Multivariate Analysis of Variance to Identify Instrumental Parameters Affecting Tandem Mass Spectra Reproducibility

Jacob R. King* and Glen P. Jackson
C. Eugene Bennett Department of Chemistry, West Virginia University, Morgantown, WV 26506

Presentation No.: 107

Assigned Category (Presentation Format): Physical Sciences (Poster Presentations)

Student’s Major: Chemistry

Standard methods for compound identification use algorithms to match an unknown mass spectrum to one in a database. Current algorithms are limited in their success because uncontrolled fluctuations in instrumental parameters create variance in peak abundances in mass spectra, and these spectral differences can make it difficult to distinguish structurally similar compounds. Therefore, our ability to understand the sources of variance seen among replicate mass spectra can lead to improved instrument control or better algorithms. To further understand the causes of variance, three sets of replicate spectra were collected in a five-factor, three-level full factorial set of experiments on an Agilent 6538 UHD QTOF, using cocaine as a model compound. Two notable cocaine fragments at m/z 82 and m/z 182, representing the low and high mass-to-charge ions, respectively, were monitored. Results showed that the collision energy and fragmentor voltage were responsible for the most variation in the recorded abundances of the ions, and skimmer voltage and capillary voltage were both responsible for a moderate amount of variance. The drying gas temperature was responsible for the least variance. This data should help to inform the community about the relative influences of mass spectral parameters.

Funding: NSF CHE-1852369

Program/mechanism supporting research/creative efforts: the WVU Chemistry REU (Brian Popp & Michelle Richards-Babb)