Statistical Evaluation of Mass Spectral Data for Seized Drug Identification

Description:
The increased emergence and use of novel psychoactive substances has resulted in an increasing need to distinguish and definitively identify structurally similar analogs, including positional isomers. Such distinction is typically based on visual assessment of electron-ionization mass spectra, which is challenging given the structural and, hence, spectral, similarity of the analogs. Our work in this area focuses on the development and validation of a method to statistically compare two mass spectra and thereby increase objectivity in such identifications.
The developed method is based on an unequal variance t-test, which is performed at all m/z values in the scan range of the two spectra being compared. The null hypothesis states that the difference in ion intensity at that m/z value is equal to zero, while the alternative hypothesis states that the difference is not equal to zero. Spectra are considered statistically equivalent if the null hypothesis is accepted at all m/z values and statistically distinct if the null hypothesis is not accepted for at least one m/z value. Our previous work described development of the method and demonstrated statistical association and discrimination of a range of seized drugs, including positional isomers.
In this presentation, our current work to further validate the method will be described, with a focus on synthetic cannabinoids, synthetic cathinones, and fentanyl analogs. The effect of sample concentration on successful association and discrimination will be discussed, identifying the ions responsible for discrimination of structurally similar compounds. Further, comparison of spectra collected on different instruments and over several months will be presented, along with discussion of those ions that are reliable and consistent in discriminating these compounds. Finally, recommendations will be made for appropriate implementation of the statistical comparison method in forensic laboratories.

Speaker: Ruth Smith - Michigan State University