In this work, principal component analysis (PCA) was utilized to analyze laser-induced breakdown spectroscopy (LIBS) signals of the extracted chicken fat, lamb fat, beef fat, and lard froze using two different freezing methods. The frozen samples were ablated using a neodymium-doped yttrium aluminum garnet (Nd:YAG) laser with a wavelength of 1064 nm, 170 mJ pulse energy, and 6 ns pulse duration to produce plasma on target surfaces. The samples were ablated using 30-60 shots of the laser beam at different spots. Stronger LIBS signals from the extracted chicken fat and lamb fat were obtained with liquid nitrogen (LN2) method. However, LIBS signals obtained from the freezer freezing method were found to be stronger for extracted beef fat and lard. The PCA was then used to visualize the LIBS spectra of extracted animal fats into a score plot. Data points of each extracted animal fat were divided into three groups representing LIBS spectra collected at the early, middle, and end part of the ablation process. The score plot revealed that the data points of the three groups of frozen extracted animal fats using the LN2 method were more closely clustered than those frozen in the freezer. Good discrimination with 97% of the variance was achieved between the extracted chicken fat, lamb fat, beef fat, and lard using the LN2 method in the three-dimensional score plot. LIBS signals of the extracted animal fats produced from the LN2 method were found to be more stable than those from the freezer method.
In this study, the performance of a thermoelectric cooler (TEC) as a simple and easy-to-assemble freezing instrument has been evaluated. Experiments were carried out using samples with different viscosity ranging from 44.07 to 16 965.80 MPa. The analysis of sodium component of the samples by direct laser irradiation of frozen samples showed emission enhancement and higher signal-to-noise ratio compared to that of liquids. This work also focused on using chemometrics methods such as principal component analysis (PCA) to compare the principal component score separation and clustering pattern between frozen and liquid samples. The PCA was constructed by dividing the samples into two different categories: (i) type (paste, cream, gel, and oil) and (ii) viscosity (more than and less than 10 000 MPa). The frozen samples showed a more established separation and clustering compared to that acquired from the liquid samples. However, poorer clustering pattern of some frozen samples could be due to the heat transfer during laser-sample interaction inducing surface melting and splashing. The average laser-induced breakdown spectroscopy (LIBS) spectra were taken at as many different surface areas as possible to ensure the sample surface always maintain similar freezing temperature. This work showed that the TEC pre-treatment method had improved the LIBS measurement of the liquid samples.
Protein drugs may encounter conformational perturbations during the formulation processing of lipid-based solid dosage forms. In aqueous protein solutions, attenuated total reflection Fourier transform infrared (ATR FT-IR) spectroscopy can investigate these conformational changes following the subtraction of spectral interference of solvent with protein amide I bands. However, in solid dosage forms, the possible spectral contribution of lipid carriers to protein amide I band may be an obstacle to determine conformational alterations. The objective of this study was to develop an ATR FT-IR spectroscopic method for the analysis of protein secondary structure embedded in solid lipid matrices. Bovine serum albumin (BSA) was chosen as a model protein, while Precirol AT05 (glycerol palmitostearate, melting point 58 ℃) was employed as the model lipid matrix. Bovine serum albumin was incorporated into lipid using physical mixing, melting and mixing, or wet granulation mixing methods. Attenuated total reflection FT-IR spectroscopy and size exclusion chromatography (SEC) were performed for the analysis of BSA secondary structure and its dissolution in aqueous media, respectively. The results showed significant interference of Precirol ATO5 with BSA amide I band which was subtracted up to 90% w/w lipid content to analyze BSA secondary structure. In addition, ATR FT-IR spectroscopy also detected thermally denatured BSA solid alone and in the presence of lipid matrix indicating its suitability for the detection of denatured protein solids in lipid matrices. Despite being in the solid state, conformational changes occurred to BSA upon incorporation into solid lipid matrices. However, the extent of these conformational alterations was found to be dependent on the mixing method employed as indicated by area overlap calculations. For instance, the melting and mixing method imparted negligible effect on BSA secondary structure, whereas the wet granulation mixing method promoted more changes. Size exclusion chromatography analysis depicted the complete dissolution of BSA in the aqueous media employed in the wet granulation method. In conclusion, an ATR FT-IR spectroscopic method was successfully developed to investigate BSA secondary structure in solid lipid matrices following the subtraction of lipid spectral interference. The ATR FT-IR spectroscopy could further be applied to investigate the secondary structure perturbations of therapeutic proteins during their formulation development.
Nutrient solution plays an essential role in providing macronutrients to hydroponic plants. Determining nitrogen in the form of nitrate is crucial, as either a deficient or excessive supply of nitrate ions may reduce the plant yield or lead to environmental pollution. This work aims to evaluate the performance of feature reduction techniques and conventional machine learning (ML) algorithms in determining nitrate concentration levels. Two features reduction techniques, linear discriminant analysis (LDA) and principal component analysis (PCA), and seven ML algorithms, for example, k-nearest neighbors (KNN), support vector machine, decision trees, naïve bayes, random forest (RF), gradient boosting, and extreme gradient boosting, were evaluated using a high-dimensional spectroscopic dataset containing measured nitrate-nitrite mixed solution absorbance data. Despite the limited and uneven number of samples per class, this study demonstrated that PCA outperformed LDA on the high-dimensional spectroscopic dataset. The classification accuracy of ML algorithms combined with PCA ranged from 92.7% to 99.8%, whereas the classification accuracy of ML algorithms combined with LDA ranged from 80.7% to 87.6%. The PCA with the RF algorithm exhibited the best performance with 99.8% accuracy.
Obesity is strongly linked with increased risk and poorer prognosis of endometrial cancer (EC). Cancer-associated fibroblasts (CAFs) are activated fibroblasts that form a large component of the tumor microenvironment and undergo metabolic reprogramming to provide critical metabolites for tumor growth. However, it is still unknown how obesity, characterized by a surplus of free fatty acids drives the modifications of CAFs lipid metabolism which may provide the mechanistic link between obesity and EC progression. The present study aims to evaluate the utility of Raman spectroscopy, an emerging nondestructive analytical tool to detect signature changes in lipid metabolites of CAFs from EC patients with varying body mass index. We established primary cultures of fibroblasts from human EC tissues, and CAFs of overweight/obese and nonobese women using antibody-conjugated magnetic beads isolation. These homogeneous fibroblast cultures expressed fibroblast markers, including α-smooth muscle actin and vimentin. Analysis was made in the Raman spectra region best associated with cancer progression biochemical changes in lipids (600-1800 cm-1 and 2800-3200 cm-1). Direct band analysis and ratiometric analysis were conducted to extract information from the Raman spectrum. Present results demonstrated minor shifts in the CH2 symmetric stretch of lipids at 2879 cm-1 and CH3 asymmetric stretching from protein at 2932 cm-1 in the overweight/obese CAFS compared to nonobese CAFs, indicating increased lipid content and a higher degree of lipid saturation. Principal component analysis showed that CAFs from overweight/obese and nonobese EC patients can be clearly distinguished indicating the capability of Raman spectroscopy to detect changes in biochemical components. Our results suggest Raman spectroscopy supported by chemometric analysis is a reliable technique for characterizing metabolic changes in clinical samples, providing an insight into obesity-driven alteration in CAFs, a critical stromal component during EC tumorigenesis.