Alphabet Projection of Spectra

Area of Science:

• Analytical Chemistry
• Computational Chemistry
• Biochemistry

Background:

• De novo analysis of structured small molecules is crucial in metabolomics and glycomics due to the vast number of possible combinations.
• Current de novo methods are often hindered by the unknown alphabet of mass differences required to link spectral peaks.
• Mass spectrometry is a key technology for analyzing small molecules in biological systems.

Purpose of the Study:

• To propose a method for identifying the alphabet of m/z differences from fragmentation mass spectra.
• To enable de novo analysis when the mass difference alphabet is unknown.
• To develop an efficient approach for finding recurring substructures within the de novo analysis results.

Main Methods:

• A novel algorithm is presented to identify a set of m/z differences from fragmentation spectra.
• This method constructs large, connected graphs by linking intense peaks based on the identified mass differences.
• The study also introduces an efficient technique for substructure discovery in the resulting graphs.

Main Results:

• The proposed method successfully identifies the alphabet of m/z differences from experimental data.
• The approach facilitates the creation of extensive molecular graphs from mass spectrometry data.
• An efficient algorithm for identifying recurring molecular substructures is demonstrated.

Conclusions:

• The developed method overcomes limitations in de novo analysis by automatically determining the mass difference alphabet.
• This facilitates more comprehensive structural elucidation of small molecules in metabolomics and glycomics.
• The efficient substructure discovery component aids in identifying common molecular patterns.