One of the greatest challenges for pioneering researchers in
new fields is establishing standards for the exchange of information.
That challenge has never been greater than for the emerging field
of metabolomics, the study of the myriad metabolites in an organism.
A recent workshop sponsored by several NIH entities (the NIH Roadmap
Metabolomics Technology Development program, NIDDK, NIGMS and NIEHS)
focused on how to share all the complicated data this new field
Whereas genomics is the study of all the genes in an organism
and proteomics the study of all the proteins, metabolomics aims
to depict the physiological states of cells and organisms by focusing
on carbohydrates, lipids, signaling molecules and other metabolites.
To a far greater extent than genomics and proteomics, metabolomics
studies also regularly include compounds from the things we eat
and breathe, and from the microbes living in our bodies.
The NIH Roadmap initiatives include metabolomics as a major priority,
with the goal of understanding and detecting differences in biological
pathways and networks between normal and disease states. The presentations
at this workshop gave a hint at the enormous potential this type
of research holds. The analysis of metabolites in blood, urine
or tissue extracts promises to be a powerful tool for disease diagnosis,
revealing a snapshot of what biological pathways are going awry.
The complexity of achieving this vision, however, is daunting.
Metabolomics data is far more disparate than DNA and protein sequence
data. The methods involved in analyzing all these metabolites include
NMR, mass spectrometry, molecular probes, electrophoresis and others.
Even within these categories, there are myriad techniques for generating
and analyzing data. Dr. Oliver Fiehn of the University of California
at Davis pointed out in his talk that there are "dozens of techniques" for
performing mass spectrometry alone.
Compounding this, metabolomics experiments are very sensitive
to protocol changes. As Dr. Wayne Matson of ESA Biosciences, Inc.,
noted, "The metabolome really depends on where you look." They
are snapshots in time, dependent on a host of environmental and
experimental design factors that all need to be included in information
databases. Something seemingly as simple as how long a sample sits
before being frozen or centrifuged can greatly influence results.
Standards-setting is an ongoing process, and this meeting was
an early step on a long path. Metabolite names and definitions
need to be standardized; databases with disparate data need to
learn to talk to each other; and perhaps most challenging, sample
descriptions and countless experimental details need to be standardized.
In light of all this activity, it's interesting to note that,
while metabolomics is among the most cutting-edge research around,
as Dr. Rima Kaddurah-Daouk of Duke University Medical Center (and
president of the Metabolomics Society, which co-organized the event)
pointed out, it actually follows in the footsteps of the biochemists
who first elucidated those metabolic pathways we all studied in
our high school and college textbooks. Metabolomics is a return,
at an "omics" level, to biochemistry. It is, in a sense, one of
the oldest sciences around.