Establishing Sufficiency Thresholds for Assessing the Quality of Mass Spectral Data

At present, no quantitative metric is available for determining the sufficiency of information in the mass spectrum for identification of a sample of an unknown compound. Thus, determination of data sufficiency is currently left to the forensic analyst’s interpretation, which may vary with the skill and experience of the analyst as well as the quality of the sample. Developing and validating such a quantitative measure will provide baseline information for forensic laboratories to use in interpretation of mass spectra.
However, there are challenges in meeting this goal. One is that the characteristic mass spectral pattern of a given substance is not constant from very low to very high concentrations. The other is that contamination and/or background noise may also change the mass spectral pattern, and do so differentially depending on the concentration. Identification difficulty also arises from the sparsity of reference standards and resulting library data. This is also an acute problem for novel substances such as synthetic opioids.
Two aspects are identified as critical to developing a quantitative measure of sufficiency: the quality and quantity of information in the mass spectral library and the experience of the analyst in the interpretation of that data. The goal of this project is to factor in these two critical elements when developing sufficiency thresholds for mass spectral data.
By treating a mass spectrum as a pattern, it is possible, both mathematically and statistically, to characterize accurately the spectral pattern variation that is solely attributable to dilution effect and to define limits on the variation in pattern (regardless of source) that allow accurate identification.
Selected known standards of natural opiates, semi-synthetic and synthetic opioids will be assessed to provide insight into the effect of concentrations, adulterants, and mixtures on the quality of mass spectral data. Additionally, we will present varying experienced forensic chemists with a series of questionnaires to test pattern recognition challenges. To evaluate the analyst identification process, analyst will be given spectra constructed to increase variances of opinion to determine the sufficiency of information needed for an identification.
We anticipate that the results and methodology from this project will be applicable to other forensic disciplines that utilize mass spectral data.

Note: This project contains a research and/or development component, as defined in applicable law, and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14).

CA/NCF