170
or relatively rare. Approximately 5,688 protein sequences are linked to these metab-
olite entries. Each MetaboCard entry contains >10 data fi elds with 2/3 of the infor-
mation being devoted to chemical/clinical data and the other 1/3 devoted to
enzymatic or biochemical data. Many data fi elds are hyperlinked to other databases.
The HMDB database supports extensive text, sequence, chemical structure and rela-
tional query searches. Four additional databases, DrugBank, T3DB, SMPDB and
FooDB are also part of the HMDB suite of databases. DrugBank contains equiva-
lent information on ~1,600 drug and drug metabolites, T3DB contains information
on 3,100 common toxins and environmental pollutants, and SMPDB contains path-
way diagrams for 440 human metabolic and disease pathways, while FooDB con-
tains equivalent information on ~28,000 food components and food additives.
Metabolomics Bridges the Gap Between Genotype
and Phenotype
In general, phenotype is not necessarily predicted by genotype. The gap between
genotype and phenotype is spanned by many biochemical reactions, each with indi-
vidual dependencies to various infl uences, including drugs, nutrition and environ-
mental factors. In this chain of biomolecules from the genes to phenotype,
metabolites are the quantifi able molecules with the closest link to phenotype. Many
phenotypic and genotypic states, such as a toxic response to a drug or disease preva-
lence are predicted by differences in the concentrations of functionally relevant
metabolites within biological fl uids and tissues.
A genome-wide association (GWA) study has been carried out with metabolic
traits as phenotypic traits (Gieger et al. 2008 ). Genetically determined variants in
metabolic phenotype (metabotype) have been identifi ed by simultaneous measure-
ments of SNPs and serum concentrations of endogenous organic compounds in a
human population. Four of these polymorphisms are located in genes. Individuals
with polymorphisms in genes coding for well-characterized enzymes of the lipid
metabolism have signifi cantly different metabolic capacities with respect to the syn-
thesis of some polyunsaturated fatty acids, the beta-oxidation of short- and medium-
chain fatty acids, and the breakdown of triglycerides. Thus, the concept of
“genetically determined metabotype” as an intermediate phenotype provides a mea-
surable quantity in the framework of GWA studies with metabolomics and might
help to better understand the pathogenesis of common diseases and gene-
environment interactions.
Use of this approach to screen previous GWA studies to look for associations
between the SNPs of interest and clinical measurements infl uencing cardiovascular
disease, revealed overlap between several SNPs that seem to affect both metabolite
biochemistry and clinical outcomes. These metabotypes, in interactions with envi-
ronmental factors such as nutrition of lifestyle, may infl uence the susceptibility of
an individual for certain phenotypes. For example, there are potential links between