Moving beyond ddMS2: Improving annotation confidence in untargeted metabolomics using higher resolution MS and parallel ion trap experiments

Significance of the Topic

The accurate identification of small molecules in untargeted metabolomics is critical for understanding biochemical pathways, biomarker discovery and quality control in pharmaceutical, environmental and clinical research. Traditional single-spectrum approaches often struggle with isobaric species, isomers and high m/z complexity. By harnessing higher-resolution mass measurements and complementary ion trap experiments, annotation confidence can be substantially enhanced, leading to more reliable metabolite profiles and deeper biological insights.

Objectives and Study Overview

This study aimed to evaluate how increased Orbitrap resolution combined with parallel ion trap fragmentation experiments improves elemental formula prediction and structural annotation in untargeted metabolomics. Key goals included:

• Resolving isotopic fine patterns at multiple resolutions (60k to 1 million) to reduce formula ambiguity
• Incorporating data-dependent MS2, MS3 and breakdown curve scans in parallel ion trap acquisitions
• Assessing the impact of these strategies on annotation confidence across a broad range of endogenous plasma metabolites

Methodology

A certified human plasma extract (NIST SRM 1950) was prepared in methanol:water and analyzed by reversed-phase UHPLC with a water–methanol gradient containing 0.1 percent formic acid. Mass spectrometric data were acquired on an Orbitrap IQ-X Tribrid instrument at five resolving powers (60 000, 120 000, 240 000, 500 000 and 1 000 000) in MS1 mode. Fragmentation experiments employed an intelligent AcquireX workflow to collect:

• Data-dependent MS2 scans
• MS2 with targeted MS3 (ddMS3)
• MS2 with stepped and individual collision energy breakdown curves

Data processing used isotopic pattern prediction and isobaric species enumeration workflows in Compound Discoverer alongside mzCloud library matching and FISh scoring to evaluate annotation confidence and structural hypotheses.

Used Instrumentation

• Thermo Scientific Orbitrap IQ-X Tribrid mass spectrometer
• Thermo Scientific Vanquish Horizon UHPLC system with Hypersil GOLD column
• Thermo Scientific AcquireX intelligent data acquisition
• Compound Discoverer, Freestyle and Mass Frontier software for data analysis

Key Results and Discussion

At higher resolution settings, isotopic fine patterns became more distinct, particularly for m/z values above 400 where the number of plausible elemental formulas at 1 ppm mass error decreased significantly. When comparing features over m/z 400 at 60k versus 1 M resolution, 85 percent of predicted compositions changed, demonstrating the resolving power’s impact on formula assignment. Parallel MS3 data enabled additional spectral criteria to distinguish isomeric species with nearly identical MS2 spectra. Breakdown curves collected at individual collision energies further enhanced discrimination by revealing unique fragment intensity trends, as shown in comparisons of theophylline versus paraxanthine and acetaminophen versus 2-acetamidophenol.

Benefits and Practical Applications

The combined high-resolution and parallel ion trap approach offers:

• Greater confidence in elemental formula predictions through resolved isotopic peaks
• Improved structural annotation for unknowns by leveraging MS3 and breakdown curve data
• Efficient use of instrument cycle time by collecting complementary scans in parallel
• Enhanced library matching performance for mzCloud and in silico fragment databases via FISh scoring

Future Trends and Applications

Emerging directions include:

• Integration of ultra-high-resolution MS data with machine learning models for automated annotation
• Expansion of spectral libraries to encompass MSn and energy-resolved fragmentation patterns
• Real-time adaptive acquisition strategies that tailor collision energies based on preliminary spectral features
• Coupling high-resolution metabolomics with ion mobility and microfluidics for multidimensional compound characterization

Conclusion

This work demonstrates that increasing Orbitrap resolution and incorporating parallel ion trap MS3 and breakdown curve experiments significantly boosts annotation confidence in untargeted metabolomics. The enhanced isotopic details and complementary fragmentation data reduce formula ambiguity and enable more reliable structural assignments, paving the way for more comprehensive metabolite profiling in diverse applications.

References

1. Bills B, Yedla S, Deshpande R, Amer B, Tautenhahn R, Bird S, Zabrouskov V. Moving beyond ddMS Improving annotation confidence in untargeted metabolomics using higher resolution MS and parallel ion trap experiments. Thermo Fisher Scientific; 2025.