Improvement of an EI-MS Peptide Fingerprinting Model

We have been developing a Python-based fragment prediction program which serves to match experimental peptide fragments from Electron Ionization (EI) mass spectra with a generated fragment “fingerprint” produced using known peptide fragmentation mechanisms during EI. EI Mass Spectrometry (EI-MS) has several limitations in regard to its use to analyze peptides. Peptide fragmentation can be helpful in determining a peptide’s initial sequence, but can make it difficult to differentiate between peaks from contaminants and peptides. The goal of this work is to eventually be able to confirm the identity of a synthetic peptide by comparing its mass spectrum with our program-generated “fingerprint”. To confirm current predictions and to identify additional fragmentation mechanisms to refine the prediction model, various dipeptides were synthesized using fluorenylmethyloxycarbonyl (Fmoc)-based solid-phase peptide synthesis (SPPS). They and single amino acids were analyzed using Direct Exposure Probe EI-MS (DEP-EI-MS) to help validate and improve our fragmentation prediction model. In addition to manual analysis of mass spectra and theoretical overlap comparison between predicted fragments, Upset plots were created from the mass spectra as a way to visualize and determine peak matches between spectra. This analysis can identify contaminant peaks and help direct further investigations to identify additional fragmentation mechanisms.