There is ample research on foods and diet patterns against obesity in the West, a medical condition that has been formally recognized as a global epidemic. A huge variety of diets has been developed to prevent obesity, which is also becoming an increasing problem in Asia. Wouldn’t it be great if we could translate those specific diets and recipes to the Asian cuisine? What if this knowledge about food could be shared more easily, if we were for example able to view in a database what ingredients are available in Asia that could replace ingredients from the West? Scientific knowledge on food is growing quickly, but we can capitalize more efficiently on the available data.
Scientific publications contain a wealth of unexplored and unstructured data. If we can rapidly search this sea of knowledge, from articles to patents and blogs, on relevant insights, it would accelerate the research process tremendously. What is more, text mining bridges the boundaries between domains, as a result of which hidden connections may emerge. Mining and cross-linking information from the domain of food with information from other domains and sources can provide valuable insight in the physiology of organisms, explain experimental data or lead to new hypotheses.
The technological developments in life science research have led to a vast increase in data that are available in public and proprietary databases. In order to efficiently capitalize on these data, dedicated vocabularies and algorithms are necessary for annotating, searching, filtering and integrating data from various sources. Although a number of generic knowledge discovery and knowledge management (KDKM) and text mining (TM) tools exist, their application in life science areas, in particular food research is limited. One reason is the absence of structured vocabularies that are of interest to specific applications in food research.
In this research project structured vocabularies covering the food domain are developed. These vocabularies will be incorporated in existing KDKM and TM tools to link potentially related research findings. Using these vocabularies, insights into the function of bacteria and organisms involved in food processing can be generated, for example. Furthermore hidden relations which might lead to a better understanding of how processes work or might lead to improved products can be identified. These relations can be used to generate hypotheses addressing important areas in food research.
The ontologies and related (web) services will be evaluated in two ways. Firstly, the ontology and associated services will be validated by measuring the quality of semi-automatic annotations and by demonstrating improved integration of food research data. Secondly, the hypotheses generated with the above computational methods will be validated in experiments in which the effects of probiotics and neutraceuticals are measured in in-vitro and in-vivo models for health. Visualizations of terms often co-occurring with bacteria do already summarize the main applications of those bacteria by a single mouse click. Introducing food concepts enables us to tag those terms, enriching the set of terms with useful concepts which enables discovery of new relations between bacteria and concepts.