Unknown Identifications Using NIST EI Hybrid Search

Significance of the Topic

Accurate identification of unknown compounds remains a cornerstone in forensic, environmental, pharmaceutical, and industrial quality-control laboratories. Traditional mass spectral library searches often fail when target analytes are absent or only structurally related compounds are present. The NIST EI Hybrid Search algorithm bridges this gap by combining fragment-ion and neutral-loss matching, thereby extending library scope and improving confidence in identifying both known and novel “known unknowns.”

Objectives and Overview of the Study

This work, presented at the Forensic@NIST 2022 Workshop, aimed to demonstrate how the hybrid search approach enhances mass spectral identification workflows. Key objectives were to illustrate the algorithm’s principles, showcase its application in GC-EI-MS and LC-MS/MS data, and highlight its integration with complementary analytical techniques and spectraless databases for robust unknown characterization.

Methodology

The hybrid search process involves two sequential comparisons: an initial standard identity search against library fragment-ion spectra, followed by a neutral-loss comparison adjusted by a single DeltaMass value. The combined match score increases the likelihood of detecting analogues that differ by a well-defined structural modification. DeltaMass values, representing nominal mass differences between query and reference compounds, are catalogued to guide interpretation of observed mass shifts.

Used Instrumentation

• Gas chromatograph coupled to electron ionization mass spectrometer (GC-EI-MS)
• Liquid chromatography with tandem mass spectrometry (LC-MS/MS), including high-resolution accurate mass analyzers
• DART source for accurate mass determination
• Chemical ionization GC-MS for molecular ion observation
• High-field and low-field NMR spectrometers (1D/2D H, C, P, F)
• Infrared (IR) spectroscopy
• Derivatization tools (e.g., silylation, methyl esterification)

Major Results and Discussion

Application of the hybrid search routinely yielded top match factors above 800, even when standard searches failed (<600). In GC-EI-MS examples, unknowns differing by a DeltaMass of 18 or 54 were rapidly linked to fluoro- or alkyl-modified analogues. A forensic case study identified a novel PCP-related compound (C21H29N) by combining hybrid search output (Δmass 54), DART accurate mass data, and substructure proposals from known illicit syntheses. Subsequent NMR validation confirmed the structure. A parallel LC-MS/MS demonstration further underscored the method’s adaptability across ionization techniques.

Benefits and Practical Application

• Extends identification capability beyond library entries, capturing structural analogues
• Accelerates forensic and regulatory workflows by suggesting substructures
• Supports mixture analysis when combined with NMR quantitation and spectraless database searches (ChemSpider, CAS Registry)
• Facilitates corporate user-library growth and continuous updates
• Integrates with fragmentation‐interpretation software (NIST MS Interpreter) for mechanistic insights

Future Trends and Potential Uses

Ongoing developments include expanding hybrid search to high-resolution MS/MS libraries, automating DeltaMass annotation through machine learning, and integrating predictive fragmentation models. Real-time implementation in screening platforms and deeper coupling with spectraless chemical registries will further streamline unknown identifications in complex matrices.

Conclusion

The NIST EI Hybrid Search algorithm represents a significant advancement in mass spectral library searching, enabling reliable identification of structurally related unknowns. When combined with accurate mass data, NMR, IR, and spectraless databases, it forms a versatile toolkit for modern analytical laboratories tackling challenging identification problems.

Reference

1. Moorthy A., Wallace W., Kearsley A. J., Tchekhovskoi D., Stein S., “Combining Fragment-Ion and Neutral-Loss Matching during Mass Spectral Library Searching: A New General Purpose Algorithm Applicable to Illicit Drug Identification,” Analytical Chemistry, 2017, 89(24):13261–13268.
2. Shulgin A. T., Mac Lean D. E., “Illicit Synthesis of Phencyclidine (PCP) and Several of Its Analogs,” Clinical Toxicology, 1976, 9(4):553–560.