Analytical chemistry finds applications in a wide range of disciplines to permit or improve analysis of compounds. Chromatography, mass spectrometry, and chemometrics are important analytical chemistry topics. Related principles are briefly introduced in Chapter 1.
In Chapter 2, the selenamide reagents N-(phenylseleno)phthalimide (NPSP), which selectively derivatizes thiols, was used in protein/peptide structure analysis, because thiol reactivity toward NPSP was able to indicate the chemical environment of the thiol in a peptide and the solvent accessibility of the thiol in a protein. Moreover, the NPSP derivatization was integrated with on-column digestion and electrochemical reduction, to realize rapid bottom-up proteomics. Results demonstrate that the NPSP derivatization protects thiols in bottom-up proteomics and is an effective method for protein structure analysis.
In Chapter 3, basil plants cultivated by organic and conventional farming practices were successfully classified by pattern recognition of their gas chromatography/mass spectrometry (GC/MS) data. Two in-house fuzzy classifiers, i.e., the fuzzy rule-building expert system (FuRES) and the fuzzy optimal associative memory (FOAM), were used to build classification models. Two crisp classifiers, i.e., soft independent modeling by class analogy (SIMCA) and partial least-squares discriminant analysis (PLS-DA), were used as control methods. Prior to data processing, baseline correction and retention time alignment were performed. Classifiers were built with the two-way, the total ion chromatogram (TIC), and the total mass spectrum (TMS) representation of data sets, separately. Bootstrapped Latin partition (BLP) was used as an unbiased evaluation of the classifiers. The established classifiers were used to classify a new validation set collected 2.5 months later with no parametric change to experimental procedure. Results indicate that the FuRES and the FOAM are powerful tools for modeling two-way GC/MS data objects of complex samples. The GC/MS approach coupled with chemometric analysis is demonstrated as a viable method for organic basil authentication.
In Chapter 4, bootstrapped FuRES and bootstrapped t-statistical weight feature selection methods were individually used to select informative features from GC/MS chemical profiles of organic and conventional basil plants. Feature subsets were selected from two-way, TIC, and TMS representations of GC/MS data objects. Four economic classifiers based on the bootstrapped FuRES approach, i.e., e-FOAM, e-FuRES, e-PLS-DA, and e-SIMCA, and four economic classifiers based on the bootstrapped t-weight approach, i.e., e-PLS-DA-t, e-FOAM-t, e-FuRES-t, and e-SIMCA-t, were constructed to be compared with full-size classifiers obtained from the entire GC/MS data objects (i.e., FOAM, FuRES, PLS-DA, and SIMCA). The established economic classifiers were used to classify a new validation set collected 2.5 months later with no parametric change to experimental procedure. Characteristic components in basil extracts were putatively identified. Feature selection may prove valuable as a rapid approach for organic basil authentication.
Two ongoing projects are summarized in Chapter 5 and 6. In Chapter 5, a computer-aided screening method was used for putatively identifying flavone/flavonol glycosides in plants. Two-way data objects of plant extracts were collected on high performance liquid chromatography (HPLC)–diode array detection (DAD)–tandem mass spectrometry (MS/MS). Results indicate that the proposed MATLAB based algorithm is able to facilitate the HPLC–DAD–MS/MS analysis of flavone/flavonol glycosides in plants to a large extent.
In Chapter 6, three baseline correction algorithms based on orthogonal basis, FOAM, and polynomial fitting were compared by using two-way HPLC–MS and GC/MS data objects. The performance of baseline correction was evaluated with respect to the signal-to-noise ratios (SNRs) of major peaks of the HPLC–MS TIC chromatogram and the classification accuracy of FuRES and PLS-DA classifiers constructed by using GC/MS data objects. Results indicate that baseline correction facilitates data analysis of two-way data objects to a large extent. Lastly, the entire dissertation is summarized in Chapter 7.