The multifactorial nature of nutrigenomics research calls for extensive phenotyping.
"Nutrition research is a mess. It involves many processes, and the effects of dietary compounds on nutrition are very subtle, so the quantification of the effect of diet on health is very difficult," says Ben van Ommen, program director of systems biology at The Netherlands Organization for Applied Scientific Research. "The study of small changes needs very accurate technology. You need to be able to very carefully, very accurately, very precisely quantify small phenotypic changes, and that's exactly where 'omics technologies pop in."
When researchers first began working on nutrigenomics more than a decade ago, they "thought we were going to conquer the world," van Ommen says. In the late '90s, while burgeoning genomic, proteomic, transcriptomic, and metabolomic approaches were on the ascent, van Ommen and his colleagues thought that by "surfing on the technologies of the biology revolution, we would revolutionize nutrition research."
That ride was short-lived. However, with time, researchers began "to apply these technologies in ways that we could start assessing minor changes," van Ommen says. "We learned to quantify health using nutrigenomics."
Still, much of researchers' early success in the field was driven by the application of single technologies. As a result, van Ommen says the resulting data painted only partial pictures. "People did, for example, whole-genome [sequencing], or they did transcriptomes, or they did proteomes or metabolomes, and they built databases for transcriptomics, databases for metabolomics, and so on," he says. "Science is not about technologies, it's about studies. So we said we had to build something essentially different — not something technology-focused, but study-centered."
Start with the study
Initiated by the European Nutrigenomics Organization, the Nutritional Phenotype Database, or dbNP, allows users to interpret the results of studies involving multiple 'omics data combined with traditional clinical metrics.
"We start with the study design, then connect all the various 'omics pipelines, and then reabsorb all that data together," van Ommen says. The dbNP, he adds, is "completely open source." Despite this, the database is not centralized. "It is a piece of software that anyone — any university, any company — can download and install on its server and upload data," van Ommen says. "It's completely study owner-driven — you are in control of your own data. ... You decide when and with whom to share it."
This philosophy has also inspired dbNP's organizers to build a governing structure to sustain it. "We're now creating the Phenotype Foundation, an organization with an open-access and public-domain theoretical basis that will sustain and govern the Nutritional Phenotype Database. This is something that is essential," he says. The foundation's mission, he adds, is "to empower scientists ... with standardized software and knowledge stores to store, manage, and retrieve information and data on genotypes and phenotypes."
It's this meticulously planned, precompetitive governing structure that van Ommen says will give the dbNP staying power. "Databases come and go," he says. "Not this time. This one is built for the future."
At present, the dbNP team is focused on expansion. Researchers are building additional modules to assess intestinal microflora and to analyze MRI imaging data, for example. Going forward, van Ommen and his colleagues have their sights set on the clinic.
"The next step will be that major step — to bring this to the clinical area," van Ommen says. With growing participation and industry interest, dbNP will make an impact on basic research and, eventually, clinical care, van Ommen says.