FindPFΔS: Non-Target Screening for PFAS─Comprehensive Data Mining for MS² Fragment Mass Differences

Area of Science:

• Environmental Chemistry
• Analytical Chemistry
• Mass Spectrometry

Background:

• Limited availability of analytical reference standards hinders the identification of unknown per- and polyfluoroalkyl substances (PFAS).
• Non-target screening using high-resolution mass spectrometry (HRMS) is crucial for detecting emerging contaminants like PFAS and their transformation products (TPs).

Purpose of the Study:

• To develop and optimize an open-source algorithm, FindPFΔS (Find PolyFluoroDeltas), for identifying unknown PFAS.
• To leverage distinct fragment mass differences in MS/MS data for enhanced PFAS detection.

Main Methods:

• Developed a vendor-independent, Python-based algorithm (FindPFΔS) to analyze fragment mass differences in .ms2 files.
• Optimized the algorithm using PFAS standards and characterized environmental samples (paper, soil) with iterative data-dependent acquisition.
• Systematically evaluated the influence of mass tolerance and intensity thresholds on identification efficiency and false positive rates using HR-MS2 spectra from MassBank.

Main Results:

• Identified key fragment differences (e.g., Δ(CF2), ΔHF, ΔCF3) relevant for PFAS identification based on collision energy.
• Achieved 94% identification rate for PFAS standards (36/38 compounds across 10 classes) in a mixed standard.
• Successfully identified unknown PFAS homologues in paper extracts, demonstrating the algorithm's applicability to real-world samples.

Conclusions:

• The FindPFΔS algorithm provides a promising, broadly applicable approach for PFAS identification using fragment mass differences.
• This method enhances non-target screening capabilities, addressing the challenge of limited analytical standards for PFAS.
• The open-source algorithm is freely available, promoting wider adoption in environmental analysis and research.