An enlarged data base of electron-ionization mass spectra

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Computational Chemistry

Background:

• Reference mass spectral databases are crucial for compound identification.
• Previous versions of the database had limitations in size and accuracy.
• The Probability Based Matching (PBM) and Quality Index (QI) algorithms aid spectral analysis.

Purpose of the Study:

• To report the expansion and improvement of a computer-searchable reference mass spectra database.
• To validate the enhanced database's accuracy and reliability for spectral matching.
• To assess the impact of increased data variety and improved algorithms on matching accuracy.

Main Methods:

• Database expansion by 76% to 139,859 spectra from 118,144 compounds.
• Error checking using Probability Based Matching (PBM), Quality Index (QI) algorithms, and human inspection.
• Incorporation of an improved QI algorithm based on Terwilliger's suggestion for saturated spectra.

Main Results:

• The database now contains 139,859 spectra of 118,144 compounds, with an average of 53 peaks per spectrum.
• The number of unique elemental compositions increased by 64%.
• Incorrect matches using PBM were reduced by 67% compared to the original database; adding spectra of the same compound under different conditions reduced incorrect matches by 42%.

Conclusions:

• The expanded and refined mass spectral database offers significantly improved accuracy and reliability for compound identification.
• Enhanced data variety and algorithmic improvements substantially reduce the probability of incorrect spectral matches.
• The updated database represents a valuable resource for analytical chemists and researchers utilizing mass spectrometry.