Nutritional studies increasingly use new technologies that result in large data sets. The power of these technologies is that many parts of the human physiology are described, but makes it difficult for researchers to properly interpret the outcomes. Part of this problem is caused by the fact that the number of parameters measured is much greater than the number of participants in the survey, which weakens the statistical results and makes it more difficult to draw biological conclusions. This problem can be overcome by combining the results of similar studies. However, this requires structured and standardized data storage and integrated analysis methods.

ENPADASI has provided an open access research infrastructure (RI) for all nutritional studies. Standardization is important in order to find comparable studies. For this reason, the metadata, describing the study and the phenotypic data are standardized. Through the implementation of ontologies in the infrastructure and the development of a new ontology for nutritional studies (ONS) and the sharing of standardized protocols, the quality of the collected data is increased. To combine studies, the Phenotype database and Opal/DataShield have been integrated in the DASH-IN infrastructure, which contains options for performing integrated analysis on multiple studies and facilitates further exploitation of data. Nutritional researchers are trained to use the system. ENPADASI partners will continue to encourage nutritional researchers to share their future studies, which will increase the future impact of the project.

Making data available after finalization of a study is now sometimes a prerequisite for funding, but this is often not clearly defined in the consortium agreement. In addition, the participants of studies have not always been informed about the fact that the researchers want to share the data later and/or to use the data for other questions. For these ethical and legal issues ENPADASI has delivered some solution.

- Study data from experimental studies in animals and humans as well as data from observational studies can be shared, searched for and jointly analyzed in the DASH-IN infrastructure.
- We have shown that the data can be reused for new research questions.
- Reusability of data is increased by use of ontologies and a definition of the minimal requirements for data sharing
- A study appraisal tool helps to judge how well a study relates to your research question
- Ethical issues challenge sharing of data but this can be overcome by careful planning and by several different strategies, e.g.
- Sharing of summarized data largely resolves ethical issues, also for non-anonymous data
- A broad informed consent makes it possible to use non-anonymous data for a broad range of scientific questions
- Sharing anonymous data ethically has no hurdles
- An embargo period should be included in all consortium agreements for publicly funded projects. This period should not be too short discouraging researchers from uploading data.
- Training is essential for rich meta-data and data sharing
- Combining local and central data sharing solutions facilitates long term sharing as it limited the resources needed (less duplications of data) and overcomes several data safety issues.