Metabolomics is the comprehensive study of small molecule metabolites present in a biological system and the changes in those metabolites associated with physiological states, environmental conditions, disease, development or genetic variation. The metabolome includes chemically diverse compounds such as amino acids, sugars, organic acids, nucleotides, lipids, vitamins, cofactors, signalling molecules and products of microbial or host metabolism.
Unlike the genome, which is relatively stable, and unlike many aspects of the proteome, which can change over longer or shorter periods depending on context, metabolite concentrations can respond rapidly to changes in substrate availability, enzyme activity, hormonal signals, cellular energy demands and environmental conditions. Metabolomics therefore provides a close biochemical view of cellular state and can reveal the consequences of multiple upstream regulatory processes. It is used in human biology, medicine, plant science, microbiology, nutrition, ecology, pharmacology and biotechnology, and it frequently forms part of multiomics studies that connect genetic information to molecular phenotype.
Metabolites are the chemical intermediates and products involved in biochemical networks, although the boundary of what counts as a metabolite can vary according to the study. Primary metabolites are generally closely connected to essential processes such as energy generation, amino acid synthesis and nucleic acid metabolism, whereas secondary or specialised metabolites often have ecological, signalling, defensive or species specific functions. Examples include glucose, lactate, citrate, alanine, ATP, cholesterol and numerous plant secondary compounds.
Metabolomics can examine endogenous compounds produced by an organism, exogenous substances introduced through diet or environment, and xenobiotic metabolites generated through chemical transformation. The distinction between metabolomics and lipidomics is partly one of scope, with lipidomics concentrating on lipid classes and molecular species, while metabolomics may encompass lipids alongside many other chemical categories. Metabolic profiling can be targeted, focusing on predefined compounds, or untargeted, attempting to measure as broad a chemical spectrum as practical.
Analytical chemistry is central to metabolomics because metabolites differ greatly in polarity, volatility, stability, concentration and chemical structure. Mass spectrometry is widely used because it provides high sensitivity and can be coupled with chromatographic separation techniques. Gas chromatography mass spectrometry is particularly useful for volatile or derivatised compounds, while liquid chromatography mass spectrometry is broadly applicable to many nonvolatile metabolites. Nuclear magnetic resonance spectroscopy provides highly reproducible measurements and structural information but generally has lower sensitivity than mass spectrometry.
Targeted metabolomics often uses authentic standards and calibration curves to quantify specific compounds accurately, whereas untargeted metabolomics generates extensive spectral data that require computational annotation. Identifying a signal is more difficult than detecting it because different molecules can produce similar mass to charge values, and definitive identification may require retention time, fragmentation spectra, authentic standards and orthogonal analytical evidence.
Metabolomics is particularly valuable because metabolite concentrations integrate information from genes, enzymes, cellular pathways and environmental inputs. A genetic variant affecting an enzyme may alter substrate and product concentrations, but the observed metabolic phenotype can also be influenced by diet, temperature, hormones, microbiota, tissue composition and other variables. Metabolic flux is distinct from metabolite concentration. Concentration measures the amount of a compound at a particular time, whereas flux concerns the rate at which material moves through a biochemical pathway.
A pathway can have high flux without unusually high metabolite concentrations if production and consumption are both rapid. Stable isotope tracing can help investigate flux by following labelled atoms through metabolic pathways. This distinction prevents a common misconception that a large concentration change necessarily means that the corresponding pathway has proportionally increased or decreased in activity.
In medicine, metabolomics is used to investigate disease mechanisms, identify candidate biomarkers and study responses to treatment. Changes in metabolites can reflect altered energy metabolism, inflammation, oxidative processes, organ function or microbial activity. Cancer metabolism is a major research area because malignant cells can reconfigure nutrient use and biosynthetic pathways. Metabolomics also contributes to research into inherited metabolic disorders, where defects in particular biochemical reactions can produce characteristic patterns of metabolite accumulation or depletion.
However, a metabolite associated statistically with disease is not automatically a diagnostic biomarker or causal factor. Biomarker development requires independent validation, assessment of specificity and sensitivity, consideration of confounding factors and evaluation in appropriate populations. Dietary intake, medications, age, sex, exercise, circadian timing and sample handling can all affect metabolomic measurements and must therefore be considered carefully in experimental design.
Plant and microbial metabolomics reveal how organisms interact with their environments. Plants produce extensive collections of specialised metabolites involved in defence, attraction, competition, signalling and interactions with herbivores or microorganisms. Microbial metabolomics can identify metabolic products associated with fermentation, nutrient acquisition, environmental adaptation and interactions between species. In environmental studies, metabolite profiles can provide information about ecosystem processes or organismal responses to pollution and environmental stress.
Metabolomics is also important in food science, where it can characterise flavour compounds, nutritional composition, fermentation processes, adulteration and changes during storage. Because many metabolites are chemically active and environmentally transferable, metabolomic data can connect molecular physiology with ecological relationships, making the field relevant beyond conventional cellular biochemistry.
Major challenges in metabolomics include chemical diversity, incomplete databases, concentration differences spanning many orders of magnitude, rapid degradation and strong sensitivity to sampling conditions. Blood, urine, tissue and cell samples can produce very different metabolic profiles, and measurements can change during collection, storage, extraction and analysis. Batch effects and instrument drift require quality control procedures and appropriate normalisation. Statistical analysis must address high dimensionality and multiple comparisons without producing models that overfit the data.
Unknown compounds are another major challenge because many detected signals cannot immediately be assigned a precise chemical identity. Researchers therefore distinguish between metabolite features, putatively annotated compounds and confidently identified metabolites according to the strength of available evidence. Careful reporting is important because an apparently precise biological interpretation can exceed what the analytical evidence actually supports.
Metabolomics has become an important component of systems biology and multiomics because metabolites represent a downstream expression of biochemical activity. Combining metabolomic data with genomics can connect genetic variants to metabolic consequences, while integration with transcriptomics and proteomics can reveal whether changes occur at the level of gene expression, protein abundance, enzyme modification or downstream chemistry. Computational pathway analysis can organise metabolite changes into biochemical networks, although pathway databases are incomplete and statistical associations do not necessarily establish causal direction.
The field continues to develop through improved mass spectrometers, chromatography, isotope tracing, spatial metabolomics and single cell approaches. Its central contribution is the ability to observe biological systems through their chemical state, providing information that cannot be inferred reliably from DNA sequence or protein abundance alone.
