs. Metagenomics analysis allows the sequencing of all the DNA contained in a sample. It gives access to the functional potential of the species present in the environment [3]. Unlike metagenomics, metabarcoding does not provide these functional profiles of a microbial population, by being restricted to o ne marker gene. However, algorithms have been developed to bypass this restriction and obtain a prediction of the functional potential of a sample, at low cost. For this purpose, PICRUSt2 has been commonly used these last years [4]. Firstly, PICRUSt2 place s the marker gene sequences (16S, ITS or 18S) of interest into its reference tree, that is used as the basis of functional predictions. After, it predicts number of marker and function copy number in each OTU/ASV. Then, for each sample, it calculates funct ions abundances and finally, pathway abundances are inferred, based on functional profile. The plus of FROGS 4.0 : additional outputs to the original PICRUSt2 tools that will guide users in the correct interpretation of their data and an ease of use thanks to the possibility of running these tools via Galaxy. Some "hidden" PICRUSt2 outputs are also exploited (reporting incongruence between taxonomic NSTI threshold confidence indicator, decision support graphic to help choosing the NSTI threshol d ...). In keeping with the FROGS philosophy, graphic outputs are also displayed to make the experience intuitive.