Automatic detection and grouping of MS1 fragmentation in LC-MS data

Importance of the Topic

In-source fragmentation generates unexpected MS1 fragment ions that complicate data interpretation by adding false positive features and obscuring true compound signals. Efficient detection and grouping of these fragments are essential for accurate untargeted LC-MS analysis.

Objectives and Study Overview

This study focused on optimizing data acquisition parameters to enhance the detection and grouping of MS1 fragment ions during untargeted component discovery. Two acquisition strategies were compared to assess their impact on fragment detection and library identification.

Applied Methodology

The workflow combined untargeted feature detection with targeted interrogation of potential in-source fragments using both internal and online spectral libraries. Data acquisition methods included:

• Stepped HCD collision energies of 30%, 50%, and 80% NCE
• Modified stepped HCD including a low-energy step (10%, 50%, 80% NCE)

The software algorithm applied peak quality scoring, isotope and adduct grouping, then searched MS2 spectra against mzCloud to identify low-energy fragments and associate them with co-eluting MS1 features.

Applied Instrumentation

A Thermo Scientific Orbitrap ID-X Tribrid mass spectrometer coupled to a Vanquish Horizon UHPLC system was used. Separation utilized an Accucore phenyl-hexyl column (100×2.1 mm, 2.6 μm) with a water–formic acid and acetonitrile–methanol mobile phase gradient. Data processing was performed with Compound Discoverer 3.4 and mzCloud spectral libraries.

Main Results and Discussion

Implementing the low-energy acquisition step led to a 55% increase in MS1 fragment detection in positive ionization and 16% in negative mode. Fragments such as m/z 388.3203 were only observed when the 10% NCE scan was included. Despite the additional low-energy data, the number and quality of library identifications remained comparable, with a slight improvement in high-confidence matches for some compounds.

Benefits and Practical Applications

The enhanced workflow reduces data complexity by grouping true in-source fragments with their parent ions, improving feature annotation and confidence in untargeted studies. It enables discovery of fragments absent from higher-energy MS2 spectra, facilitating more comprehensive metabolite profiling and unknown compound characterization.

Future Trends and Possibilities

Future work may explore:

• Separate low-energy fragmentation runs to avoid potential trade-offs in library ID
• Integration with data-independent acquisition strategies
• Advanced software algorithms for real-time fragment association

These developments could further streamline MS1 fragment detection without compromising spectral library performance.

Conclusion

Inclusion of a low-energy stepped collision energy significantly improves in-source fragment detection and grouping in LC-MS data with minimal impact on library-based compound identification, offering a robust approach for untargeted analyses.

Reference

Stratton T, Pedišius V. Automatic detection and grouping of MS1 fragmentation in LC-MS data. Thermo Fisher Scientific Application Note, 2025.