Self-Supervised Peak Prediction for Mass Spectrometry
Self-supervised GPT2-based model for predicting missing peaks in mass spectrometry data.
Proposed, trained, and evaluated a self-supervised GPT2-based plug-in model for mass spectrometry processing, focused on predicting missing peaks in GC-EI-MS spectra. The model leverages the sequential structure of spectral data to fill in missing information, improving downstream molecule identification.
At: Institute of Computer Science, Masaryk University, Brno