FoodMicrobionet v4: a large, integrated, open and transparent database for food bacterial communities

With the availability of high-throughput sequencing techniques our knowledge of the structure and dynamics of food microbial communities has made a quantum leap. However, this knowledge is dispersed in a large number of papers and hard data are only partly available through powerful on-line databases and tools such as QIITA, MGnify and the Integrated Microbial Next Generation Sequencing platform, whose annotation is not optimized for foods. Here, we present the 4 th iteration of FoodMicrobionet, a database of the composition of bacterial microbial communities of foods and food environments. With 180 studies and 10,151 samples belonging to 8 major food groups FoodMicrobionet 4.1.2 is arguably the largest and best annotated database on food bacterial communities. This version includes 1,684 environmental samples and 8,467 food samples, belonging to 16 L1 categories and 196 L6 categories of the EFSA FoodEx2 classification and is approximately 4 times larger than previous version (3.1, https://doi.org/10.1016/j.ijfoodmicro.2019.108249 ). Using data in FoodMicrobionet we confirm that taxonomic assignment at the genus level can be performed confidently for the majority of amplicon sequence variants using the most commonly used 16S RNA gene target regions (V1-V3, V3-V4, V4), with best results with higher quality sequences and longer fragment lengths, but that care should be exercised in confirming the assignment at species level. Both FoodMicrobionet and related data and software conform to FAIR (findable, accessible, interoperable, reusable/reproducible) criteria for scientific data and software and are freely available on public repositories (GitHub, Mendeley data). Even if FoodMicrobionet does not have the sophistication of QIITA, IMNGS and MGnify, we feel that this iteration, due to its size and diversity, provides a valuable asset for both the scientific community and industrial and regulatory stakeholders.