Naive Bayes classification model for isotopologue detection in LC-HRMS data

Isotopologue identification or removal is a necessary step to reduce the number of features that need be identified in samples analyzed with non-targeted analysis. Currently available approaches rely on either predicted isotopic patterns an arbitrary mass tolerance, requiring information molecular formula instrumental error, respectively. Therefore, Naive Bayes isotopologue classification model was developed does not depend any thresholds information. This uses elemental defects six ratios and successfully isotopologues for both theoretical wastewater influent samples, outperforming one most commonly used (i.e., 1.0033 ?Da difference method - CAMERA). For isotopologues, outperformed “in-house” true positive rate (TPr) 99.0% false (FPr) 1.8% compared TPr 16.2% FPr 0.02%, assuming no error. As model, 99.8% detection (FDr) 0.5%, again performed better than method, 96.3% FDr 4.8%. it can concluded identification, formula.