Chemovar Typing of Cannabis Strains with MarkerView™ Software and the SCIEX X500R QTOF System

Importance of the topic

Cannabis consumers often rely on traditional cultivar labels—indica, sativa, hybrid—to predict effects. However, these horticultural classes fail to reflect the underlying chemical diversity of strains. As legal markets expand and regulations tighten, robust analytical methods that characterize chemical fingerprints, or chemovars, have become essential for patient treatment, intellectual property protection, quality control, and detection of adulteration. A chemistry-based classification can more accurately guide cultivation, medical dosing, and product development.

Objectives and overview of the study

This work presents a nontargeted workflow to differentiate seven commercially available cannabis strains based on their comprehensive chemical profiles. Instead of targeting a predefined list of cannabinoids and terpenes, the study employs high-resolution mass spectrometry (HRMS) with SWATH® Acquisition combined with multivariate statistics to discover novel marker compounds. The goals include demonstrating strain clustering by principal component analysis (PCA), identifying unique peaks via t-tests and volcano plots, and tentatively assigning structures using MS/MS libraries and database searches.

Methodology

• Sample preparation: Triplicate extracts of seven strains were prepared in acetonitrile following a modified vMethod™ protocol, then diluted 200× in methanol.
• LC separation: A biphenyl column (150×4.6 mm, 2.6 µm) with a water/ammonium acetate/formic acid and methanol/water mobile phase gradient optimized for cannabinoid and terpene isomers. Flow rate was 0.8 mL/min over a 30 min run.
• MS acquisition: A SCIEX X500R QTOF system with APCI in positive mode collected full-scan MS (50–1000 m/z) and variable-window SWATH® MS/MS using optimized windows for high specificity.
• Data processing: MarkerView™ Software performed PCA and t-test analyses to highlight features distinguishing strain groups without prespecifying targets. Volcano plots visualized fold changes and statistical significance. SCIEX OS Software and FormulaFinder provided MS/MS library searches and empirical formula generation for annotation.

Instrumentation

• SCIEX X500R QTOF system with Turbo V™ Ion Source and APCI probe
• ExionLC™ AD ultrahigh-performance liquid chromatography system
• Phenomenex Kinetex Biphenyl LC column (150×4.6 mm, 2.6 µm)
• MarkerView™ Software for multivariate analysis
• SCIEX OS Software (FormulaFinder, ChemSpider search, MS/MS library)

Main results and discussion

• PCA achieved clear clustering of replicate extractions within each strain, confirming reproducible chemovar signatures and revealing relative chemical similarities.
• T-tests and volcano plots identified features uniquely up- or down-regulated in individual strains. For example, a feature at m/z 313.1794 was enriched in one strain, while a peak at m/z 341.2107 was depleted.
• MS/MS library matching tentatively identified known compounds such as cannabinol and sesquiterpenoids in specific strains.
• For unknown features lacking library spectra, FormulaFinder determined empirical formulas (e.g., C22H30O3), which were searched against ChemSpider. In-silico fragmentation suggested candidates like Myrsinoic Acid, pending confirmation with reference standards.

Benefits and practical applications

• A nontargeted, data-driven approach enables discovery of novel biomarkers beyond common cannabinoids and terpenes.
• SWATH® Acquisition captures comprehensive MS/MS data in a single run with minimal method development.
• Statistical tools streamline feature selection, reducing manual processing of thousands of signals.
• Chemovar classification supports tailored medical dosing, product standardization, and protection of breeding innovations.

Future trends and potential uses

• Integration of genomics and metabolomics to refine chemovar definitions and link genetic markers to chemical phenotypes.
• Development of standardized chemovar databases to enable cross-laboratory comparisons and regulatory compliance.
• Application of machine learning to predict strain properties and consumer effects from large HRMS datasets.
• Adoption of real-time monitoring techniques for in-field chemotyping during cultivation and quality control.

Conclusion

The described nontargeted workflow, combining HRMS SWATH® Acquisition with advanced statistical analysis and database-driven annotation, effectively differentiates cannabis strains by their chemical fingerprints. This chemovar-centric approach enhances understanding of strain variability, supports regulatory and clinical requirements, and opens avenues for discovering novel bioactive markers.

References

• 1. Henry P. Cannabis chemovar classification: terpenes hyper-classes and targeted genetic markers for accurate discrimination of flavours and effects. PeerJ Preprints (2017).
• 2. Quantitation of Oregon List of Pesticides and Cannabinoids in Cannabis Matrices by LC-MS/MS. SCIEX Technical Note RUO-MKT-02-6729-B.
• 3. Comprehensive Cannabis Analysis: Pesticides, Aflatoxins, Terpenes, and High Linear Dynamic Range Potency from One Extract Using One Column and One Solvent System. SCIEX Technical Note RUO-MKT-02-7218-A.
• 4. Elzinga S., Fischedick J., Podkolinski R., Raber J. Cannabinoids and Terpenes as Chemotaxonomic Markers in Cannabis. Natural Products Chemistry & Research, 3(4):181 (2015).
• 5. Improved Data Quality Using Variable Q1 Window Widths in SWATH Acquisition. SCIEX Application Note RUO-MKT-02-2879-C.