Optimal analysis of complex protein mass spectra

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Biomarker Discovery

Background:

• Mass spectrometry often produces complex spectra with numerous peaks, exceeding the number of actual molecular species.
• Physical and chemical phenomena can lead to spectral complexity, complicating data interpretation.

Purpose of the Study:

• To develop an advanced analysis approach for reliable estimation of molecule masses and abundances from complex mass spectra.
• To reduce the complexity of mass spectra by linking peaks and identifying underlying molecular species.

Main Methods:

• A peak-linking strategy was employed to connect related peaks within and between spectra.
• This method allows for the accurate positioning of multiple-charge peaks.
• Complex spectral regions were deconvoluted using information from simpler, higher-resolution regions.

Main Results:

• The analysis successfully modeled 29,952 peaks across 64 spectra using only 39 location and 1 shape parameter.
• A significant reduction in the number of observed peaks to a limited set of molecular species was achieved.
• The approach demonstrated the ability to locate multiple-charge peaks and deconvolute overlapping peaks.

Conclusions:

• The developed peak-linking method effectively reduces spectral complexity, simplifying mass spectrometry data.
• This reduction in complexity is expected to significantly enhance the power of biomarker discovery by reducing statistical test multiplicity.