Improvement of drug identification in urine by LC-QqTOF using a probability-based library search algorithm

Area of Science:

• Analytical Chemistry
• Biochemistry
• Forensic Science

Background:

• Compound identification commonly relies on comparing mass spectra to spectral libraries.
• The accuracy of this identification is influenced by the spectral library, test spectrum quality, and the search algorithm used.

Purpose of the Study:

• To introduce and evaluate a redesigned probability-based library search algorithm (ProLS).
• To compare ProLS performance against existing algorithms (NIST, LV/MV) for drug identification in human urine samples.

Main Methods:

• Mass spectrometry (quadrupole-time of flight) was used to analyze human urine samples for drugs.
• The ProLS algorithm was developed and compared with AMDIS from NIST and LibraryView/MasterView algorithms.
• Spectral data was searched against an in-house spectral library.

Main Results:

• ProLS demonstrated superior efficiency in drug detection compared to NIST and LV/MV.
• ProLS exhibited an improved scoring profile, yielding lower match scores for absent compounds.
• Enhanced scoring accuracy has the potential to decrease manual data review time for analysts.

Conclusions:

• The redesigned ProLS algorithm significantly impacts the accuracy of small molecule identification in biological matrices.
• Improved search algorithms can enhance the overall utility of bioanalytical methods.
• This work highlights the underappreciated role of search algorithms in mass spectrometry-based identification.