*3.2.2 Mass defect filter*

Mass defect filter (MDF) is a software-based data filter technique developed for the detection of metabolites using full-scan HR-MS data. In this approach, metabolite ions will be differentiated from matrix ions based on the mass defect value of metabolites from their parent drug. Mass defect is defined as the difference between the exact mass and nominal mass of an element (e.g., 1 H and 14N have an

exact mass of 1.0078 and 14.00307 Da and nominal mass of 1 and 14 Da, therefore the calculated mass defect of <sup>1</sup> H and 14N is +7.8 and + 3.07 mDa, respectively). It is based on the understanding that mass defect values of metabolites fall within a defined narrow window related to that of the parent drug. A narrow mass defect window (40–50 mDa) of MDF removes unwanted signals and causes enrichment of metabolites [91].

Phase I and Phase II metabolites are generally having mass defect values of less than 50 mDa relative to that of the structure of the parent drug. MDF has been applied for the identification of drug metabolites in plasma, urine, feces, bile, and in incubates of liver microsomes and hepatocytes [62, 92, 93]. All the metabolites generated are not structurally similar to the parent drug, some varies slightly (e.g., oxidation), and some show a significant variation (e.g., GSH adduct 68 mDa). If the MDF window is set at ±50 mDa, it excludes all the metabolites which have a mass defect value of more than 50 mDa and if the MDF window is broader, interference ions from the endogenous matrix will be included. So as to avoid it, multiple narrow MDF windows are developed and applied over a certain mass range. Drug filter, substructure filter, and conjugate filter are the commonly used MDF templates. Structures of metabolites that are generated by oxidation or reduction are slightly different from their parent drug structure for such type of metabolites, drug filter template is used. Metabolites that are generated by cleavage of the drug molecule are substructure metabolites of the parent drug compound, for such types of metabolites substructure filters are used. Metabolites that are generated due to conjugation reactions (phase II biotransformation reactions) are called conjugation metabolites, for this type of metabolites conjugate filter templates are used.

#### *3.2.3 Product ion filter (PIF) and NL filter (NLF)*

The mechanism for the identification of metabolites by PIF and NLF is based on the predicted product ion and predicted neutral loss fragmentation, respectively. Both known and unknown metabolites are determined by using these techniques. High-resolution product ion filter (PIF) and high-resolution neutral loss filter (NLF) are highly selective and sensitive techniques, and sometimes these are helpful to determine the trace amounts of unexpected metabolites that are not detected by MDF [94].

PIF is like a precursor ion filter scanning, a data acquisition technique, but PIF is a post-acquisition data mining technique, and metabolites are identified by applying multiple filters. On the other hand, multiple injections are required for the detection of multiple metabolites by precursor ion filter scanning. Likewise, NLF is like a neutral loss scanning; but the difference is avoiding the use of multiple injections because multiple filters are used to identify multiple desired metabolites [94, 95]. PIF and NLF are commonly used for the identification of Phase II metabolites (conjugated metabolites) [85, 96, 97].

#### *3.2.4 Isotope pattern filtering (IPF)*

This technique is useful for the identification of unexpected metabolites which have a distinct isotopic pattern. Metabolite ions that have a distinct isotopic pattern are extracted by applying the filters to full MS scan data. Most of the background peaks are eliminated by isotope pattern filtering (IPF) because many of the endogenous components do not show isotopic patterns [71]. IPF is applicable to the compounds containing distinct natural isotopes (Cl or Br) or synthetically incorporated isotopes (2 H,13C and 15N), or radiolabeled compound (14C) with a distinct isotopic pattern. IPF is a valuable data mining tool for the identification and characterization of conjugated metabolites and reactive metabolites with improved selectivity and sensitivity [98–100].
