Interactive Design and Application of MassQL Queries after Preprocessing for the Annotation of PFAS in LC-TIMS-PASEF data

Importance of the Topic

Per- and polyfluoroalkyl substances (PFAS) represent a class of persistent environmental contaminants with growing regulatory and public health concerns. Rapid, reliable annotation of PFAS in complex samples such as drinking water is critical for monitoring exposure and guiding remediation efforts. Integrating a flexible query language with advanced data processing greatly enhances the capacity to discover both known and novel PFAS candidates in high-dimensional mass spectrometry datasets.

Study Objectives and Overview

This work describes an interactive workflow to design, validate and apply Mass Spectrometry Query Language (MassQL) queries on preprocessed LC-TIMS-PASEF data within the MetaboScape 2023b environment. The main goals are:

• To extract characteristic fragmentation and mobility patterns for PFAS compounds.
• To formalize these patterns as MassQL queries.
• To screen both known and potential new PFAS in drinking water samples.

Methodology and Instrumentation

Samples were analyzed on a Bruker timsTOF Pro equipped with a VIP-HESI source and coupled to an Elute UHPLC system using an Intensity Solo C18 column with a water/methanol gradient containing 5 mM ammonium acetate. Data were acquired in negative ion mode using PASEF, covering m/z 30–1000 and ion mobility 0.5–1.3 V/cm².

Raw data processing employed the T-ReX 4D algorithm in MetaboScape 2023b, generating a feature table. Known PFAS were initially dereplicated via internal target lists and spectral libraries, yielding 975 features. Interactive filtering and manual inspection guided the derivation of fragment and neutral loss rules, which were then expressed as a MassQL query.

Key software tools:

• MassQL: domain-specific language for MS pattern queries.
• MetFrag In Silico Fragmentation: for matching experimental MS/MS spectra.
• CCS-Predict Pro: for collision cross-section prediction.
• SmartFormula with adapted Seven Golden Rules: for molecular formula generation.
• Compound Crawler accessing ChemSpider, PubChem, ChEBI.

Main Results and Discussion

Applying the custom MassQL query to negative-mode features filtered the table down to seven candidates, two of which were previously unreported PFAS analogs. In silico fragmentation of a target compound (1-hydro-pentadecafluoroheptane) showed fragment ions corresponding to C2F5, C3F7 and C4F9 losses within 5 mDa tolerance. Measured CCS for this feature (155.7 Ų) agreed within 1.7% of the predicted value (153.1 Ų), supporting its structural assignment.

This interactive workflow demonstrates a closed-loop process: (1) explore features, (2) derive rules, (3) formalize MassQL query, (4) filter features, (5) validate candidate structures. It accelerates the discovery of both known and novel PFAS in complex matrices.

Benefits and Practical Applications

• Interactive rule derivation reduces trial-and-error in query design.
• Integration of MS/MS fragmentation and CCS prediction enhances annotation confidence.
• Shareable MassQL queries promote reproducibility and community adoption.
• Applicable to routine PFAS screening in environmental and regulatory labs.

Future Trends and Perspectives

Emerging directions include integration of AI-driven spectral interpretation, automated query optimization, expansion to multi-omic data types, and development of public MassQL repositories. Enhanced collaboration among software platforms and community-curated query libraries will further streamline non-targeted contaminant screening workflows.

Conclusion

The presented workflow leverages MetaboScape’s interactive environment and MassQL to enable dynamic design and validation of queries on processed LC-TIMS-PASEF data. Combining feature exploration, in silico tools and predictive CCS accelerates PFAS annotation and the discovery of novel analogs, offering a powerful strategy for environmental monitoring.

Instrumentation

• Bruker timsTOF Pro with VIP-HESI
• Elute UHPLC system
• Intensity Solo C18 column
• MetaboScape 2023b with T-ReX 4D preprocessing
• MassQL query engine
• MetFrag In Silico Fragmentation
• CCS-Predict Pro
• SmartFormula with Seven Golden Rules
• Compound Crawler accessing ChemSpider, PubChem, ChEBI

References

1. Jarmusch AK, et al. A Universal Language for Finding Mass Spectrometry Data Patterns. 2022.
2. Ruttkies C, et al. MetFrag relaunch: incorporating strategies beyond in silico fragmentation. Journal of Cheminformatics. 2016;8:1–16.
3. Kind T, Fiehn O. Seven Golden Rules for heuristic filtering of molecular formulas obtained by accurate mass spectrometry. BMC Bioinformatics. 2007;8:1–20.
4. ChemSpider. www.chemspider.com.
5. PubChem. pubchem.ncbi.nlm.nih.gov.
6. ChEBI. www.ebi.ac.uk/chebi/.