Multivariate Raman mapping for phenotypic characterization in plant tissue sections

Abstract

The pre-processing and analysis of spectrometric and spectroscopic data of plant tissue are important in a wide variety of research areas, such as plant biology, agricultural science, and climate research. The focus of this thesis is the optimized utilization of data from plant tissues, which include matrix-assisted laser desorption/ionization mass spectrometry (MALDI-TOF MS), Raman spectroscopy, and FTIR spectroscopy. The ability to attain a classification using these methods is compared, in particular after combination of the data with each other and with additional chemical and biological information. The discussed are concerned with the investigation and classification within a particular plant species, such as the distinction of samples from different populations, growth conditions, or tissue substructures. The samples comprise grass pollens from large greenhouse experiments and from environmental samples, as well as tissue sections of the species Sorghum bicolor and Cucumis sativus from dedicated physiological and ecological studies. In this way, several results of this work contribute directly to these projects. The data were analyzed by exploratory tools such as principal component analysis and hierarchical cluster analysis, as well as by predictive tools that included partial least square-discriminant analysis and machine learning approaches.Specifically, the results show that combination of the methods with additional plant-related information in a consensus principal component analysis leads to a comprehensive characterization of the samples. This was indicated by the discrimination of pollen from the grass species Poa alpina regarding different populations and environmental conditions. As another application of a multimodal analysis for classification, the combination of FTIR microspectra from individual pollen grains with their Raman mapping data and with MALDI-MS data of pollen extract is discussed. Moreover, as shown for Raman mapping data of tissue sections of Sorghum bicolor, the data of many Raman maps can be combined with other phenotypical data for an extensive insight into tissue biochemical composition. Moreover, some important problems of the individual methods with regard to non-relevant variances in the respective data sets are addressed. In the case of Raman microspectra, a high non-Raman based variance in the data set can be caused by a strong fluorescence background, as is discussed here for the spectra of individual pollen grains. On the other hand, FTIR microspectra of individual pollen grains display scattering artifacts, unless they are obtained from samples that are embedded in paraffin, which, in turn, leads to paraffin signals in the pollen spectra. In MALDI-MS imaging data, the extracts of pollen mixtures can overlap and lead to superposition of unknown origin and extent. Optimized data pre-treatment strategies are discussed for each of these problems. They include advanced baseline correction methods by asymmetric least squares, decomposition by non-negative matrix factorizati...

Deposition of silica in sorghum root endodermis modifies the chemistry of associated lignin

“…Raman maps were pre-processed and analyzed using Matlab R2020b (The MathWorks, Inc, Natick, MA, USA). Data pre-processing steps included the data managing with sorting the spectra and eventually extracting selective spectra corresponding to specific spots or tissues using approaches for multivariate classification ( Liedtke et al., 2021 ). Spectra were interpolated with a distance of 1.8 cm -1 in the range of 350 to 1800 cm- 1 in order to assure equal distances between data points.…”

A new approach to Fourier transform Raman: Identification of haploids in maize (Zea mays)

“…The final spectra for each sample were obtained by the average of 48 scans with a resolution of 4 cm −1 in a spectral range of 200–2400 cm −1 . The spectra were frequency‐calibrated using the spectrum of a mixture of toluene and acetonitrile (Almeida et al, 2010; Liedtke et al, 2021).…”

Physicochemical Understanding of Biomineralization by Molecular Vibrational Spectroscopy: From Mechanism to Nature

“…The acquired tissue needs to be measured for Raman spectroscopy promptly to prevent tissue denaturation, which can lead to inaccurate results. [106,107] If the timely measurement is not possible, the tissue can be fixed and decalcified by organic matter and placed in cryopreservation or directly in liquid nitrogen for preservation. [108][109][110][111][112] It is noteworthy that during the decalcification process, the hydrochloric acid/ethylenediamine tetraacetic acid (HCl/EDTA) mixture removed most of the minerals.…”