Untargeted Analysis of Biospecimens

How does untargeted chemical profiling improve understanding of the impact of environmental exposures on human health?

The targeted lab hubs provide measures of biomarkers arising from exposure to well-recognized pollutants and exogenous chemicals or their metabolites. These panels of biomarkers include polychlorinated biphenyls, pesticides, brominated flame retardants, tobacco smoke, volatile organic chemicals, plasticizers and other chemicals used in commercial products. However, in the United States alone, over 85,000 chemicals are registered with the EPA for manufacture, import and use in commercial products. This is in addition to 40,000 registered pesticide formulations, approximately 100,000 dietary phytochemicals, 5000 chemicals approved for use as inert ingredients and 7500 compounds registered as drugs or food additives with the FDA. Thus, it is not feasible to develop and apply targeted assays to measure such a large number of chemicals using the limited biological material available from population studies. Untargeted chemical profiling techniques have the potential to provide measures of 1000’s to tens of thousands of chemicals in a biological sample. With this information, it is possible to identify unexpected, previously uncharacterized exposures associated with a given health outcome. This complements targeted assays in that measured chemicals do not have to be selected a priori and allows a data-driven approach to understanding how environmental exposures influence children’s health.

What is the difference between targeted and untargeted measures?

Approaches that use targeted assays for chemical measurement are designed to measure the absolute concentration (in terms of mass/volume, i.e. ng/mL plasma or mass/mass, i.e. ng/g lipids) of specific exposure biomarkers. Methods are developed based upon a priori selected chemicals; sample preparation and analytical methods are then developed to maximize sensitivity, selectivity and quantitative reproducibility. Concentrations are provided by comparing intensity of the signal in each study sample to either a calibration curve consisting of additional samples containing a known analyte concentration or isotopically labeled standards added to each sample prior to processing. The resulting measures provide useful information for determining whether an exposure has occurred and to what extent.

Untargeted chemical profiling uses what is referred to as a top-down or data-driven approach. Rather than designing strategies for measuring specific compounds, laboratory hubs within the HHEAR network have designed methodologies that maximize detection of chemical signals while providing semi-quantitative information on levels in a biological sample. Results are provided as a feature table that contains both annotated (correct/probable identity assigned) and raw (defined by unique spectral signature) chemical signals. Based upon the study design, these signals can then be prioritized based upon their association with an outcome of interest or used for screening to determine what exposures are occurring in a population. Since analytes were not selected prior to measuring the samples, there is potential to identify new and previously uncharacterized exposures to exogenous chemicals in human populations.

What chemicals are detected by untargeted platforms?

The exact chemical species measured within the HHEAR network will vary based upon lab hub and instrumental methods used. In general, untargeted platforms will provide measures of both endogenous metabolites and exogenous chemicals. These include central metabolic pathways essential for life and produced within the body. In addition, dietary chemicals, compounds produced from microbes present in the gut and chemicals that are present due to the environment. Examples of measures potentially detected by the HHEAR untargeted lab hubs include:

  • Nutrient and intermediate metabolites: Carbohydrate metabolism, amino acid metabolism, nucleotide metabolism, cofactor metabolism, steroid metabolism, lipid/fatty acid metabolism, terpenoid metabolism, nutrients, additional secondary and amino-group metabolism
  • Dietary and food-related metabolites: Phytochemicals, essential amino acid metabolism, diet-related biomarkers, fatty acid metabolism, supplements
  • Pharmaceuticals: Non-steroidal anti-inflammatory drugs, platelet aggregation inhibitors, analgesics, antitussives, acetaminophen, anticonvulsants, antihistamines, bronchodilators, anesthetics, vasodilators
  • Environmental chemicals: Phosphoester flame retardants, carbamate pesticides, phenylurea pesticides, neonicotinoid insecticides, phthalates, pesticide synergists, thiocarbamates, chloroacetanilide herbicides, organophosphates

What is the metabolome and how does it relate to what is measured by untargeted chemical profiling techniques?

The human metabolome includes all chemicals present in a biological sample with a molecular weight less than 2,000 da, including endogenous biological metabolites, the chemicals from human–environment interaction, and reactants arising from interaction of these compounds with enzymatic and bacterial processes occurring within multiple body components. The metabolome does not include large macromolecules such as proteins, nucleic acids or polymers. Thus, the data provided by the untargeted lab hubs in HHEAR provides measures of both endogenous and exogenous chemical species. By measuring both within a singular analysis framework, it is possible to extend beyond just exposure or bioeffect and use the measures available from untargeted profiling to link exposure to internal dose, biological response and disease (Walker, Go et al. 2016). Biological changes are determined using metabolic pathway maps or enrichment tools (i.e., KEGG, BioCyc, Mummichog), while databases of exogenous chemicals are used for assessing exposures (T3DB, DrugBank, DSSTox, Exposome-Explorer, METLIN).

What type of samples can be measured?

Capabilities exist for most biological samples within the HHEAR network. These include, but are not limited to, blood (serum/plasma/red blood cells), urine, saliva, feces, cord blood, placenta, amniotic fluid, sweat, cerebrospinal fluid, exhaled breath condensate, bronchial lavage fluid and tissues from most human organs. Additional biological samples not mentioned here may have potential to be characterized by the untargeted lab hubs and can be discussed during the laboratory feasibility stage.

What sample quality and quantity are necessary?

The sample quantity will depend on the scientific question but ranges from 10 to 250 µl. Patient fasting, sample collection, processing, the preservative used, and storage can influence the quality and composition of biological specimens. Thus, information related to sample collection, processing, storage temperature, and duration of storage is important when evaluating the suitability of biological specimens for untargeted analyses.

Considerations include number of freeze-thaw cycles, consistency of sample preservatives across all study samples, time to sample processing, and length and temperature of storage. Untargeted lab hubs within the HHEAR network will assist in determining the suitability of biological specimens for specific analyses.

How does HHEAR ensure the quality of its analyses?

Quality control (QC) procedures have been designed and optimized based on instrumental platform, biological matrix, and study design. Well-established protocols, defined QC procedures, use of isotopic standards, analytical replicates, coefficient of variation evaluation, metabolite quality scoring, and replicate analyses of pooled samples are completed to assess data quality. The HHEAR laboratories are participating in international efforts to assess the comparability of untargeted metabolomics methods, with some participating in efforts to assess the comparability for untargeted profiling of environmental chemicals.

How is untargeted metabolomics assessed?

Metabolites in biological fluids (e.g., blood, urine, saliva, feces) or tissue extracts (e.g., biopsy specimens) are measured using high-resolution mass spectrometry (coupled with liquid and gas chromatography) or NMR spectroscopy. In most cases, the results will be generated using high-resolution mass spectrometry due to the need for increased sensitivity to detect low-abundance exposure biomarkers.


Niedzwiecki MM, et al The Exposome: Molecules to Populations. Annu Rev Pharmacol Toxicol. 2018.

Sud M, Fahy E, Cotter D, et al. Metabolomics Workbench: An international repository for metabolomics data and metadata, metabolite standards, protocols, tutorials and training, and analysis tools. Nucleic Acids Research. 2016;44(D1),D463–D470.

Walker DI, Go Y, Liu K, et al, Eds. Population Screening for Biological and Environmental Properties of the Human Metabolic Phenotype: Implications for Personalized Medicine. Metabolic Phenotyping in Personalized and Public Healthcare. 2016; Elsevier.

Walker DI, et al. The Metabolome: A Key Measure for Exposome Research in Epidemiology. Current Epidemiology Reports. 2019;6(2):93-103.