Untargeted Metabolomics Service for Unbiased Small Molecule Profiling

What Is Untargeted Metabolomics?

Unlike targeted metabolomics, which quantifies a predefined set of compounds, untargeted metabolomics captures as many detectable metabolites as possible in a single run. This unbiased, high-coverage strategy enables:

• Discovery of unexpected or novel metabolites
• Identification of key biological pathways under different conditions
• Data-driven insights when prior knowledge is limited

Using a combination of LC–MS/MS, GC–MS, dual-mode chromatography, and curated databases, we help researchers reveal hidden biological shifts at the metabolite level.

Why Choose Our Untargeted Metabolomics Service?

Comprehensive & Confident Metabolite Identification

• Detects 7,000+ metabolites, with 600+ MSI Level 1 identifications
• Matches MS1, MS2, and RT against curated databases (NovoMetDB, HMDB, METLIN)
• Rich annotations: CAS, InChIKey, classification, bilingual (EN/CN)
• Supports HILIC + C18 dual-mode for broad polarity coverage
• AI-assisted classification for unknowns

Advanced Instruments & Automated Workflow

• Platforms: Q Exactive™ HF-X, QE Plus, Q-TOF, GC-TOF/MS
• Dual ion modes (ESI±), ≥120,000 resolution, high stability
• Automated sample prep (Gerstel, Beckman)
• Handles 15,000+ samples/year, covering blood, urine, tissues, cells, exosomes, plants, microbes
• Scalable for small or large cohort studies

Flexible Analysis Pipeline & Visual Reporting

End-to-end data support:

• Peak alignment, normalization, batch correction
• Multivariate statistics: PCA, PLS-DA, volcano plots
• Functional enrichment: KEGG pathway mapping, correlation analysis

Compatible with major bioinformatics tools for downstream integration

Reliable QA/QC and Expert Support

• Dual QC system: internal standards + pooled QC samples
• RT drift correction, blank subtraction, batch normalization
• SOP-driven sample handling and instrument tracking

Untargeted Metabolomics Solutions to Support Your Biological Discovery

Whether you're launching a discovery-phase study or evaluating a complex biological system, our service suite is built to support your real-world research goals.

Global Metabolite Profiling

• Wide chemical space coverage: HILIC + C18, ESI+/-, GC–MS
• Compatible with plasma, urine, tissues, feces, saliva, exosomes, plants, microbial and fermentation matrices

Group Comparison & Statistical Analysis

• Differential metabolite screening using t-test, ANOVA, PCA, PLS-DA, OPLS-DA
• Visual outputs: volcano plots, heatmaps, clustering, VIP ranking
• Supports biomarker screening, phenotypic discrimination, and mechanism studies

Biological Pathway Interpretation

• Metabolite mapping to KEGG, HMDB, and Reactome
• Enrichment analysis, pathway topology, class-based categorization
• Outputs include pathway impact plots and integrated functional summaries

Unknown Feature Prioritization

• In-silico MS/MS matching and chemical class prediction
• Molecular networking for structural inference
• Feature ranking for future structure validation or targeted quantification

Optional: Targeted Panel Development

• Build custom quantitative panels based on untargeted results
• Select high-priority markers for validation
• Support for biomarker assay design and follow-up quantification

Untargeted Metabolomics Analytical Setup: Instruments, Coverage & Throughput

Thermo Fisher Q Exactive (Figure from Thermo Fisher)

SCIEX Triple Quad™ 6500+ (Figure from Sciex)

Agilent 1260 Infinity II HPLC (Figure from Agilent)

Waters ACQUITY UPLC System (Figure from Waters)

Untargeted Metabolomics Analysis Workflow — A Step-by-Step Guide

1

Scoping & Experimental Design

Define biological contrasts, randomization, replicates, and covariates; select RP/HILIC and ion modes; plan QC cadence.

2

Sample Preparation

Protein precipitation and biphasic extraction options; inclusion of internal standards aligned to matrix.

3

UHPLC-HRMS Acquisition

Orthogonal methods in positive/negative ESI; DDA/DIA for MS/MS depth; blanks and QCs run at defined intervals.

4

Data Processing

Peak detection, deconvolution, retention-time alignment, feature consolidation, adduct/fragment de-isotoping, blank subtraction, normalization.

5

Annotation & Curation

Library matching with mz/RT/MSMS tolerances; in-silico scoring; chemical class assignment; MSI-level tagging.

6

Statistics & Pathway Biology

Unsupervised structure (PCA, clustering), supervised modeling (with cross-validation), univariate testing with multiple-testing control, pathway enrichment & network mapping.

This Service Is Ideal For

• Hypothesis-free discovery across phenotypes
• Early-phase biomarker screening
• Dynamic metabolic response analysis (e.g., drug dosing, stress, infection)
• Microbiome, nutrition, or host–pathogen interaction studies
• Untargeted profiling before targeted panel development

How to Prepare and Submit Samples for Untargeted Metabolomics

Notes

• Avoid using glass or foil containers—use screw-cap centrifuge tubes only.
• Label all samples clearly and consistently (e.g., Group-A1, Control-3).
• For targeted metabolomics or flux analysis, contact us for specific instructions.
• A backup sample is recommended when possible.
• For unlisted sample types, please consult our technical team.

Metabolomics Sample Submission Guidelines

To ensure reliable and high-quality metabolomics results, proper sample handling is essential. We've prepared a comprehensive submission guide detailing:

• Sample types, volumes & preparation steps
• Storage & shipping recommendations
• Precautions for special or targeted analysis
• Biological replicate suggestions

Get the Guide

Deliverables: What You Receive from Untargeted Metabolomics Analysis

• Raw and Preprocessed Data: Includes original mass spectrometry files (e.g., Thermo RAW) and converted open formats (mzML, mzXML, CSV), along with cleaned and normalized outputs.
• Feature Table: A structured table of aligned metabolite features with m/z, retention time, intensity values, and annotation level (when available).
• Metabolite Annotation Summary: Identification results with confidence levels (MSI standards), matched database entries (e.g., HMDB, METLIN), and relevant compound metadata.
• Quality Control Overview: Comprehensive QC documentation, including results from blanks, pooled QCs, system performance checks, and batch correction details.
• Statistical Analysis Outputs: Key exploratory analyses such as PCA, PLS-DA, volcano plots, and clustering — supporting pattern discovery and sample differentiation.
• Final Analysis Report (PDF): A complete, structured report covering the workflow, methodologies, key results, interpretation highlights, and visual figures.

Overlay of TIC profiles from representative samples, illustrating signal stability and reproducibility across groups during LC-MS acquisition.

Volcano plot highlighting significantly up- and downregulated metabolites.

Enriched metabolic pathways ranked by significance (FDR).

PCA plots before and after normalization. QC samples cluster tightly post-normalization.

Untargeted metabolomics yields insight into ALS disease mechanisms

Background

Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disorder, and understanding its metabolic profile can offer insights into disease mechanisms and identify novel therapeutic opportunities. Metabolomic analyses provide a comprehensive view of endogenous and exogenous influences on ALS, allowing the identification of biomarkers and potential drug targets.

Samples:

The study included 125 ALS and 71 control participants, matched for key demographics. ALS cases represented a typical population with median diagnostic age, symptom onset-to-diagnosis interval, and distribution across onset segments. Metabolite profiling involved the identification of 1051 metabolites, with 144 excluded due to high missingness. A total of 899 metabolites were used for downstream analysis.

Technical Methods

Metabolite Profiling: 1051 metabolites were identified, and 144 with >60% missingness were excluded.

Drug Metabolite Exclusion: Eight drug metabolites were removed from analysis due to weak correlations and high missingness.

Differential Analysis: Wilcoxon rank-sum tests identified 303 significant metabolites. Logistic regression models, adjusted for sex, age, and BMI, further identified 300 metabolites. Partial Least Squares Discriminant Analysis (PLS-DA) and Group Lasso methods were also employed.

Pathway Enrichment Analysis: Identified significantly over-represented sub-pathways among differential metabolites.

Machine Learning Classification: Seven algorithms were applied to predict ALS cases using 259 Group Lasso-selected metabolites.

Results

Differential Metabolites: 303 metabolites were identified as differentially expressed in ALS, with overlaps across various analytical methods.

Pathway Enrichment: Shared sub-pathways, including 'sphingomyelins,' 'ceramides,' 'benzoate metabolism,' and 'fatty acid metabolism,' were enriched across different models. Group Lasso uniquely identified 'diacylglycerol,' 'chemical,' and other sub-pathways.

Metabolite Correlations: Interconnections between significant metabolites and their sub-pathways were visualized, revealing associations in sphingolipid metabolism, polyamine metabolism, and more.

Diagnostic Potential: Machine learning models using 259 Group Lasso-selected metabolites demonstrated high diagnostic potential, with Receiver Operating Characteristic (ROC) analysis showing an AUC of 0.98.

Partial least squares-discriminant analysis (PLS-DA) analysis of amyotrophic lateral sclerosis (ALS) cases versus controls. (A) PLS-DA score plot of ALS cases (red) versus controls (blue); each dot represents an individual subject. (B) The variable importance in projection (VIP) score plot of the top 30 PLS-DA metabolites, which most significantly separate cases from controls.

Pathway enrichment of adjusted logistic regression-selected, partial least squares-discriminant analysis (PLS-DA)-selected and group lassoselected metabolites

Metabolite correlation analysis of group lasso-selected metabolites

Reference

1. Goutman, Stephen A., et al. "Untargeted metabolomics yields insight into ALS disease mechanisms." Journal of Neurology, Neurosurgery & Psychiatry 91.12 (2020): 1329-1338.