*2.2. EuroFIR AISBL's Food Data Banks: Main Features and Updates* 2.2.1. FoodEXplorer

FoodEXplorer [1] is an innovative interface for searching simultaneously food composition data in most publicly available national FCDBs in the European Union (EU) Member States as well as Canada, the United States, New Zealand, and Japan. Currently, FoodEXplorer [60] host 40 interoperable national FCDBs (EuroFIR AISBL FoodExplorer, https://www.eurofir.org/foodexplorer/foodgroups.php, accessed on 6 June 2022). Food and nutrient data are linked throughout LanguaLTM.

For the search, "access African and EMR data" (https://www.eurofir.org/FoodEXplorer/ foodgroups.php?data=D2, accessed on 6 June 2022) was also created, in addition to "access on FoodEXplorer". Open (publicly available) datasets have been developed and published for Australia and New Zealand, Iran, Iraq, Kuwait, Morocco, Pakistan, South Africa, and Tunisia, supported by projects including EMR (Eastern Mediterranean Regional data, funded by UK Medical Research Council Global Challenges Research Fund in collaboration with the World Health Organization's Eastern Mediterranean Regional Office), African data (funded by the UK Biotechnology and Biological Sciences Research Council Global Challenges Research Fund in collaboration with the FAO INFOODS AFROFOODS network), and Food Standards Australia New Zealand (funded by the Commonwealth of Australia and Food Standards Australia New Zealand, 2018).

In this regard, it is worth mentioning the work of Ene-Obong et al. [61], which describes the importance and use of reliable food composition data by nutrition/dietetic professionals in solving Africa's nutrition problems and focuses on constraints and the roles of FAO INFOODS and AFROFOODS as well as other stakeholders in future initiatives. The authors noted how AFROFOODS recommended that compilation, dissemination, and use of food composition tables (FCTs)/FCDBs should be given priority and included in country and regional development and investment plans. Similarly, AFROFOODS has called on governments to incorporate food composition into curricula for higher education, particularly nutrition and dietetics professional learning, but also health and agriculture [61]. More recently, EuroFIR AISBL and Quadram Institute Bioscience (QIB, UK) have assisted AFROFOODS in capacity building and development of a website with help from Premotec GmbH (PMT, CH)—a Swiss company experienced into implementation of software solutions for food data, i.e., food composition, food consumption and total diet studies—to increase visibility and enhance networking, and development of a road map for future activities.

In 2019, analysis of harmonized EuroFIR documentation for macronutrient values in 26 European FCDBs was carried out by Westenbrink et al. [41] to evaluate the impact of harmonized documentation and its usefulness for research and/or policy; documentation of most properties describing nutrient values was complete, even if the percentage coded as unknown varied from 14% to 49% for value and method types, method indicator, and acquisition type. The same authors reported some inconsistencies and incomplete information (about 65% missing) in coding and documentation [41]. Additionally, they noted how easy data exchange was supported by harmonized procedures for data documentation according to EuroFIR guidelines, even if comparability of carbohydrate, dietary fiber, protein, and energy values remained difficult due to multiple definitions and formulae, particularly lack of details about analytical and calculation methods [41].

A potential solution to improve harmonization was defined and published in 2020 in EuroFIR FoodEXplorer Standard [42], providing updated guidelines for collecting, compiling, and updating food composition data. In particular, the following actions were proposed for datasets being uploaded to FoodEXplorer: (i) before uploading, EuroFIR will standardize data units; (ii) energy will be re-calculated using European labelling legislation EU Regulation No. 1169/2011 (https://bit.ly/3g5yegE, accessed on 26 October 2022) recommendations and presented as both kcal and kJ; and data on less common components, such as polyols, organic acids, and salatrims, should be provided and included in recalculation; (iii) vitamin A will be presented as retinol activity equivalents (RAE); (iv) for calculated components, only one value per component ID will be shown; and (v) the use of both LanguaLTM food description coding and FoodEx2 classification and description coding is recommended but not mandatory [42]. In 2020, following user feedback, functionalities of FoodEXplorer were updated, specifically: (i) advanced search functionalities; (ii) formatting of downloads for Excel; (iii) options for sorting components; (iv) presentation of component values and documentation; and (v) selection of foods for comparison.

Elaborations and applications using data from FoodExplorer were carried out among users and compilers. An example of a FoodEXplorer application for creating specialized food composition datasets, in this case for vitamin D in foods based on European standards for dietary intake assessment, was described by Mileševi´c et al. [62] while Gurinovi´c et al. [63] elaborated development, functionalities, and application of DIET ASSESS & PLAN (DAP) software, a platform for standardized and harmonized food consumption collection, comprehensive dietary intake assessment, and nutrition planning to support public health nutrition research in Central Eastern European Countries (CEEC). DAP enabled exploitation of national FCDBs from FoodEXplorer and their exploration using other online tools [63].

Another example of the utilization of data from FoodExplorer was given by Fish-Choice 2.0 (www.fishchoice.eu, accessed on 26 October 2022) [64]. FishChoice 2.0 is a tool, relaunched by Marquès et al. [64] as a tool for consumers and nutrition professionals, which delivers information about health benefits/risks as well as some sustainability information for fish and seafood on an individual basis, based on calculation of nutrients and contaminant intakes [64]; FoodEXplorer was used to collect nutrient data for fish and seafood species typically consumed in Europe for inclusion in FishChoice 2.0 [64].

#### 2.2.2. eBASIS—Bioactive Substances in Food Information System

Demand for easily accessible information on composition, intakes, and activities of bioactive compounds is significant among researchers. Bioactive Substances in Food Information System (eBASIS) [65] is a web-based database containing scientifically validated information describing the composition of bioactive compounds in major European plant foods. eBASIS was launched in 2006 [66,67] as a user-friendly, efficient, and flexible interface for the scientific community and food industry. It was the first EU harmonized database combining composition data and biological effects for compound classes, including polyphenols, isoflavones, glucosinolates, phytosterols, glycoalkaloids, and xanthine alkaloids, in 15 languages [68,69].

EuroFIR eBASIS was compiled using data from the peer-reviewed literature evaluated critically by experts. Tutorials for users are available online (https://www.eurofir.org/ourtools/ebasis/, accessed on 26 October 2022) as well as via a short video demonstrating how eBASIS can be used (*Introduction to eBASIS*, https://ebasis.eurofir.org/Default.asp, accessible on 6 June 2022). Currently, eBASIS contains 44,664 datapoints for bioactive compounds for 276 plant-based foods, distributed in main classes, e.g., 677 datapoints for phenols, 3945 datapoints for flavonols, 4581 datapoints for anthocyanins, 881 datapoints for carotenoids, 2695 datapoints for lignans, and 2654 datapoints for glucosinolates (https: //ebasis.eurofir.org/Default.asp, accessed on 6 jube 2022).

Information included in eBASIS was described by Pilegaard et al. [70] and, in 2011, the utility of eBASIS tested in a phytosterols case study [71]. In 2017, a new interface linking the eBASIS bioactives database and the Creme Nutrition® model was developed for the BACCHUS project (http://bacchus.cremeglobal.com/bacchus/, accessed on 26 October 2022) [72]. The eBASIS-Creme Global exposure tool enables users to assess compound intakes from various foods across populations to determine whether compounds required to obtain a claimed effect can be reasonably consumed within a balanced diet [72]. In 2018, an update on extractable and non-extractable antioxidants was completed [73] with the addition of 437 quality-evaluated datapoints. This update was the first example of building a resource dedicated to antioxidant properties within the existing resource. An updated eBASIS user guide was published at the same time, covering data concerning antioxidant properties and extractable and non-extractable compounds (https://ebasis.eurofir.org/ files/basis\_antiox.pdf, accessed on 6 June 2022).

The input form for data includes bibliographic references, food information (i.e., plant, part, subspecies/cultivar, maturity, season, growing conditions, etc.), processing (i.e., shape, state or form, heat treatment, cooking method, treatment applied, preservation method), sampling information (i.e., primary sample unit size, analytical sample size, sample plan, sample handling, etc.), compositional information (i.e., compound class, analytical method, concentration, extraction, and preparation, identification, etc.), and quality assessment. For each eBASIS section (plant/food description, processing defined, sampling plan, sample handling, compound identification, analytical method, analytical performance), transparent quality systems are included, ensuring eBASIS as a reliable resource for research with upto-date information about plant food phytochemicals.

eBASIS was developed to present raw rather than aggregated data, reflecting variations in bioactive compositions related to cultivar, plant part, growing conditions, processing, and country of origin; there are multiple datapoints for each compound/food combination. To better meet requirements for aggregated bioactive composition data in dietary intake assessment, eBASIS data structures are being organized to link plant food data and bioactives with dietary intake assessment outputs and coding systems. At the same time, the architecture permits future inclusion of food data from animal origins and/or addition of new data on other plant foods/products or classes of compounds, emphasizing the need to envisage potential needs and gaps during development.
