Statistical Analysis of Photon-Activated Chimeric Tandem Mass Spectra

Photon activation is a powerful means of inducing controlled fragmentation in mass spectrometry, but the simultaneous activation of multiple precursor ions may produce multiplexed tandem mass spectra that are difficult to interpret. In this work, we show that such photon-activated multiple precursor spectra can be decomposed into individual precursor contributions using correlation-based and information-theoretic approaches. We propose a general, data-driven framework for the statistical resolution of multiplexed photon-activated MS/MS spectra based on the statistical treatment of the spectra followed by clustering analysis. This approach allows statistical tandem mass spectra to be obtained without prior knowledge of the precursor identities.