AI-based identification of grain cultivars via non-target mass spectrometry

Abstract

Detection of food fraud and geographical traceability of ingredients is a continually sought goal for government institutions, producers, and consumers. Herein we explore the use of non-target high-resolution mass spectrometry approaches and demonstrate its utility through a particularly challenging case study -to distinguish wheat and spelt cultivars. By employing a data-independent acquisition (DIA) approach for sample measurement, the spectra are of considerable size and complexity. We utilize artificial intelligence (AI) algorithms (artificial neural networks) to evaluate the extensive proteomic footprint of several wheat and spelt cultivars. The AI model thus obtained is used to classify newer varieties of spelt, processed flour, and bread samples. Additionally, we discuss the validation of such a method coupling DIA and AI approaches. The novel framework for method validation enables calculation of precision parameters for facile comparison of the discriminatory power of the method and in the development of a reliable decision rule.

Essential Terminology and Considerations for Validation of Non-targeted Methods

“…More generally, the numerical demarcation of class definition can be based on adulteration level, concentration, contamination, etc. Such numerically delimited class definitions also have been reported elsewhere in the literature [47][48][49][50]. Alternatively, suppose the method tests whether the examined olive oil sample originates from Italy or Turkey -the classes cannot be delimited numerically.…”

“…It can be evaluated that the FPR and FNR are below 5%, when the seed oil adulteration is below 13% and above 17% respectively. On similar lines, a recent study described a preliminary method performance characterization study, using quantitative decision scores (called D scores in the study [47]).…”

Towards a Mathematical Framework for Method Validation and Performance Characteristics of Non-targeted Methods for Food Authenticity

“…More generally, the numerical demarcation of class definition can be based on adulteration level, concentration, contamination, etc. Such numerically delimited class definitions also have been reported elsewhere in the literature [45][46][47][48]. Alternatively, suppose the method tests whether the examined olive oil sample originates from Italy or Turkey -the classes cannot be delimited numerically.…”

“…It can be evaluated that the FPR and FNR are below 5%, when the seed oil adulteration is below 13% and above 17% respectively. On similar lines, a recent study described a preliminary method performance characterization study, using quantitative decision scores (called D scores in the study [45]).…”