Discrimination and prediction of Qingzhuan tea storage year using quantitative chemical profile combined with multivariate analysis: Advantages of MRMHR based targeted quantification metabolomics

The duration of storage significantly influences the quality and market value of Qingzhuan tea (QZT). Herein, a high-resolution multiple reaction monitoring (MRMHR) quantitative method for markers of QZT storage year was developed. Quantitative data alongside multivariate analysis were employed to discriminate and predict the storage year of QZT. Furthermore, the content of the main biochemical ingredients, catechins and alkaloids, and free amino acids (FAA) were assessed for this purpose. The results show that targeted marker-based models exhibited superior discrimination and prediction performance among four datasets. The R2Xcum, R2Ycum and Q2cum of orthogonal projection to latent structure-discriminant analysis discrimination model were close to 1. The correlation coefficient (R2) and the root mean square error of prediction of the QZT storage year prediction model were 0.9906 and 0.63, respectively. This study provides valuable insights into tea storage quality and highlights the potential application of targeted markers in food quality evaluation.

Enhancing breast cancer screening with urinary biomarkers and Random Forest supervised classification: A comprehensive investigation

OBJECTIVES

Urinary sex hormones are investigated as potential biomarkers for the early detection of breast cancer, aiming to evaluate their relevance and applicability, in combination with supervised machine-learning data analysis, toward the ultimate goal of extensive screening.

METHODS

Sex hormones were determined on urine samples collected from 250 post-menopausal women (65 healthy - 185 with breast cancer, recruited among the clinical patients of Candiolo Cancer Institute FPO-IRCCS (Torino, Italy). Two analytical procedures based on UHPLC-MS/HRMS were developed and comprehensively validated to quantify 20 free and conjugated sex hormones from urine samples. The quantitative data were processed by seven machine learning algorithms. The efficiency of the resulting models was compared.

RESULTS

Among the tested models aimed to relate urinary estrogen and androgen levels and the occurrence of breast cancer, Random Forest (RF) proved to underscore all the other supervised classification approaches, including Partial Least Squares - Discriminant Analysis (PLS-DA), in terms of effectiveness and robustness. The final optimized model built on only five biomarkers (testosterone-sulphate, alpha-estradiol, 4-methoxyestradiol, DHEA-sulphate, and epitestosterone-sulphate) achieved an approximate 98% diagnostic accuracy on replicated validation sets. To balance the less-represented population of healthy women, a Synthetic Minority Oversampling TEchnique (SMOTE) data oversampling approach was applied.

CONCLUSIONS

By means of tunable hyperparameters optimization, the RF algorithm showed great potential for early breast cancer detection, as it provides clear biomarkers ranking and their relative efficiency, allowing to ground the final diagnostic model on a restricted selection five steroid biomarkers only, as desirable for noninvasive tests with wide screening purposes.

Identification of provenance of Basmati rice grown in different regions of Punjab through multivariate analysis

High-priced Basmati rice is vulnerable to deliberate mislabeling to increase profits. This type of fraud may lower consumers' confidence as inferior products can affect brand reputation. To address this problem, there is a need to devise a method that can efficiently distinguish Basmati rice grown in regions that are famous versus the regions that are not suitable for their production. Therefore, in this investigation, thirty-six samples of Basmati rice were collected from two zones of Punjab province (one known for Basmati rice) of Pakistan which is the major producer of Basmati rice. The elemental composition of rice samples was assessed using inductively coupled plasma-optical emission spectrometry and an organic elemental analyzer, whereas data on δ13C was acquired using isotopic ratio-mass spectrometry. Regional clustering of samples based on their respective cultivation zones was observed using multivariate data analysis techniques. Partial least squares-discriminant analysis was found to be effective in grouping rice samples from the different locations and identifying unknown samples belonging to these two regions. Further recommendations are presented to develop a better model for tracing the origin of unidentified rice samples.

Prospective exploration of hazelnut's unsaponifiable fraction for geographical and varietal authentication: A comparative study of advanced fingerprinting and untargeted profiling techniques

This study compares two data processing techniques (fingerprinting and untargeted profiling) to authenticate hazelnut cultivar and provenance based on its unsaponifiable fraction by GC-MS. PLS-DA classification models were developed on a selected sample set (n = 176). As test cases, cultivar models were developed for "Tonda di Giffoni" vs other cultivars, whereas provenance models were developed for three origins (Chile, Italy or Spain). Both fingerprinting and untargeted profiling successfully classified hazelnuts by cultivar or provenance, revealing the potential of the unsaponifiable fraction. External validation provided over 90 % correct classification, with fingerprinting slightly outperforming. Analysing PLS-DA models' regression coefficients and tentatively identifying compounds corresponding to highly relevant variables showed consistent agreement in key discriminant compounds across both approaches. However, fingerprinting in selected ion mode extracted slightly more information from chromatographic data, including minor discriminant species. Conversely, untargeted profiling acquired in full scan mode, provided pure spectra, facilitating chemical interpretability.

Low-abundance proteins-based label-free SERS approach for high precision detection of liver cancer with different stages

BACKGROUND

Surface-enhanced Raman scattering (SERS) technology have unique advantages of rapid, simple, and highly sensitive in the detection of serum, it can be used for the detection of liver cancer. However, some protein biomarkers in body fluids are often present at ultra-low concentrations and severely interfered with by the high-abundance proteins (HAPs), which will affect the detection of specificity and accuracy in cancer screening based on the SERS immunoassay. Clearly, there is a need for an unlabeled SERS method based on low abundance proteins, which is rapid, noninvasive, and capable of high precision detection and screening of liver cancer.

RESULTS

Serum samples were collected from 60 patients with liver cancer (27 patients with stage T1 and T2 liver cancer, 33 patients with stage T3 and T4 liver cancer) and 40 healthy volunteers. Herein, immunoglobulin and albumin were separated by immune sorption and Cohn ethanol fractionation. Then, the low abundance protein (LAPs) was enriched, and high-quality SERS spectral signals were detected and obtained. Finally, combined with the principal component analysis-linear discriminant analysis (PCA-LDA) algorithm, the SERS spectrum of early liver cancer (T1-T2) and advanced liver cancer (T3-T4) could be well distinguished from normal people, and the accuracy rate was 98.5% and 100%, respectively. Moreover, SERS technology based on serum LAPs extraction combined with the partial least square-support vector machine (PLS-SVM) successfully realized the classification and prediction of normal volunteers and liver cancer patients with different tumor (T) stages, and the diagnostic accuracy of PLS-SVM reached 87.5% in the unknown testing set.

SIGNIFICANCE

The experimental results show that the serum LAPs SERS detection combined with multivariate statistical algorithms can be used for effectively distinguishing liver cancer patients from healthy volunteers, and even achieved the screening of early liver cancer with high accuracy (T1 and T2 stage). These results showed that serum LAPs SERS detection combined with a multivariate statistical diagnostic algorithm has certain application potential in early cancer screening.

Development of a LC-QTOF-MS based dilute-and-shoot approach for the botanical discrimination of honeys

Honeys of particular botanical origins can be associated with premium market prices, a trait which also makes them susceptible to fraud. Currently available authenticity testing methods for botanical classification of honeys are either time-consuming or only target a few "known" types of markers. Simple and effective methods are therefore needed to monitor and guarantee the authenticity of honey. In this study, a 'dilute-and-shoot' approach using liquid chromatography (LC) coupled to quadrupole time-of-flight-mass spectrometry (QTOF-MS) was applied to the non-targeted fingerprinting of honeys of different floral origin (buckwheat, clover and blueberry). This work investigated for the first time the impact of different instrumental conditions such as the column type, the mobile phase composition, the chromatographic gradient, and the MS fragmentor voltage (in-source collision-induced dissociation) on the botanical classification of honeys as well as the data quality. Results indicated that the data sets obtained for the various LC-QTOF-MS conditions tested were all suitable to discriminate the three honeys of different floral origin regardless of the mathematical model applied (random forest, partial least squares-discriminant analysis, soft independent modelling by class analogy and linear discriminant analysis). The present study investigated different LC-QTOF-MS conditions in a "dilute and shoot" method for honey analysis, in order to establish a relatively fast, simple and reliable analytical method to record the chemical fingerprints of honey. This approach is suitable for marker discovery and will be used for the future development of advanced predictive models for honey botanical origin.

Trace the origin of yak meat in Xizang based on stable isotope combined with multivariate statistics

In this study, the feasibility of tracing the origin of yak meat in Xizang Autonomous Region based on stable isotope combined with multivariable statistics was researched. The δ13C, δ15N, δ2H and δ18O in yak meat were determined by stable isotope ratio mass spectrometry, and the data were analyzed by analysis of variance, fisher discriminant analysis (FDA), back propagation (BP) neural network and orthogonal partial least squares discrimination analysis (OPLS-DA). The results showed that the δ13C, δ15N, δ2H and δ18O had significant differences among different origins (P < 0.05). The overall original correct discrimination rate of fisher discriminant analysis was 89.7 %, and the correct discrimination rate of cross validation was 88.2 %. The correct classification rate of BP neural network based on training set was 93.38 %, and the correct classification rate of BP neural network based on test set was 89.83 %. The OPLS-DA model interpretation rate parameter R2Y was 0.67, the model prediction rate parameter Q2 was 0.409, which could distinguish yak meat from seven different producing areas in Xizang Autonomous Region. The results showed that the origin of yak meat in Xizang Autonomous Region can be traced based on stable isotope combined with multivariate statistics.

Multidimensional discriminant analysis of species, strains and culture age of closely related entomopathogenic fungi using reflectance spectroscopy

The diversity of fungal strains is influenced by genetic and environmental factors, growth conditions and mycelium age, and the spectral features of fungal mycelia are associated with their biochemical, physiological, and structural traits. This study investigates whether intraspecific differences can be detected in two closely related entomopathogenic species, namely Cordyceps farinosa and Cordyceps fumosorosea, using ultraviolet A to shortwave infrared (UVA-SWIR) reflectance spectra. Phylogenetic analysis of all strains revealed a high degree of uniformity among the populations of both species. The characteristics resulting from variation in the species, as well as those resulting from the age of the cultures were determined. We cultured fungi on PDA medium and measured the reflectance of mycelia in the 350-2500 nm range after 10 and 17 days. We subjected the measurements to quadratic discriminant analysis (QDA) to identify the minimum number of bands containing meaningful information. We found that when the age of the fungal culture was known, species represented by a group of different strains could be distinguished with no more than 3-4 wavelengths, compared to 7-8 wavelengths when the age of the culture was unknown. At least 6-8 bands were required to distinguish cultures of a known species among different age groups. Distinguishing all strains within a species was more demanding: at least 10 bands were required for C. fumosorosea and 21 bands for C. farinosa. In conclusion, fungal differentiation using point reflectance spectroscopy gives reliable results when intraspecific and age variations are taken into account.

Rapid analysis of Radix Astragali using a portable Raman spectrometer with 1064-nm laser excitation and data fusion with PLS-DA

Radix Astragali is a medicinal herb with various physiological activities. There were high similarities among Radix Astragali samples from different regions owing to similarities in their major chemical compositions. Raman spectroscopy is a non-invasive and non-des- tructive technique that can be used in in-situ analysis of herbal samples. Dispersive Raman scattering, excited at 1064 nm, produced minimal fluorescence background and facilitated easy detection of the weak Raman signal. By moving the portable Raman probe point-by- point from the centre of the Radix Astragali sample to the margin, the spectral fingerprints, composed of dozens of Raman spectra representing the entire Radix Astragali samples, were obtained. Principal component analysis and partial least squares discriminant analysis (PLS-DA) were applied to the Radix Astragali spectral data to compare classification results, leading to efficient discrimination between genuine and counterfeit products. Furthermore, based on the PLS-DA model using data fusion combined with different pre- processing methods, the samples from Shanxi Province were separated from those belonging to other habitats. The as-proposed combination method can effectively improve the recognition rate and accuracy of identification of herbal samples, which can be a valuable tool for the identification of genuine medicinal herbs with uneven qualities and various origins.

Distinction and quantification of Panax polysaccharide extracts via attenuated total reflectance-Fourier transform infrared spectroscopy with first-order derivative processing

Derivative spectroscopy is used to separate the small absorption peaks superimposed on the main absorption band, which is widely adopted in modern spectral analysis to increase both the valid spectral information and the identification accuracy. In this study, a method based on attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) with first-order derivative (FD) processing combined with chemometrics is proposed for rapid qualitative and quantitative analysis of Panax ginseng polysaccharides (PGP), Panax notoginseng polysaccharides (PNP), and Panax quinquefolius polysaccharides (PQP). First, ATR-FTIR with FD processing was used to establish the discriminant model combined with principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA) and linear discriminant analysis (LDA). After that, two-dimensional ATR-FTIR based on single-characteristic temperature as external interference (2D-sATR-FTIR) was established using ATR-FTIR with FD processing. Then, ATR-FTIR with FD processing was combined with PLS to establish and optimize the quantitative regression model. Finally, the established discriminant model and 2D-sATR-FTIR successfully distinguished PGP, PNP and PQP, and the optimal PLS regression model had a good prediction ability for the Panax polysaccharide extracts content. This strategy provides an efficient, economical and nondestructive method for the distinction and quantification of PGP, PNP and PQP in a short detection time.

Classification of asbestos and their nonasbestiform analogues using FTIR and multivariate data analysis

This study presents a possible application of Fourier transform infrared (FTIR) spectrometry and multivariate data analysis, principal component analysis (PCA), and partial least squares-discriminant analysis (PLS-DA) for classifying asbestos and their nonasbestiform analogues. The objectives of the study are: 1) to classify six regulated asbestos types and 2) to classify between asbestos types and their nonasbestiform analogues. The respirable fraction of six regulated asbestos types and their nonasbestiform analogues were prepared in potassium bromide pellets and collected on polyvinyl chloride membrane filters for FTIR measurement. Both PCA and PLS-DA classified asbestos types and their nonasbestiform analogues on the score plots showed a very distinct clustering of samples between the serpentine (chrysotile) and amphibole groups. The PLS-DA model provided ∼95% correct prediction with a single asbestos type in the sample, although it did not provide all correct predictions for all the challenge samples due to their inherent complexity and the limited sample number. Further studies are necessary for a better prediction level in real samples and standardization of sampling and analysis procedures.

Sex and blood group determination from hair using ATR-FTIR spectroscopy and chemometrics

Examination of hair with its intact root is commonly used for DNA profiling of the donor. However, its use for gathering other types of information is less explored. Using attenuated total reflectance-Fourier transform infrared spectroscopy, the present study aims to explore other relevant aspects in a non-destructive manner for forensics. Determining the sex and blood group of human hair samples were the major goals of the study. Sex determination was accomplished by analyzing the differential vibrational intensities and stretching of various chemical groups associated with hair and its proteins. Statistical inference of spectral data was performed using chemometric algorithms such as PCA and PLS-DA. The PLS-DA model determined sex with 100% accuracy and blood grouping with an average accuracy of 95%. The present study is the first of its kind to determine sex and blood grouping from human scalp hair shafts, as far as the author knows. By acting as a preliminary screening test, this study could have significant implications for forensic analysis of crime scene samples. Human and synthetic hair were used in validation studies, resulting in 100% accuracy, specificity, and sensitivity, with 0% false positives and false negatives. The technique ATR FTIR spectroscopy could complement the currently used methods of hair analysis such as physical examination and mitochondrial or genomic DNA analysis.

Identifying hypothermia death in a mouse model by ATR-FTIR

The identification of hypothermia death (HD) is difficult for cadavers, especially the distinction from death due to alternative causes. A large number of studies have shown that brown adipose tissue (BAT) plays critical roles in thermoregulation of mammals. In this study, BAT of mice was used for the discrimination of HD using attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR). A modified mouse HD model conducted by Feeney DM was used in this study to obtain infrared spectra of BAT. Principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA) were used to establish discrimination models. The PLS-DA and OPLS-DA models exhibit prominent discriminative efficiency, and the accuracy of HD identification using fingerprint regions and ratios of absorption intensity is near 100% in both the calibration and validation sets. Our preliminary study suggests that BAT may be an extremely effective target tissue for identification of cadavers of HD, and ATR-FTIR spectra combined with chemometrics have also shown potential for cadaver identification in forensic practice in a fast and accurate manner.

Morphometric analysis for determination of larval instars in Dermestes frischii Kugelann and Dermestes undulatus Brahm (Coleoptera: Dermestidae)

Dermestes frischii Kugelann, 1792 and Dermestes undulatus Brahm, 1790 are the most abundant species worldwide at outdoor or indoor crime scenes during the dry and skeletal stages of decomposition. The attribution of larval age in these beetles is problematic due to the variable number of instars, which is influenced by environmental factors. In this study, a morphometric approach was used to look for potential morphological features as evidence of larval stages. Breeding and monitoring were performed for both species in an incubator with a preset temperature of 28°C ± 0.5 without a photoperiod. Morphometric measurements were made on 10 larvae per instar for each species using length, width, and thickness parameters. Linear discriminant analysis was then used to generate decision boundaries that clearly separated larval stages. The cross-validation procedure demonstrated that the morphometric approach successfully discriminated adjacent larval stages in both species with high values of sensitivity and specificity. This less-invasive approach could improve the ability to estimate minPMI in forensic studies of Dermestidae beetles. Future studies may extend this approach to other species and establish good practices for collecting and storing specimens for morphometric analysis.

Detection of carbon nanotubes in bovine raw milk through Fourier transform Raman spectroscopy

The potential use of carbon-based methodologies for drug delivery and reproductive biology in cows raises concerns about residues in milk and food safety. This study aimed to assess the potential of Fourier transform Raman spectroscopy and discriminant analysis using partial least squares (PLS-DA) to detect functionalized multiwalled carbon nanotubes (MWCNT) in bovine raw milk. Oxidized MWCNT were diluted in milk at different concentrations from 25.00 to 0.01 µg/mL. Raman spectroscopy measurements and PLS-DA were performed to identify low concentrations of MWCNT in milk samples. The PLS-DA model was characterized by the analysis of the variable importance in projection (VIP) scores. All the training samples were correctly classified by the model, resulting in no false-positive or false-negative classifications. For test samples, only one false-negative result was observed, for 0.01 µg/mL MWCNT dilution. The association between Raman spectroscopy and PLS-DA was able to identify MWCNT diluted in milk samples up to 0.1 µg/mL. The PLS-DA model was built and validated using a set of test samples and spectrally interpreted based on the highest VIP scores. This allowed the identification of the vibrational modes associated with the D and G bands of MWCNT, as well as the milk bands, which were the most important variables in this analysis.

Advances of machine learning-assisted small extracellular vesicles detection strategy

Detection of extracellular vesicles (EVs), particularly small EVs (sEVs), is of great significance in exploring their physiological characteristics and clinical applications. The heterogeneity of sEVs plays a crucial role in distinguishing different types of cells and diseases. Machine learning, with its exceptional data processing capabilities, offers a solution to overcome the limitations of conventional detection methods for accurately classifying sEV subtypes and sources. Principal component analysis, linear discriminant analysis, partial least squares discriminant analysis, XGBoost, support vector machine, k-nearest neighbor, and deep learning, along with some combined methods such as principal component-linear discriminant analysis, have been successfully applied in the detection and identification of sEVs. This review focuses on machine learning-assisted detection strategies for cell identification and disease prediction via sEVs, and summarizes the integration of these strategies with surface-enhanced Raman scattering, electrochemistry, inductively coupled plasma mass spectrometry and fluorescence. The performance of different machine learning-based detection strategies is compared, and the advantages and limitations of various machine learning models are also evaluated. Finally, we discuss the merits and limitations of the current approaches and briefly outline the perspective of potential research directions in the field of sEV analysis based on machine learning.

Quality difference analysis of raw and vinegar-processed products of Qingpi based on color and component correlation

The dried young fruit of Citrus reticulata Blanco, known as Qingpi, is commonly used in clinic both with its raw and vinegar-processed products. However, the distinctions in quality between these two products remain unclear, and the methods for identification are considerably intricate. In this study, an electronic eye technique was applied to assess the overall color of Qingpi products before and after processing. The luminosity (L*) and yellow-blue (b*) values of Qingpi decreased after vinegar processing, while red-green (a*) values increased. The discriminant function models based on color parameters were established to effectively classify the two products. The chemical compositions of different Qingpi products were characterized using ultra-high performance liquid chromatography fingerprint technology, and 10 distinct components were considered as potential chemical markers. The correlation analysis revealed a significant relationship between chromatic values and chemical components. In conclusion, the results of this study suggested that chromaticity can be effectively considered as a valuable instrument for the prediction of component content in both raw and vinegar-processed Qingpi products. This study will provide new ideas and methods for identification and quality evaluation of Qingpi processed products, as well as provide a reference for standardizing traditional Chinese medicine processing techniques.

Comparison of classification methods performance for defining the best reuse of waste wood material using NIR spectroscopy

Recycling of post-consumer waste wood material is becoming an increasingly appealing alternative to disposal. However, its huge heterogeneity is calling for an assessment of the material characteristics in order to define the best recycling option and intended reuse. In fact, waste wood comes into a variety of uses/types of wood, along with several levels of contamination, and it can be divided into different categories based on its composition and quality grade. This study provides the measurement of more than a hundred waste wood samples and their characterisation using a hand-held NIR spectrophotometer. Three classification methods, i.e. K-nearest Neighbours (KNN), Principal Component Analysis - Linear Discriminant Analysis (PCA-LDA) and PCA-KNN, have been compared to develop models for the sorting of waste wood in quality categories according to the best-suited reuse. In addition, the classification performance has been investigated as a function of the number of the spectral measurements of the sample and as the average of the spectral measurements. The results showed that PCA-KNN performs better than the other classification methods, especially when the material is ground to 5 cm of particle size and the spectral measurements are averaged across replicates (classification accuracy: 90.9 %). NIR spectroscopy, coupled with chemometrics, turned out to be a promising tool for the real-time sorting of waste wood material, ensuring a more accurate and sustainable waste wood management. Obtaining real-time information about the quality and characteristics of waste wood material translates into a decision of the best recycling option, increasing its recycling potential.

MultiToxPred 1.0: a novel comprehensive tool for predicting 27 classes of protein toxins using an ensemble machine learning approach

Protein toxins are defense mechanisms and adaptations found in various organisms and microorganisms, and their use in scientific research as therapeutic candidates is gaining relevance due to their effectiveness and specificity against cellular targets. However, discovering these toxins is time-consuming and expensive. In silico tools, particularly those based on machine learning and deep learning, have emerged as valuable resources to address this challenge. Existing tools primarily focus on binary classification, determining whether a protein is a toxin or not, and occasionally identifying specific types of toxins. For the first time, we propose a novel approach capable of classifying protein toxins into 27 distinct categories based on their mode of action within cells. To accomplish this, we assessed multiple machine learning techniques and found that an ensemble model incorporating the Light Gradient Boosting Machine and Quadratic Discriminant Analysis algorithms exhibited the best performance. During the tenfold cross-validation on the training dataset, our model exhibited notable metrics: 0.840 accuracy, 0.827 F1 score, 0.836 precision, 0.840 sensitivity, and 0.989 AUC. In the testing stage, using an independent dataset, the model achieved 0.846 accuracy, 0.838 F1 score, 0.847 precision, 0.849 sensitivity, and 0.991 AUC. These results present a powerful next-generation tool called MultiToxPred 1.0, accessible through a web application. We believe that MultiToxPred 1.0 has the potential to become an indispensable resource for researchers, facilitating the efficient identification of protein toxins. By leveraging this tool, scientists can accelerate their search for these toxins and advance their understanding of their therapeutic potential.

Evaluation of Hand Action Classification Performance Using Machine Learning Based on Signals from Two sEMG Electrodes

Classification-based myoelectric control has attracted significant interest in recent years, leading to prosthetic hands with advanced functionality, such as multi-grip hands. Thus far, high classification accuracies have been achieved by increasing the number of surface electromyography (sEMG) electrodes or adding other sensing mechanisms. While many prescribed myoelectric hands still adopt two-electrode sEMG systems, detailed studies on signal processing and classification performance are still lacking. In this study, nine able-bodied participants were recruited to perform six typical hand actions, from which sEMG signals from two electrodes were acquired using a Delsys Trigno Research+ acquisition system. Signal processing and machine learning algorithms, specifically, linear discriminant analysis (LDA), k-nearest neighbors (KNN), and support vector machines (SVM), were used to study classification accuracies. Overall classification accuracy of 93 ± 2%, action-specific accuracy of 97 ± 2%, and F1-score of 87 ± 7% were achieved, which are comparable with those reported from multi-electrode systems. The highest accuracies were achieved using SVM algorithm compared to LDA and KNN algorithms. A logarithmic relationship between classification accuracy and number of features was revealed, which plateaued at five features. These comprehensive findings may potentially contribute to signal processing and machine learning strategies for commonly prescribed myoelectric hand systems with two sEMG electrodes to further improve functionality.

Recognition of breast cancer subtypes using FTIR hyperspectral data

Fourier-transform infrared spectroscopy (FTIR) is a powerful, non-destructive, highly sensitive and a promising analytical technique to provide spectrochemical signatures of biological samples, where markers like carbohydrates, proteins, and phosphate groups of DNA can be recognized in biological micro-environment. However, method of measurements of large cells need an excessive time to achieve high quality images, making its clinical use difficult due to speed of data-acquisition and lack of optimized computational procedures. To address such challenges, Machine Learning (ML) based technologies can assist to assess an accurate prognostication of breast cancer (BC) subtypes with high performance. Here, we applied FTIR spectroscopy to identify breast cancer subtypes in order to differentiate between luminal (BT474) and non-luminal (SKBR3) molecular subtypes. For this reason, we tested multivariate classification technique to extract feature information employing three-dimension (3D)-discriminant analysis approach based on 3D-principle component analysis-linear discriminant analysis (3D-PCA-LDA) and 3D-principal component analysis-quadratic discriminant analysis (3D-PCA-QDA), showing an improvement in sensitivity (98%), specificity (94%) and accuracy (98%) parameters compared to conventional unfolded methods. Our results evidence that 3D-PCA-LDA and 3D-PCA-QDA are potential tools for discriminant analysis of hyperspectral dataset to obtain superior classification assessment.

Structural characterization of DNA amplicons by ATR-FTIR spectroscopy as a guide for screening metainflammatory disorders in blood plasma

Attenuated total reflectance (ATR) Fourier transform infrared (FTIR) spectroscopy is a promising rapid, reagent-free, and low-cost technique considered for clinical translation. It allows to characterize biofluids proteome, lipidome, and metabolome at once. Metainflammatory disorders share a constellation of chronic systemic inflammation, oxidative stress, aberrant adipogenesis, and hypoxia, that significantly increased cardiovascular and cancer risk. As a result, these patients have elevated concentration of cfDNA in the bloodstream. Considering this, DNA amplicons were analyzed by ATR-FTIR at 3 concentrations with 1:100 dilution: (IU/mL): 718, 7.18, and 0.0718. The generated IR spectrum was used as a guide for variable selection. The main peaks in the biofingerprint (1800-900 cm-1) give important information about the base, base-sugar, phosphate, and sugar-phosphate transitions of DNA. To validate our method of selecting variables in blood plasma, 38 control subjects and 12 with metabolic syndrome were used. Using the wavenumbers of the peaks in the biofingerprint of the DNA amplicons, was generated a discriminant analysis model with Mahalanobis distance in blood plasma, and 100 % discrimination accuracy was obtained. In addition, the interval 1475-1188 cm-1 showed the greatest sensitivity to variation in the concentration of DNA amplicons, so curve fitting with Gaussian funcion was performed, obtaining adjusted-R2 of 0.993. PCA with Mahalanobis distance in the interval 1475-1188 cm-1 obtained an accuracy of 96 % and PLS-DA modeling in the interval 1475-1088 cm-1 obtained AUC = 0.991 with sensitivity of 95 % and specificity of 100 %. Therefore, ATR-FTIR spectroscopy with variable selection guided by DNA IR peaks is a promising and efficient method to be applied in metainflammatory disorders.

Early diagnosis of thyroid-associated ophthalmopathy using label-free Raman spectroscopy and multivariate analysis

Thyroid-associated ophthalmopathy (TAO) is the most common orbital disease in adults, with complex clinical manifestations and significant impacts on the life quality of patients. The current diagnosis of TAO lacks reliable biomarkers for early and non-invasive screening and detection, easily leading to poor prognosis. Therefore, it is essential to explore new methods for accurately predicting TAO development in its early stage. In this study, Raman spectroscopy, with non-destructive, label-free, and high-sensitivity characteristics, was used to analyze the differences in biochemical components of orbital adipocyte and tear samples between TAO and control groups. Furthermore, a multivariate analysis method (i.e., Principal Component Analysis-Linear Discriminant Analysis (PCA-LDA)) was applied for data processing and analysis. Compared with controls, PCA-LDA yielded TAO diagnostic accuracies of 72.7% and 75.0% using orbital adipocytes and tears, respectively. Our proof-of-concept results suggest that Raman spectroscopy holds potential for exploring the underlying pathogenesis of TAO, and its potential application in early screening of other thyroid-associated diseases can be further expanded.

Discrimination of opium from Afghanistan and Myanmar by infrared spectroscopy coupled with machine learning methods

Afghanistan and Myanmar are two overwhelming opium production places. In this study, rapid and efficient methods for distinguishing opium from Afghanistan and Myanmar were developed using infrared spectroscopy (IR) coupled with multiple machine learning (ML) methods for the first time. A total of 146 authentic opium samples were analyzed by mid-IR (MIR) and near-IR (NIR), within them 116 were used for model training and 30 were used for model validation. Six ML methods, including partial least squares discriminant analysis (PLS-DA), orthogonal PLS-DA (OPLS-DA), k-nearest neighbour (KNN), support vector machine (SVM), random forest (RF), and artificial neural networks (ANNs) were constructed and compared to get the best classification effect. For MIR data, the average of precision, recall and f1-score for all classification models were 1.0. For NIR data, the average of precision, recall and f1-score for different classification models ranged from 0.90 to 0.94. The comparison results of six ML models for MIR and NIR data showed that MIR was more suitable for opium geography classification. Compared with traditional chromatography and mass spectrometry profiling methods, the advantages of MIR are simple, rapid, cost-effective, and environmentally friendly. The developed IR chemical profiling methodology may find wide application in classification of opium from Afghanistan and Myanmar, and also to differentiate them from opium originating from other opium producing countries. This study presented new insights into the application of IR and ML to rapid drug profiling analysis.

Serum Raman spectroscopy: Evaluation of tumour load variations in experimental carcinogenesis

Several serum Raman spectroscopy (RS) studies have demonstrated its potential as an oral cancer screening tool. This study investigates influence of low tumour load (LTL) and high tumour load (HTL) on serum RS using hamster buccal pouch model of experimental oral carcinogenesis. Sera of untreated control, LTL, and HTL groups at week intervals during malignant transformation were employed. Serum Raman spectra were subjected to multivariate analyses-principal component analysis, principal component-based linear discriminant analysis (for stratification of study groups), and multivariate curve resolution-alternating least squares (MCR-ALS) (to comprehend biomolecular differences). Multivariate analysis revealed misclassifications between LTL and HTL at all week intervals. MCR-ALS components showed statistically significant abundances between control versus LTL and control versus HTL, but could not discern LTL and HTL. MCR-ALS components exhibited spectral mixtures of proteins, lipids, heme and nucleic acids. Thus, these findings support use of serum RS as a screening tool as varying tumour load is not a confounding factor influencing the technique.

A rapid method for identification of Lanxangia tsaoko origin and fruit shape: FT-NIR combined with chemometrics and image recognition

Lanxangia tsaoko's accurate classifications of different origins and fruit shapes are significant for research in L. tsaoko difference between origin and species as well as for variety breeding, cultivation, and market management. In this work, Fourier transform-near infrared (FT-NIR) spectroscopy was transformed into two-dimensional and three-dimensional correlation spectroscopies to further investigate the spectral characteristics of L. tsaoko. Before building the classification model, the raw FT-NIR spectra were preprocessed using multiplicative scatter correction and second derivative, whereas principal component analysis, successive projections algorithm, and competitive adaptive reweighted sampling were used for spectral feature variable extraction. Then combined with partial least squares-discriminant analysis (PLS-DA), support vector machine (SVM), decision tree, and residual network (ResNet) models for origin and fruit shape discriminated in L. tsaoko. The PLS-DA and SVM models can achieve 100% classification in origin classification, but what is difficult to avoid is the complex process of model optimization. The ResNet image recognition model classifies the origin and shape of L. tsaoko with 100% accuracy, and without the need for complex preprocessing and feature extraction, the model facilitates the realization of fast, accurate, and efficient identification.

Rapid and non-invasive detection of cystic echinococcosis in sheep based on serum fluorescence spectrum combined with machine learning algorithms

Cystic echinococcosis (CE) is a grievous zoonotic parasitic disease. Currently, the traditional technology of screening CE is laborious and expensive, developing an innovative technology is urgent. In this study, we combined serum fluorescence spectroscopy with machine learning algorithms to develop an innovative screening technique to diagnose CE in sheep. Serum fluorescence spectra of Echinococcus granulosus sensu stricto-infected group (n = 63) and uninfected E. granulosus s.s. group (n = 60) under excitation at 405 nm were recorded. The linear support vector machine (Linear SVM), Quadratic SVM, medium radial basis function (RBF) SVM, K-nearest neighbor (KNN), and principal component analysis-linear discriminant analysis (PCA-LDA) were used to analyze the spectra data. The results showed that Quadratic SVM had the great classification capacity, its sensitivity, specificity, and accuracy were 85.0%, 93.8%, and 88.9%, respectively. In short, serum fluorescence spectroscopy combined with Quadratic SVM algorithm has great potential in the innovative diagnosis of CE in sheep.

Discriminant analysis and logistic regression on genetic history and environmental factors in children with asthma

BACKGROUND

Asthma is known to be related to genetic and environmental factors, we aimed to identify the predictors discriminating between children with asthma and a control group in order to build typical profiles of these children.

METHODS

A multidimensional analysis covered children (58 with asthma and 217 as control group), under 17 years of age, involving environmental variables and medical history of these children and their families.

RESULTS

Chi-square tests highlighted significant links between variables as rhinitis and conjunctivitis (P<0.001). The results showed, in group of asthmatic children, significant high frequencies of allergies, mainly seasonal (P<0.001), rhinitis, family history more present in mothers (P=0.002) and in maternal aunts and uncles (P<0.02). Allergies were mostly present in mothers of asthmatic children (P=0.03). Children whose father, mother or both had asthma were significantly more numerous in asthmatic group (P=0.0007). A multiple correspondence analysis (MCA) identified two typical profiles of children, a first group of asthmatic children with positive modalities of family history, medical and environmental factors, a second, the control group (nA, non-asthmatic children), with essentially negative modalities of the variables. Logistic regression (LR) resulted in a final model which retained four significant predictors, rhinitis (P=0.01), atopic dermatitis (P=0.04), mother antecedents (P=0.03) and paternal uncle antecedents (P=0.008) with a globally appreciable predictive value (82%) of the Hosmer-Lemeshow Test.

CONCLUSIONS

These results allowed the drafting of a typical profile quantifying through a function of a few predictors, the variation of the probability for a child to develop an asthma.

Determination of the geographical origin of chicken (breast and drumstick) using ICP-OES and ICP-MS: Chemometric analysis

This study aimed to develop a geographical origin discrimination analytical method for chicken breasts and drumsticks based on inductively coupled plasma (ICP). The sixty elements were set as variables, and the geographical origin discrimination analysis was conducted through chemometrics. In orthogonal partial least square discriminant analysis (OPLS-DA), twenty-three variable importance in projection (VIP) elements were selected in chicken breasts, and twenty-eight VIP elements were selected in drumsticks. The importance of the selected elements was displayed by the area under the curve (AUC) value of the receiver operating characteristic (ROC). Verification of OPLS-DA was performed through permutation test and good results were obtained. A heatmap was also used as a method for determining the geographical origin, and each top element discriminant classification was 100 % accurate, as determined through canonical discriminant analysis (CDA). This method shows potential as a food analysis tool and can accurately determine the geographic origin of chicken.

Combined quantitative lipidomics and back-propagation neural network approach to discriminate the breed and part source of lamb

The study successfully utilized an analytical approach that combined quantitative lipidomics with back-propagation neural networks to identify breed and part source of lamb using small-scale samples. 1230 molecules across 29 lipid classes were identified in longissimus dorsi and knuckle meat of both Tan sheep and Bahan crossbreed sheep. Applying multivariate statistical methods, 12 and 7 lipid molecules were identified as potential markers for breed and part identification, respectively. Stepwise linear discriminant analysis was applied to select 3 and 4 lipid molecules, respectively, for discriminating lamb breed and part sources, achieving correct rates of discrimination of 100 % and 95 %. Additionally, back-propagation neural network proved to be a superior method for identifying sources of lamb meat compared to other machine learning approaches. These findings indicate that integrating lipidomics with back-propagation neural network approach can provide an effective strategy to trace and certify lamb products, ensuring their quality and protecting consumer rights.

1-Year Mortality Prediction through Artificial Intelligence Using Hemodynamic Trace Analysis among Patients with ST Elevation Myocardial Infarction

Background and Objectives: Patients presenting with ST Elevation Myocardial Infarction (STEMI) due to occlusive coronary arteries remain at a higher risk of excess morbidity and mortality despite being treated with primary percutaneous coronary intervention (PPCI). Identifying high-risk patients is prudent so that close monitoring and timely interventions can improve outcomes. Materials and Methods: A cohort of 605 STEMI patients [64.2 ± 13.2 years, 432 (71.41%) males] treated with PPCI were recruited. Their arterial pressure (AP) wave recorded throughout the PPCI procedure was analyzed to extract features to predict 1-year mortality. After denoising and extracting features, we developed two distinct feature selection strategies. The first strategy uses linear discriminant analysis (LDA), and the second employs principal component analysis (PCA), with each method selecting the top five features. Then, three machine learning algorithms were employed: LDA, K-nearest neighbor (KNN), and support vector machine (SVM). Results: The performance of these algorithms, measured by the area under the curve (AUC), ranged from 0.73 to 0.77, with accuracy, specificity, and sensitivity ranging between 68% and 73%. Moreover, we extended the analysis by incorporating demographics, risk factors, and catheterization information. This significantly improved the overall accuracy and specificity to more than 76% while maintaining the same level of sensitivity. This resulted in an AUC greater than 0.80 for most models. Conclusions: Machine learning algorithms analyzing hemodynamic traces in STEMI patients identify high-risk patients at risk of mortality.

Identification of Vernonia patula Merr. and Its Similar Varieties Based on a Combination of HPLC Fingerprinting and Chemical Pattern Recognition

Vernonia patula Merr. (VP) is a traditional medicine used by the Zhuang and Yao people, known for its therapeutic properties in treating anemopyretic cold and other diseases. Distinguishing VP from similar varieties such as Praxelis clematidea (PC), Ageratum conyzoides L. (AC) and Ageratum houstonianum Mill (AH) was challenging due to their similar traits and plant morphology. The HPLC fingerprints of 40 batches of VP and three similar varieties were established. SPSS 20.0 and SIMCA-P 13.0 were used to statistically analyze the chromatographic peak areas of 37 components. The results showed that the similarity of the HPLC fingerprints for each of the four varieties was >0.9, while the similarity between the control chromatogram of VP and its similar varieties was <0.678. Cluster analysis and partial least squares discriminant analysis provided consistent results, indicating that all four varieties could be individually clustered together. Through further analysis, we found isochlorogenic acid A and isochlorogenic acid C were present only in the original VP, while preconene II was present in the three similar varieties of VP. These three components are expected to be identification points for accurately distinguishing VP from PC, AC and AH.

Strategy for Identifying a Robust Metabolomic Signature Reveals the Altered Lipid Metabolism in Pituitary Adenoma

Despite the well-established connection between systematic metabolic abnormalities and the pathophysiology of pituitary adenoma (PA), current metabolomic studies have reported an extremely limited number of metabolites associated with PA. Moreover, there was very little consistency in the identified metabolite signatures, resulting in a lack of robust metabolic biomarkers for the diagnosis and treatment of PA. Herein, we performed a global untargeted plasma metabolomic profiling on PA and identified a highly robust metabolomic signature based on a strategy. Specifically, this strategy is unique in (1) integrating repeated random sampling and a consensus evaluation-based feature selection algorithm and (2) evaluating the consistency of metabolomic signatures among different sample groups. This strategy demonstrated superior robustness and stronger discriminative ability compared with that of other feature selection methods including Student's t-test, partial least-squares-discriminant analysis, support vector machine recursive feature elimination, and random forest recursive feature elimination. More importantly, a highly robust metabolomic signature comprising 45 PA-specific differential metabolites was identified. Moreover, metabolite set enrichment analysis of these potential metabolic biomarkers revealed altered lipid metabolism in PA. In conclusion, our findings contribute to a better understanding of the metabolic changes in PA and may have implications for the development of diagnostic and therapeutic approaches targeting lipid metabolism in PA. We believe that the proposed strategy serves as a valuable tool for screening robust, discriminating metabolic features in the field of metabolomics.

Comparing morphological (os coxae) and metric (long bone length) sex estimation methods in archaeological collections

Objectives: The aim of this study is to evaluate the internal consistency of sex estimation using metric (long bone length) and morphological (os coxae) methodologies from different bones and across different archaeological populations from different regions. Materials and Methods: Sex was estimated using characteristics of the pelvis and compared with sex estimations using long bone length. Portuguese (659 females; 906 males) and English (141 females; 277 males) archaeological collections were analysed in this study. A set of long bone length functions were developed using one of the archaeological collections (531 females; 600 males) and its coincidence with sex estimated from the pelvis was compared to the coincidence between the pelvis and long bone length sex estimations using functions developed from contemporary collections. Intra- and inter-observer errors were calculated, as well as the sexual dimorphism index for each bone and osteological collection. Results: The accuracy of the developed functions and the other methods tested is highly variable, ranging between 25 and 100%. The accuracy of the standard forensic methods varied between collections and analysed bones. Discussion: This study reinforces that long bone length is highly population-specific, even between samples of close chronology and geography. Metric methods are good options to strengthen the sex estimations, but they need to be carefully chosen and always report the estimated probability of being male or female in either forensic or archaeological analysis.

Testing the discrepancy between physical and virtual linear measurements

Virtual calculations of bone morphology are increasingly becoming the golden standard in anthropological sciences, gradually replacing the performance of direct physical measurements. Nevertheless, the potential discrepancy between the two approaches is rarely addressed. Here, we address this question focusing on the second thoracic vertebrae of 24 well-preserved individuals from the skeletal collection of the Forensic Anthropology Unit of Medical School at the National and Kapodistrian University of Athens, Greece. Following traditional osteometric methods, a series of measurements were taken on the vertebral body, both directly (using a digital caliper) as well as on high-resolution 3D surface models. The arithmetic results of the two measuring techniques were then compared through a number of statistical analyses evaluating inter-method precision (Bland-Altman plots, TEM, %TEM and Wilcoxon test). Moreover, the values obtained from each approach were used to develop discriminant function equations for sex determination to evaluate if both approaches provide the same assessment. Both intraobserver and interobserver tests were performed. Although most statistical analyses showed a significant difference between the two measuring techniques, the discriminant function equations for sex determination provided the same assessment. Overall, the results of this experiment support the use of virtual linear measurements, also suggesting that a refinement of digital measuring protocols could improve their level of agreement with traditional direct osteometry.

Insight into the effect of cultivar and altitude on the identification of EnshiYulu tea grade in untargeted metabolomics analysis

The accurate identification of tea grade is crucial to the quality control of tea. However, existing methods lack sufficient generalization ability in identifying tea grades due to the effect of temporal and spatial factors. In this study, we analyzed the effect of cultivar and altitude on EnshiYulu (ESYL) tea grades and established a robust model to evaluate their quality. Principal component analysis (PCA) revealed that differences in variety and elevation can mask grade differences. Orthogonal projection to latent structure-discriminant analysis (OPLS-DA) was used for grade identification of samples from different altitudes. For ESYL tea samples above and below 800 m altitude, 75 and 35 grade differentiated metabolites were discovered, with 14 common differentiated metabolites. Based on reconstructed OPLS-DA models, the grades of multi-altitude sources ESYL were discriminated with a rate > 85%. These results demonstrate the potential of a grade discrimination model based on common differential metabolites, which exhibits generalization ability.

Fingerprinting based on spectral reflectance and chemometrics - An analytical approach aimed at combating the illegal trade of stingray meat in the Amazon

The survival of Amazon stingrays is threatened due to excessive fishing and habitat degradation. To address this issue, this study developed a groundbreaking method to authenticate and differentiate Amazon stingray meats using a portable spectrophotometer and chemometrics. Samples were collected from various species, including an endangered one with a commercialization ban and no population reduction records. Principal Component Analysis (PCA), identified natural groupings based on the meat's commercial origin, while Partial Least Squares-Discriminant Analysis (PLS-DA), accurately discriminated the commercial and geographic origins with 100 % accuracy. Moreover, Data-Driven Soft Independent Modeling of Class Analogy (DD-SIMCA), effectively distinguished Amazon stingray meat from other marketable species. This approach offers a rapid, precise, and non-destructive means for monitoring and controlling the illegal trade of these species, thereby supporting decision-making in the field and promoting the conservation and sustainability of freshwater stingrays in the Amazon region.

PLA2G4A and ACHE modulate lipid profiles via glycerophospholipid metabolism in platinum-resistant gastric cancer

BACKGROUND

Bioactive lipids involved in the progression of various diseases. Nevertheless, there is still a lack of biomarkers and relative regulatory targets. The lipidomic analysis of the samples from platinum-resistant in gastric cancer patients is expected to help us further improve our understanding of it.

METHODS

We employed LC-MS based untargeted lipidomic analysis to search for potential candidate biomarkers for platinum resistance in GC patients. Partial least squares discriminant analysis (PLS-DA) and variable importance in projection (VIP) analysis were used to identify differential lipids. The possible molecular mechanisms and targets were obtained by metabolite set enrichment analysis and potential gene network screened. Finally, verified them by immunohistochemical of a tissue microarray.

RESULTS

There were 71 differential lipid metabolites identified in GC samples between the chemotherapy-sensitivity group and the chemotherapy resistance group. According to Foldchange (FC) value, VIP value, P values (FC > 2, VIP > 1.5, p < 0.05), a total of 15 potential biomarkers were obtained, including MGDG(43:11)-H, Cer(d18:1/24:0) + HCOO, PI(18:0/18:1)-H, PE(16:1/18:1)-H, PE(36:2) + H, PE(34:2p)-H, Cer(d18:1 + hO/24:0) + HCOO, Cer(d18:1/23:0) + HCOO, PC(34:2e) + H, SM(d34:0) + H, LPC(18:2) + HCOO, PI(18:1/22:5)-H, PG(18:1/18:1)-H, Cer(d18:1/24:0) + H and PC(35:2) + H. Furthermore, we obtained five potential key targets (PLA2G4A, PLA2G3, DGKA, ACHE, and CHKA), and a metabolite-reaction-enzyme-gene interaction network was built to reveal the biological process of how they could disorder the endogenous lipid profile of platinum resistance in GC patients through the glycerophospholipid metabolism pathway. Finally, we further identified PLA2G4A and ACHE as core targets of the process by correlation analysis and tissue microarray immunohistochemical verification.

CONCLUSION

PLA2G4A and ACHE regulated endogenous lipid profile in the platinum resistance in GC patients through the glycerophospholipid metabolism pathway. The screening of lipid biomarkers will facilitate earlier precision medicine interventions for chemotherapy-resistant gastric cancer. The development of therapies targeting PLA2G4A and ACHE could enhance platinum chemotherapy effectiveness.

Application of surface-enhanced Raman spectroscopy to human serum for diagnosing liver cancer

This study investigates the application of surface-enhanced Raman spectroscopy (SERS) in the diagnosis of liver cancer using Ag@SiO2 nanoparticles as SERS substrates. A SERS test was conducted on serum samples obtained from patients with liver cancer and healthy individuals. After repeated several times experiments, it was found that the best SERS spectrum was obtained when the volume ratio of serum to deionized water was 1:2. Moreover, data preprocessing was performed on the tested SERS spectrum, and the preprocessed spectral data were combined with principal component analysis (PCA), partial least-squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA) for further analysis to classify the serum samples of patients with liver cancer and healthy individuals. The results showed that the classification effect of standard normal variate spectral data combined with the OPLS-DA was the best for the serum samples, with a classification accuracy of 97.98%, sensitivity of 97.14%, and specificity of 98.44%. Therefore, the SERS technology can be developed as a favorable method for the accurate diagnosis of liver cancer in the future.

Genome-wide analysis of population structure, effective population size and inbreeding in Iranian and exotic horses

Population structure and genetic diversity are the key parameters to study the breeding history of animals. This research aimed to provide a characterization of the population structure and to compare the effective population size (Ne), LD decay, genetic diversity, and genomic inbreeding in Iranian native Caspian (n = 38), Turkmen (n = 24) and Kurdish (n = 29) breeds and some other exotic horses consisting of Arabian (n = 24), Fell pony (n = 21) and Akhal-Teke (n = 20). A variety of statistical population analysis techniques, such as principal component analysis (PCA), discriminant analysis of principal component (DAPC) and model-based method (STRUCTURE) were employed. The results of the population analysis clearly demonstrated a distinct separation of native and exotic horse breeds and clarified the relationships between studied breeds. The effective population size (Ne) for the last six generations was estimated 54, 49, 37, 35, 27 and 26 for the Caspian, Kurdish, Arabian, Turkmen, Akhal-Teke and Fell pony breeds, respectively. The Caspian breed showed the lowest LD with an average r2 value of 0.079, while the highest was observed in Fell pony (0.148). The highest and lowest average observed heterozygosity were found in the Kurdish breeds (0.346) and Fell pony (0.290) breeds, respectively. The lowest genomic inbreeding coefficient based on run of homozygosity (FROH) and excess of homozygosity (FHOM) was in the Caspian and Kurdish breeds, respectively, while based on genomic relationship matrix) FGRM) and correlation between uniting gametes) FUNI) the lowest genomic inbreeding coefficient was found in the Kurdish breed. The estimation of genomic inbreeding rates in the six breeds revealed that FROH yielded lower estimates compared to the other three methods. Additionally, the Iranian breeds displayed lower levels of inbreeding compared to the exotic breeds. Overall, the findings of this study provide valuable insights for the development of effective breeding management strategies aimed at preserving these horse breeds.

Enhancing forensic investigations: Identifying bloodstains on various substrates through ATR-FTIR spectroscopy combined with machine learning algorithms

The forensic analysis of bloodstains on various substrates plays a crucial role in criminal investigations. This study presents a novel approach for analyzing bloodstains using Attenuated Total Reflectance Fourier Transform Infrared spectroscopy (ATR-FTIR) in combination with machine learning. ATR-FTIR offers non-destructive and non-invasive advantages, requiring minimal sample preparation. By detecting specific chemical bonds in blood components, it enables the differentiation of various body fluids. However, the subjective interpretation of the spectra poses challenges in distinguishing different fluids. To address this, we employ machine learning techniques. Machine learning is extensively used in chemometrics to analyze chemical data, build models, and extract useful information. This includes both unsupervised learning and supervised learning methods, which provide objective characterization and differentiation. The focus of this study was to identify human and porcine blood on substrates using ATR-FTIR spectroscopy. The substrates included paper, plastic, cloth, and wood. Data preprocessing was performed using Principal Component Analysis (PCA) to reduce dimensionality and analyze latent variables. Subsequently, six machine learning algorithms were used to build classification models and compare their performance. These algorithms comprise Partial Least Squares Discriminant Analysis (PLS-DA), Decision Trees (DT), Logistic Regression (LR), Naive Bayes Classifier (NBC), Support Vector Machine (SVM), and Neural Network (NN). The results indicate that the PCA-NN model provides the optimal solution on most substrates. Although ATR-FTIR spectroscopy combined with machine learning effectively identifies bloodstains on substrates, the performance of different identification models still varies based on the type of substrate. The integration of these disciplines enables researchers to harness the power of data-driven approaches for solving complex forensic problems. The objective differentiation of bloodstains using machine learning holds significant implications for criminal investigations. This technique offers a non-destructive, simple, selective, and rapid approach for forensic analysis, thereby assisting forensic scientists and investigators in determining crucial evidence related to bloodstains.

Nasopharyngeal carcinoma cell screening based on the electroporation-SERS spectroscopy

Nasopharyngeal carcinoma (NPC) is a malignant tumor in head and neck. Early diagnosis can effectively improve the survival rate of patients. Nasopharyngeal exfoliative cytology, as a convenient and noninvasive auxiliary diagnostic method, is suitable for the population screening of NPC, but its diagnostic sensitivity is low. In this study, an electroporation-based SERS technique was proposed to detect and screen the clinical nasopharyngeal exfoliated cell samples. Firstly, nasopharyngeal swabs was used to collected the nasopharyngeal exfoliated cell samples from NPC patients (n = 54) and healthy volunteers (n = 60). Then, gold nanoparticles, as the Raman scattering enhancing substrates, were rapidly introduced into cells by electroporation technique for surface-enhanced Raman scattering (SERS) detection. Finally, SERS spectra combined with principal component analysis (PCA) and linear discriminant analysis (LDA) were employed to diagnose and distinguish NPC cell samples. Raman peak assignments combined with spectral differences reflected the biochemical changes associated with NPC, including nucleic acid, amino acid and carbohydrates. Based on the PCA-LDA approach, the sensitivity, specificity and accuracy of 98.15 %, 96.67 % and 97.37 %, respectively, were achieved for screening NPC. This study offers valuable assistance for noninvasive NPC auxiliary diagnosis, and has grate potential in expanding the application of the SERS technique in clinical cell sample testing.

Identification of Commercial Antimalarial Herbal Drugs Using Laser-Induced Autofluorescence Technique and Multivariate Algorithms

In malaria-prone developing countries the integrity of Anti-Malarial Herbal Drugs (AMHDs) which are easily preferred for treatment can be compromised. Currently, existing techniques for identifying AMHDs are destructive. We report on the use of non-destructive and sensitive technique, Laser-Induced-Autofluorescence (LIAF) in combination with multivariate algorithms for identification of AMHDs. The LIAF spectra were recorded from commercially prepared decoction AMHDs purchased from accredited pharmacy shop in Ghana. Deconvolution of the LIAF spectra revealed secondary metabolites belonging to derivatives of alkaloids and classes of phenolic compounds of the AMHDs. Principal Component Analysis (PCA) and Hierarchical Clustering Analysis (HCA) were able to discriminate the AMHDs base on their physicochemical properties. Based on two principal components, the PCA- QDA (Quadratic Discriminant Analysis), PCA-LDA (Linear Discriminant Analysis), PCA-SVM (Support Vector Machine) and PCA-KNN (K-Nearest Neighbour) models were developed with an accuracy performance of 99.0, 99.7, 100.0, and 100%, respectively, in identifying AMHDs. PCA-SVM and PCA-KNN provided the best classification and stability performance. The LIAF technique in combination with multivariate techniques may offer a non-destructive and viable tool for AMHDs identification.

Sex estimation from the hyoid bone measurements in an adult Eastern Turkish population using 3D CT images, discriminant function analysis, support vector machines, and artificial neural networks☆

The hyoid bone is one of the bones in the human body that shows sexual dimorphism. The anthropological and anthropometric characteristics that determine sexual dimorphism are influenced by demographic differences. The aim of this study was to investigate the rate of sexual dimorphism of the hyoid bone in the adult Eastern Turkish population from the examination of the 3D computed tomography images of 240 patients, using discriminant function analysis (DFA), support vector machines (SVM), and artificial neural networks (ANN). These evaluations were based on eight hyoid measurements that have been frequently used in previous CT studies. The results showed that all eight measurements were higher in males than in females (p = 0.000). It was determined that sex could be estimated accurately at up to 93.3 % using DFA, 93.8 % using SVM and 95.4 % using ANN. The maximum accuracy rate achieved to 94.2 % in males using SVM, and 95.8 % in females using ANN. These high rates of sexual dimorphism found using DFA, SVM, and ANN in this study indicate that characteristics of the hyoid bone can be utilized to determine sex in the Eastern Turkish population.

Unraveling volatilomics profiles of milk products from diverse regions in China

To distinguish Chinese milks from different regions, 13 milk samples were gathered from 13 regions of China in this study: Inner Mongolia (IM), Xinjiang (XJ), Hebei (HB), Shanghai (SH), Beijing (BJ), Sichuan (SC), Ningxia (NX), Henan (HN), Tianjin (TJ), Qinghai (QH), Yunnan (YN), Guangxi (GX), and Tibet (XZ). Headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry (HS-SPME-GC-MS) combined with the electronic nose (E-nose) technology, was used to detect and analyze the volatile compounds in these milk samples. The qualitative and quantitative results identified 29 volatile chemicals, and we established a database of flavor profiles for the main milk-producing regions in China. E-nose analysis revealed variations in the odor of milk across different areas. Furthermore, results from partial least squares discriminant analysis (PLS-DA) and odor activity values (OAVs) suggested that seven volatile compounds: decane, 2-heptanone, 2-undecanone, 2-nonanone, 1-hexadecanol, 1-octen-3-ol, and (E)-2-nonenal, could be considered as key flavor compounds in Chinese milk products.

Cookie composition analysis by Fourier transform near infrared spectroscopy coupled to chemometric analysis

The consumption ofcookies is ever growing and during the COVID-19 pandemic reached record consumption values and it is imperative to guarantee the quality and safety of the products.Fourier transform near infrared (FT-NIR) spectroscopy, combined with chemometric techniques, provides a promising solution in that regard, due to its speed and simple sample preparation. The objective of this study was to investigate the possibilities of using FT-NIR to predict lipids, carbohydrates, fibers, proteins, salt and energy contents, as well as to identify cookies type and main cereals present in a batch of 120 commercially acquired samples. The prediction models were performed using ordinary least squares (OLS), partial least squares (PLS), and PLS based classification models including discriminant analysis (PLS-DA), k-nearest neighbors (PLS-kNN) and naïve Bayes (PLS-NB). The best prediction models allowed for good accuracies, with correlation coefficients higher than 0.9 for all studied nutritional parameters. PLS-kNN methodology was able to identify all 5 main cereals (wheat, integral wheat, oat, corn and rice) as well as the 14 types of cookies based on the nutritional contents. The developed methods were able to accurately identify the cookies type and composition, confirming the proposed methodology as a fast, reliable, environmentally friendly and non-destructive alternative to standard analytical methods.

Authenticity assessment of ground black pepper by combining headspace gas-chromatography ion mobility spectrometry and machine learning

Currently, the authentication of ground black pepper is a major concern, creating a need for a rapid, highly sensitive and specific detection tool to prevent the introduction of adulterated batches into the food chain. To this aim, head space gas-chromatography ion mobility spectrometry (HS-GC-IMS), combined with machine learning, is tested in this initial, proof-of-concept study. A broad variety of authentic samples originating from eight countries and three continents were collected and spiked with a range of adulterants, both endogenous sub-products and an assortment of exogenous materials. The method is characterized by no sample preparation and requires 20 min for chromatographic separation and ion mobility data acquisition. After an explorative analysis of the data, those were submitted to two different machine learning algorithms (partial least squared discriminant analysis-PLS-DA and support vector machine-SVM). While the PLS-DA model did not provide fully satisfactory performances, the combination of HS-GC-IMS and SVM successfully classified the samples as authentic, exogenously-adulterated or endogenously-adulterated with an overall accuracy of 90 % and 96 % on withheld test set 1 and withheld test set 2, respectively (at a 95 % confidence level). Some limitations, expected to be mitigated by further research, were encountered in the correct classification of endogenously adulterated ground black pepper. Correct categorization of the ground black pepper samples was not adversely affected by the operator or the time span of data collection (the method development and model challenge were carried out by two operators over 6 months of the study, using ground black pepper harvested between 2015 and 2019). Therefore, HS-GC-IMS, coupled to an intelligent tool, is proposed to: (i) aid in industrial decision-making before utilization of a new batch of ground black pepper in the production chain; (ii) reduce the use of time-consuming conventional analyses and; (iii) increase the number of ground black pepper samples analyzed within an industrial quality control frame.

Demonstrating the empirical effect of population specificity of anthropological standards in a contemporary Australian population

The ability to differentiate individuals based on their biological sex is essential for the creation of an accurate anthropological assessment; it is therefore crucial that the standards that facilitate this are likewise accurate. Given the relative paucity of population-specific anthropological standards formulated specifically for application in the contemporary Australian population, forensic anthropological assessments have historically relied on the application of established methods developed using population geographically and/or temporally disparate. The aim of the present paper is, therefore, to assess the accuracy and reliability of established cranial sex estimation methods, developed from geographically distinct populations, as applied to the contemporary Australian population. Comparison between the original stated accuracy and sex bias values (where applicable) and those achieved after application to the Australian population provides insight into the importance of having anthropological standards optimised for application in specific jurisdictions. The sample analysed comprised computed tomographic (CT) cranial scans of 771 (385 female and 386 male) individuals collected from five Australian states/territories. Cranial CT scans were visualised as three-dimensional volume-rendered reconstructions using OsiriX®. On each cranium, 76 cranial landmarks were acquired, and 36 linear inter-landmark measurements were calculated using MorphDB. A total of 35 predictive models taken from Giles and Elliot (1963), Iscan et al. (1995), Ogawa et al. (2013), Steyn and İşcan (1998) and Kranioti et al. (2008) were tested. Application to the Australian population resulted in an average decrease in accuracy of 21.2%, with an associated sex bias range between - 64.0 and 99.7% (average sex bias value of 29.6%), relative to the original studies. The present investigation has highlighted the inherent inaccuracies of applying models derived from geographically and/or temporally disparate populations. It is, therefore, imperative that statistical models developed from a population consistent with the decedent be used for the estimation of sex in forensic casework.

ML-FGAT: Identification of multi-label protein subcellular localization by interpretable graph attention networks and feature-generative adversarial networks

The prediction of multi-label protein subcellular localization (SCL) is a pivotal area in bioinformatics research. Recent advancements in protein structure research have facilitated the application of graph neural networks. This paper introduces a novel approach termed ML-FGAT. The approach begins by extracting node information of proteins from sequence data, physical-chemical properties, evolutionary insights, and structural details. Subsequently, various evolutionary techniques are integrated to consolidate multi-view information. A linear discriminant analysis framework, grounded on entropy weight, is then employed to reduce the dimensionality of the merged features. To enhance the robustness of the model, the training dataset is augmented using feature-generative adversarial networks. For the primary prediction step, graph attention networks are employed to determine multi-label protein SCL, leveraging both node and neighboring information. The interpretability is enhanced by analyzing the attention weight parameters. The training is based on the Gram-positive bacteria dataset, while validation employs newly constructed datasets: human, virus, Gram-negative bacteria, plant, and SARS-CoV-2. Following a leave-one-out cross-validation procedure, ML-FGAT demonstrates noteworthy superiority in this domain.

Applying Unsupervised Machine Learning Models to Identify Serve Performance Related Indicators in Women's Volleyball

In volleyball, the effect of different factors on serve performance has usually been analyzed with traditional statistical techniques such as logistic regression or discriminant analysis. Purpose: In this study, two of the main models used in unsupervised machine learning (cluster and principal component analysis) were applied to achieve these objectives: (a) to create groups of players considering their serve coefficient, age, height, and team ranking, and (b) to identify which variables related to the serve (type and performance), the players (role, age, and height), and the teams (ranking, match location, and quality of opposition) most explained the total variance of the data during an entire women's volleyball season. Method: A total of 20,936 serves were analyzed during the 132 matches played in the 2017-2018 season in the Liga Iberdrola (women Spanish first division). The variables were related to the serving action (type of serve and performance), the players' traits (player role, age, and height), and the teams' characteristics (final ranking, match location, quality of opposition, and tournament). Results: Cluster analysis showed five groups of players differing in age, serve coefficient, team ranking, and height. Principal component analysis showed how the first five components explained 72.12% of the total variance. From these components, serve coefficient, team ranking, match location, quality of opposition, and player role each contributed more than 10%. Conclusions: These findings can help coaches to improve talent selection and players' development according to competition demands.