Development of a Gaussian-Based Alignment Algorithm for the Ultrahigh-Resolution Mass Spectra of Dissolved Organic Matter

The alignment of ultrahigh-resolution mass spectra (UHR-MS)
is
critical to inspect the presence of unique and common peaks across
multiple UHR-MS spectra. However, few attempts have been conducted
to develop an automated alignment method. In this study, a novel automated
alignment algorithm, namely, FTMSCombine, that follows a Gaussian
distribution of mass errors was developed and then integrated with
existing FTMSCalibrate and TRFu algorithms to establish an open-source
analysis platform, namely, FTMSAnalysis, for the UHR-MS analysis of
the dissolved organic matter. The developed FTMSCombine was capable
of automatically aligning peaks across different UHR-MS spectra by
averaging the m/z values of each
peak cluster, although the alignment should be restricted to Fourier
transform ion cyclotron resonance mass spectrometry (FTICR-MS) spectra
collected by instruments under similar conditions. The FTMSCombine
exhibited an insignificant difference in the reproducibility of chemical
formulae but significantly higher mass accuracy than the ICBM-OCEAN.
In addition to improving the overall mass accuracy of the whole UHR-MS
dataset, the FTMSCombine could effectively exclude scatters or noise
peaks using an optional rule that restricts peaks (continuously) detected
in at least a certain number of spectra in the UHR-MS spectra dataset.
The successfully established FTMSAnalysis (freely available in the
Supporting Information of this study) is of great potential in automatically
analyzing UHR-MS spectra for dissolved organic matter (DOM) and will
largely facilitate the elucidation of DOM chemodivesity by UHR-MS
techniques including FTICR-MS.