Quantcast

Non-Integer Root Transformations for Preprocessing Nano-Electrospray Ionization High-Resolution Mass Spectra for the Classification of Cannabis.

Research paper by Yue Y Tang, Peter de Boves PB Harrington

Indexed on: 20 Dec '18Published on: 20 Dec '18Published in: Analytical Chemistry



Abstract

Typically, for measurements with high dynamic range, the range is reduced by using the square root transform. By using non-integer roots coupled with systematic experimental design, improvements to the measurements may be obtained. The effect of using non-integer root transformation was evaluated using high-resolution mass spectrometry (HRMS) combined with nano-electrospray ionization (Nano-ESI) to differentiate 23 samples of Cannabis. The mass spectra were evaluated and classified using different mass resolving powers and non-integer root transformations. Classification was achieved by super partial least squares discriminant analysis (sPLS-DA), support vector machine (SVM), and SVM classification tree type entropy (SVMTreeH). The 2.5 root transformation gave the best overall performance at different resolving powers for chemical profiling from a multilevel factorial experimental design using 2 factors and more than 4 levels. Response surface modeling using a cubic polynomial model of the sPLS-DA prediction accuracies yielded optima at 0.005 for resolving power and 2.3 for the root transformation. Root transformation is an important spectral preprocessing tool for decreasing the dynamic range, so that the relative variance of smaller but more important features may be inflated.