Using In Silico Fragmentation to Improve Routine Residue Screening in Complex Matrices

Targeted residue screening requires the use of reference substances in order to identify potential residues. This becomes a difficult issue when using multi-residue methods capable of analyzing several hundreds of analytes. Therefore, the capability of in silico fragmentation based on a structure database (suspect screening) instead of physical reference substances for routine targeted residue screening was investigated. The detection of fragment ions that can be predicted or explained by in silico software was utilized to reduce the number of false positives. These proof of principle experiments were done with a tool that is integrated into a commercial MS vendor instrument operating software (UNIFI) as well as with a platform-independent MS tool (Mass Frontier). A total of 97 analytes belonging to different chemical families were separated by reversed phase liquid chromatography and detected in a data-independent acquisition (DIA) mode using ion mobility hyphenated with quadrupole time of flight mass spectrometry. The instrument was operated in the MSE mode with alternating low and high energy traces. The fragments observed from product ion spectra were investigated using a chopping bond disconnection algorithm and a rule-based algorithm. The bond disconnection algorithm clearly explained more analyte product ions and a greater percentage of the spectral abundance than the rule-based software (92 out of the 97 compounds produced ae1 explainable fragment ions). On the other hand, tests with a complex blank matrix (bovine liver extract) indicated that the chopping algorithm reports significantly more false positive fragments than the rule based software.