Discriminant Analysis Research Articles

Sex estimation through discriminant analysis of the scapula in a contemporary Northern Thai population

Sex estimation through discriminant analysis of the scapula in a contemporary Northern Thai population

Non-targeted metabolomics revealed the effect of starvation to juvenile Onychostoma sima liver.

Artificial breeding and releasing can effectively restore fishery resources. However, it is important to note that released juvenile fish were highly susceptible to starvation during their adaptation to the natural environment. This study investigated the metabolomic changes in the liver of Onychostoma sima after 14days using ultra-high pressure liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF/MS) analysis under starvation exposure. The experiment was divided into a control group (C Group) and a starvation group (S Group), with six biological replicates in each group and one fish per replicate. The results indicated significant changes in the starvation group compared to the control group, as demonstrated by the principal component analysis (PCA) score plots and orthogonal partial least squares discriminant analysis (OPLS-DA). The 297 differential metabolites screened were mainly involved in the metabolism of organic acids and derivatives, and lipids and lipid-like molecules. Furthermore, KEGG results revealed that differential metabolites were primarily enriched in 33 metabolic pathways. The majority of the amino acid metabolic pathways in the liver were significantly affected by starvation stress. Moreover, biosynthesis of amino acids, protein digestion and absorption, and mineral absorption were upregulated, while glycerophospholipid metabolism and the hedgehog signaling pathway were downregulated in response to energy demands during starvation. In conclusion, these findings provide physiological insights into the metabolism of juvenile O. sima under starvation stress, offering new perspectives for the optimization of fish proliferation and release technology.

Traceability and discrimination of opium poppy shell analogues using HS-GC-IMS combined with machine learning algorithms.

Illegal adulteration has been a critical issue in food safety, emerging as a focal point in forensic science. This situation has led to an increased demand for effective detection and monitoring technologies. Opium poppy shells are a critical source of drugs, and the accurate tracing and identification of their analogues are essential in drug-related cases. The features of volatile compounds in six opium poppy shell analogues (OPSA) were characterized using headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS) in this study, and an accurate model for origin tracing was established through the integration of machine learning algorithms. A total of 213 volatile compounds were accurately identified, with esters, ketones, aldehydes, alcohols, and alkenes being the most abundant compounds. Additionally, two supervised machine learning algorithm classification models were established based on the HS-GC-IMS dataset to predict the categories of OPSA, including the orthogonal partial least squares discriminant analysis (OPLS-DA) and random forest models, and were subsequently compared with unsupervised models. By employing the random forest classification model, significant volatile compound characteristics were recognized, resulting in enhanced efficiency. Furthermore, the model achieved an out-of-bag (OOB) error value of 0, indicating excellent predictive capability for tracing and distinguishing OPSA. Our research findings indicate that the integration of HS-GC-IMS with machine learning is expected to enhance the efficiency and accuracy of tracing and identifying the categories of OPSA, thereby providing theoretical support for litigation and judicial processes.

Indirect Detection of Swine Influenza Activity in Porcine Blood Using Raman Spectroscopy and Machine Learning.

Over the past decade, several swine influenza variants, including H1N1 and H1N2, have been periodically detected in swine. Raman spectroscopy (RS) offers a non-destructive, label-free, and rapid method for detecting pathogens by analyzing molecular vibrations to capture biochemical changes in samples. In this study, we examined blood serum from swine under different conditions: healthy, unvaccinated, or vaccinated against porcine reproductive and respiratory syndrome, and vaccinated swine infected with H1N1 and H1N2 variants of swine influenza. Our findings demonstrate that RS, when combined with machine learning algorithms such as partial least squares discriminant analysis and eXtreme gradient boosting discriminant analysis, can achieve accuracy rates of up to 97.8% in identifying the infection status and specific variant within porcine blood serum. This research highlights RS as a useful, novel tool for the detection of influenza variants in swine, significantly enhancing surveillance efforts by identifying animal health threats.

GC-MS based nutritional and aroma profiling of date palm seeds collected from different Egyptian cultivars for valorization purposes

Date palm (Phoenix dactylifera L.) is a globally edible fruit and a traditional dietary component in various cultures. The fruit’s fleshy part is consumed for its nutritional value, while the seeds are discarded or valorized for oil production and as a coffee substitute. The current study aimed to investigate date seeds’ metabolome, in addition to macro- and micro-elements composition within12 major Egyptian cultivars (cvs.) for the first time using gas chromatography coupled with mass spectrometry (GC-MS). Post-silylation GC-MS analysis and headspace coupled with solid-phase microextraction (HS-SPME) were used for nutrients and aroma profiling in roasted seeds, respectively. Furthermore, multivariate data analyses were employed for samples classification and markers identification via principal component analysis (PCA) and orthogonal projection to least square discriminant analysis (OPLS-DA). Models are further validated by permutation test. Moreover, absolute quantification of potential markers was attempted based on reference standards A total of 101 and 65 nutrient and aroma metabolites were annotated, respectively. Fatty acids/esters (38 peaks), sugars (18), organic acids (17), sugar alcohols (7), steroids/triterpenoids (5), alcohols and aldehydes (6), in addition to flavonoids (1) and phenolic acids (3) were identified as major components in GC-MS post-silylation platform. ‘’Khalas’’ cv. seed appeared the most nutritive being enriched in sugars and fatty acids/esters. Moreover, date seed volatiles from different cvs. were dominated by fatty acids/esters (19 peaks), esters (6), and phenols/ethers (9). Anethole (peak 47) was the most abundant at 9.1–23.3% of seeds contributing to their unique aroma, especially ‘’Barhi’’ a premium date cv. PCA score plot of primary metabolites’ dataset revealed for 1-monopalmitin and monostearin as potential markers for ‘’Aref’’ and ‘’Khalas’’. Furthermore, ‘’Barhi’’, ‘’Omeldehn’’, and ‘’Lolo’’ cvs. showed comparable aroma profile and in partial agreement with nutrient results. OPLS-DA model revealed that anethole, estragole, methyl esters of dodecanoic acid and octanoic acid were characteristic in case of ‘’Barhi’’ cv. which are likely to impart a fine aroma and flavor. With regards to minerals, ‘’Zamli’’, ‘’Barhi’’, and ‘’Hasawi’’ cvs. were most rich in calcium, copper, and selenium, respectively. This study offers new perspectives for the phytochemical makeup and valorization potentials of date palm seeds. Fatty acids/esters and sugars were the major components in date palm seeds found enriched in ‘’Khalas’’ cv, while anethole, estragole, methyl esters of dodecanoic acid and octanoic acid were potential markers of ‘’Barhi’’ cultivar. Such extensive profiling identified premium cvs. to be considered for food applications.

Leaf Size Indices and Outline-Based Geomorphometric Analysis of Five Philippine Endemic Saurauia Willd. (Actinidiaceae)

Species discrimination among species of Saurauia is challenging due to large morphological variation. This study examines the intraspecific variations of the 5 Philippine endemic Saurauia species using leaf size indices (LSI) and outline-based geometric morphometrics to facilitate species discrimination. Leaf samples were measured using the traditional method, scanned, converted to binary images, subjected to elliptic Fourier Analyses, and quantitatively analyzed using principal component analysis (PCA). The leaf morphology significantly differed among species based on the results of LSI and leaf shape outline analyses. The results showed 7 effective principal components (PCs), which accounted for 94.16% of the total variation. Significant differences were observed in all PCs. Discriminant analysis of the leaf shape outline confirmed the delimitation of species with scores relatively higher than the cut-off value. The tree topology from leaf shape outline, and leaf size indices all exhibited similarity in the clustering at the species level. A key to the species based on leaf morphology is also provided. Keywords: elliptic fourier analysis, kiwi, leaf size index, leaf variation, principal component analysis

Multivariate analysis of DTMS data for elucidating the chemistry behind soft and dripping paint

The problem of soft and dripping oil paint in mid-20th-century paintings was investigated using high-resolution direct temperature-resolved mass spectrometry (DTMS) and multivariate data analysis techniques. Nineteen paintings by Asger Jorn, Karel Appel, Pierre Alechinsky and Jean-Paul Riopelle were examined, and selected areas were tested for material properties and water sensitivity. Samples were taken from these areas and subjected to DTMS, and the data collected were analysed with principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). PCA loadings showed grouping of variables belonging to similar chemical structures, but the scores showed only clear clustering of samples classified as solid and healthy. The other categories showed large variability in chemical composition. No clustering on water sensitivity was found. With PLS-DA, chemical markers were detected for the solid and healthy samples that are correlated to the presence of fish oil. Further investigation showed strong indications for the presence of chemical components related to fish oil in many samples from the other categories as well. Earlier hypotheses on the issue of soft and dripping oil paint, based on observations on a single or small group of paintings, could not be confirmed. In this study, it was also shown that using low resolution DTMS may lead to incorrect conclusions, because different chemical components with the same nominal mass can be anti-correlated.

Machine-Learning-Assisted Aroma Profile Prediction in Five Different Quality Grades of Nongxiangxing Baijiu Fermented During Summer Using Sensory Evaluation Combined with GC×GC–TOF-MS

Flavor is one of the crucial factors that influences the quality and consumer acceptance of baijiu. In this study, we analyzed the volatile organic compound (VOC) profiles of five different quality grades of Nongxiangxing baijiu (NXB), fermented during the summer of 2024, using 2D gas chromatography time-of-flight mass spectrometry (GC×GC–TOF-MS). We employed machine-learning (ML)-based classification and prediction models to evaluate the flavor. For TW, the scores of the sensory evaluation for coordination and overall evaluation were the highest. TW contained the highest concentration of ethyl caproate; we detected 965 VOCs in total, including several pyrazine compounds with potential health benefits. Principal component analysis (PCA) combined with orthogonal partial least squares discriminant analysis (OPLS-DA) enabled us to distinguish different samples, with eight VOCs emerging as primary contributors to the aroma of the samples, possessing variable importance in projection (VIP) values > 1. Furthermore, we tested eight ML models; random forest (RF) demonstrated the best classification performance, effectively discriminating samples based on their VOC profiles. The key VOC contributors that showed quality-grade specificity included 1-butanol, 3-methyl-1-butanol, and ethyl 5-methylhexanoate. The results elucidate the flavor-based features of NXB and provide a valuable reference for discriminating and predicting baijiu flavors.

The impact of spectral data pre-processing on the assessment of red wine vintage through spectroscopic methods.

Red wine is a common target of fraudulent acts considering its high market value and popularity. Although there has been much effort to assess the geographical and varietal origin of wine, this is not the case for wine vintage. Vintage is a crucial parameter for the market price, especially in the case of reputable wines. Considering the season-to-season variations affecting wine quality and the ever-occurring unstable climatological conditions due to climate change, developing analytical strategies to accurately assess wine vintage is topical and of high interest. In this study, we successfully employed ultraviolet-visible spectroscopy, fluorescence spectroscopy and mid-infrared spectroscopy to identify the vintage of a protected designation of origin red wine produced during four different vintages (n = 36). Class-based clustering and great discriminatory performance was achieved for the majority of the developed multivariate models and the impact of the applied spectral pre-processing was significant. Importantly, the tested scatter correction methods resulted in the best cross-validation parameters (goodness of fit, R2Y > 0.9 and goodness of prediction, Q2Y > 0.8) with calculated recognition and prediction abilities in the range 77-100% and 65-96%, respectively, when using partial least squares discriminant analysis. In addition, in the case of fluorescence spectroscopy, a batch effect was revealed, which was compensated by the spectral pre-processing methods. Spectral feature selection was performed in all cases to use only the analytically important spectral signals and omit model overfitting. The developed method is simple, cost-efficient and non-destructive, indicating its high potential for industrial applications as a rapid screening tool. © 2025 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Rapid Detection of Antibiotic Mycelial Dregs Adulteration in Single-Cell Protein Feed by HS-GC-IMS and Chemometrics

Single-cell protein feed (SCPF) is an important supplement to protein feed materials, but its authenticity is often affected by antibiotic mycelial dregs (AMD). Headspace-gas chromatography–ion mobility spectrometry (HS-GC-IMS), integrated with chemometrics, was utilized to differentiate nucleotide residue (NR), three AMDs, and adulterated samples with concentrations ranging from 0.1% to 20% (w/w). Orthogonal partial least squares discriminant analysis (OPLS-DA) and principal component analysis (PCA) were applied to classify the adulterated samples. In addition, the feasibility of quantitative analysis of the AMDs content in adulterated SCPF based on partial least squares regression (PLSR) algorithm. In total, 88 volatile organic compounds (VOCs) were detected. The differences in VOCs between NR and AMD mainly came from aldehydes, alcohols, and esters. The OPLS-DA models effectively identified AMD in adulterated NR samples (Accuracy = 100%), demonstrating the HS-GC-IMS data’s good application potential for the SCPF adulteration. Nine VOCs, i.e., 2-ethyl-3-methylpyrazine, dihydro-5-methyl-2(3H)-furanone, 2-methylpropanol, (E,E)-2,4-heptadienal, linalool, 2,3,5-trimethylpyrazine, citronellol, acetoin, and 3-methylbutan-1-ol, were proposed as key markers for detecting NR adulterated with AMDs. The PLSR algorithm was further used to determine the AMD content in NR (R2cal = 0.96, R2cv = 0.94). This study validated HS-GC-IMS’s ability to analyze volatile organic compounds in feed and showcased its utility as a convenient, quick, and affordable tool for SCPF authenticity screening.

Serum lipid metabolites predict the efficacy of intravenous immune globulin therapy in patients with pediatric dilated cardiomyopathy

The response of pediatric dilated cardiomyopathy (PDCM) patients to intravenous immune globulin (IVIG) varies from cardiac functional recovery to heart transplantation and even death. IVIG therapy can significantly improve the left ventricular ejection fraction (LVEF) in some PDCM patients, but indicators for evaluating the response to IVIG therapy are lacking. Lipid metabolic disturbance is associated with changes in cardiac function, but no studies have examined the associations between lipidomics markers and the efficacy of IVIG. Discovery analyses were based on 322 targeted lipids in a retrospective cohort study. T tests, orthogonal‒orthogonal projections to latent structures discriminant analysis (OPLS-DA) and random forest (RF) analysis were used to screen the candidate lipids. Associations of the candidate lipids were examined via Cox proportional hazards regression models. We subsequently developed a score for the candidate lipid metabolites, which was then used to classify children into two groups (0–3 and 4), and the survival curves of the two groups were drawn. There were 31 patients in the discovery set and 24 patients in the validation set. Adverse events which were defined as death, heart transplant, and rehospitalization for HF were observed in 23 patients (41.8%). After adjusting for age, LVEF, and the LV end-diastolic diameter z score, CE-16:1, PE36:4p, PE40:6p, and PE40:6p (22:6) were significantly associated with the occurrence of adverse events in the discovery set. The results were also consistent in the validation set, and the hazard ratios (HRs) were 2.638 (95% CI 1.042–6.683; p = 0.041), 0.549 (95% CI 0.340–0.886; p = 0.014), 0.271 (95% CI 0.109–0.672; p = 0.005), and 0.299 (95% CI 0.121–0.741; p = 0.009), respectively. We developed a score for these four lipid metabolites. When the score reached 4, adverse events were more likely to be observed in both sets. Serum lipid metabolites can be used to predict the efficacy of IVIG in children with DCM.

Identification of Occupations in Different Populations Based on Skin Microbial Characteristics.

The composition and diversity of the skin microbiome are affected by several factors, including the working environment, which plays an active role in shaping microbial communities in human skin. Previous studies have shown that residual microbial communities on personal items can be used to identify their owners. However, few studies have used the skin microbiome to identify occupations in different populations or evaluate whether the skin microbiome can be used as a tool for forensic investigations. Here, we collected palm and cuff swabs from three occupational groups-cooks, medical staff, and students-and performed next-generation sequencing targeting the 16S rRNA gene to characterise the microbial communities associated with each profession. We found that different occupational environments resulted in different skin microbial community compositions. Actinobacteria and Firmicutes were the dominant phyla in the student samples. Compared with the other two occupations, cooks had the highest relative abundances of Bacteroides and Cyanobacteria. Additionally, cuff samples from medical staff had the highest relative abundances of Proteobacteria. Principal co-ordinate analysis results indicated that the samples were roughly divided into three clusters according to their occupation. Furthermore, linear discriminant analysis effect size results showed that cooks, medical staff, and students had their own unique biomarkers, cooks exhibited seven shared biomarkers between palm and cuff samples, medical staff showed 2, while students demonstrated the highest congruence which was 13 shared biomarkers. This suggested that some palm skin microbial communities could be transferred to the cuffs through contact friction. Thus, there were also microbial communities present in cuff samples that could be used to identify the owner's occupations, suggesting that skin microorganisms left on personal items via daily contact could also be used to provide information about an individual's occupation. Finally, we constructed a random forest model based on the composition and relative abundance of the microbiota to infer the subject's occupation, achieving an accuracy of 76.92% for the palm testing dataset and 73.33% for the cuff testing dataset; all of the cuff sample datasets showed an accuracy of 70.97%. These findings suggested that an individual's occupation can be inferred not only from the skin microbiota but also from the microbiota left on the cuffs of the individual's clothes. Further studies are needed; however, these results demonstrate the potential of the skin microbiota as a forensic tool for predicting population occupations.

Non-targeted metabolomic analysis of follicular fluid in infertile individuals with poor ovarian response

BackgroundPoor ovarian response (POR) is a pathological condition characterized by inadequate ovarian response to gonadotropin stimulation in patients undergoing in vitro fertilization and embryo transfer. It represents a primary cause of failure in many assisted reproductive technology treatments. Utilizing non-targeted metabolomics technology applied to follicular fluid, this research aims to elucidate the metabolic characteristics associated with POR, explore the underlying molecular mechanisms, and identify potential biomarkers. By analyzing metabolic factors that influence oocyte quality, we aspire to provide insights for the early detection and intervention of patients with POR.MethodsIn this research, 60 follicular fluid samples were collected for a non-targeted metabolomic study, including 30 samples from POR patients and 30 from women with normal ovarian reserve. The orthogonal partial least squares discriminant analysis model was employed to discern separation trends between the two groups. Pathway enrichment analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Additionally, random forest and logistic regression models were utilized to identify biomarkers indicative of POR within the follicular fluid.ResultsBased on data from the Human Metabolome Database, our metabolomic analysis identified 40 differential metabolites associated with POR, including 18 up-regulated and 22 down-regulated metabolites. KEGG pathway analysis revealed that these metabolites predominantly participate in glycerophospholipid metabolism, choline metabolism in cancer, autophagy processes. Notably, perillyl aldehyde emerged as a potential biomarker for POR.ConclusionsThis study represents the first comprehensive examination of metabolic alterations in follicular fluid among patients with POR using non-targeted metabolomics technology. We have identified significant metabolic changes within the follicular fluid of individuals affected by POR which may offer valuable insights into therapeutic strategies for managing this condition as well as improving outcomes in assisted reproductive technologies.

Intelligent Selection of Spectral Bands from High-Precision Spectroradiometer Measurements for Optimizing Cocoa Bean Classification

Evaluating the spectral properties of cocoa beans based on their fermentation state (fermented, in a poor state, unfermented) is essential for ensuring their quality in the cocoa industry. This study examined the spectral response of beans in the range of 380 nm to 780 nm using the Konica-Minolta CS-2000 spectrophotometer comes from Dijon, France, a device designed to measure the spectrum of objects and sources in the visible range. Different spectral band selection methods have been applied to identify the most discriminating wavelengths for their classification. Several techniques were used: ANOVA, F-score, Lasso, Linear Discriminant Analysis (LDA), Mutual Information, and Partial Least Squares (PLS). A band selector voting process was implemented to determine standard wavelengths identified using the different methods. The selected spectral bands were then leveraged to train classification models, including Random Forest, SVM, and XGBoost. The results show that a restricted subset of wavelengths allows for effective class separation, thereby improving model performance. Among the approaches tested, ANOVA and F-score combined with Random Forest achieved an accuracy of 92.59%, while F-score and Mutual Information coupled with SVM and voting associated with SVM obtained an accuracy of 96.30%. These feature selection methods have effectively reduced dimensionality while maintaining high classification accuracy. These results open up promising prospects for the automation of quality control of cocoa beans, thus contributing to the optimization of industrial processes.

Detection of 2,4,6-Trichloroanisole in Sparkling Wines Using a Portable E-Nose and Chemometric Tools

This study addresses the contamination of sparkling wines by 2,4,6-trichloroanisole (TCA), a compound responsible for the “cork taint” or musty aroma in wines. Currently, its detection requires complex and expensive techniques such as chromatography and sensory panels. An innovative method is proposed using an electronic nose (e-nose) prototype, offering objective, non-destructive, and cost-effective analysis. The e-nose’s ability to detect TCA at various concentrations was evaluated in sparkling wines from different batches and a spiked wine sample. The results analyzed using Principal Component Analysis (PCA) successfully differentiated the samples. An Artificial Neural Network Discriminant Analysis (ANNDA) classified wines based on whether their TCA concentration exceeded 2 ng/L, achieving 88% accuracy. A quantitative predictive model using Partial Least Squares (PLS) analysis yielded an R2 of 0.84 across wines and 0.95 in a single sample. These advances highlight the potential of the e-nose to improve quality control in the wine industry.

Comparative analysis of composition and spatial variations in the foregut microbiota of male and female donkeys

Donkeys, as significant herbivorous mammals, also serve as valuable companion animals. Research on gut microbiota has underscored the essential role of microorganisms in maintaining gut health, supporting nutrient metabolism, and regulating immune function. As the gut microbiota is also shaped by factors such as sex, age, diet, environment and genetics, many studies have on the complexity and diversity of hindgut microbial communities, while few studies have focused on the foregut microbiota of donkeys. To address this gap, we conducted high-throughput sequencing of the highly variable V3-V4 region of the 16S rRNA gene from the donkey small intestine (duodenum, jejunum, and ileum) to characterize and compare microbiota composition and abundance between male and female donkeys. A total of 12 healthy and uniformly conditioned Dezhou donkeys (six males and six females, aged 2–3 years, 250 ± 10 kg in weight) were included in the study. The results showed that albumin (ALB), total cholesterol (TC), and high-density lipoprotein cholesterol (HDL-C) levels were significantly higher (p < 0.05) in the female group compared to the male group. Additionally, α-diversity indices (Ace, Chao, Simpson, and Sobs) were significantly different (p < 0.05) between the groups. The PCoA results indicated significant differences (p < 0.05) between male and female donkeys across all intestinal locations (R2 = 0.2372, p < 0.001). Similarly, the microbial composition of the jejunum (R2 = 0.1875, p = 0.019) and ileum (R2 = 0.1776, p = 0.007) showed significant differences between male and female donkeys. Additionally, Firmicutes, Fusobacteriota, Proteobacteria, and Actinobacteriota were the dominant phyla across all gut regions. In male and female donkeys, key genera included Lactobacillus, Streptococcus, Sarcina, and Escherichia-Shigella. Linear discriminant analysis effect size (LEfSe) analysis revealed gender-specific enrichment, with Clostridium_sensu_stricto_1, Acinetobacter, and NK4A214_group dominant in female duodenum and jejunum, while Streptococcus and Erysipelotrichaceae_UCG-002 were enriched in males. Similarly, female ileum had enriched Amnipila, Terrisporobacter, and Luteimonas, whereas males showed higher levels of Sarcina and Streptococcus. Blautia and Mogibacterium were enriched in female duodenum and jejunum, while Fusobacterium, Actinobacillus, and Moraxella were more abundant in male ileum. These findings characterize the gut microbiota of healthy donkeys and provide novel insights into the differences between male and female donkeys, offering previously unknown information about donkey gut microbiota.

Diet composition and feeding habits of Meretrix meretrix and Mactra veneriformis in the northern Bohai Sea based on high-throughput sequencing

Understanding the diet composition and feeding habits of bivalve shellfish is crucial for developing conservation measures to enhance their resources. This is particularly important for the main economic species in shellfish-producing regions. In this study, we analyzed the stomach contents composition of the two main economic shellfish in Geligang, specifically Meretrix meretrix and Mactra veneriformis, using high-throughput sequencing. The results revealed that 956 operational taxonomic units (OTUs) were common to both M. meretrix and M. veneriformis, with 1117 OTUs unique to M. meretrix and 412 OTUs unique to M. veneriformis. We identified a total of 50 bait organisms from 11 phyla. The main taxa in the stomach contents of M. meretrix were Chlorophyta, Cryptophyta, Pyrrophyta and Bacillariophyta, while Cryptophyta, Chlorophyta, Pyrrophyta and Chrysophyta dominated the stomach contents of M. veneriformis. Non-metric multidimensional scaling (NMDS) analysis indicated less compositional variety in the stomach contents of M. meretrix compared to M. veneriformis. Additionally, the Linear Discriminant Analysis Effect Size (LEfSe) results showed a significant difference in food composition between the two species. Specifically, M. meretrix and M. veneriformis preferred feeding on Bacillariophyta, Chlorophyta, and Cryptophyta, while M. veneriformis favored Chrysophyta. Overall, our study provides fundamental insights for ecological research on feeding habits and resource conservation of M. meretrix and M. veneriformis in Geligang, which can inform the development of effective conservation measures for the shellfish resources.

Assessing and validating machine learning-enhanced imputation of admission American Spinal Injury Association Impairment Scale grades for spinal cord injury.

The American Spinal Injury Association Impairment Scale (AIS) assigned at patient admission is an important predictor of outcomes following spinal cord injury (SCI). However, nearly 80% of records in the Spinal Cord Injury Model Systems (SCIMS) database-a multicenter prospective database of patients with SCI-lack admission AIS grades. Accurate imputation of this missing data could enable more robust analyses and insights into SCI recovery. This study aims to develop and validate methods for imputing missing admission AIS data in the SCIMS database. The study included 16,062 patients with SCI from the publicly available SCIMS database (1988-2020). Five machine learning algorithms-random forest (RF), linear discriminant analysis, K-nearest neighbors, naive Bayes, and support vector machine-were compared using performance metrics (accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and multiclass area under the receiver operating characteristic curve) using five-fold cross-validation on a training subset of 6054 patients with complete AIS admission grades. The model with the highest performance was trained on all 16,062 patients. The imputed AIS grades were validated by predicting discharge functional independence measure (FIM) scores (range 13-91) with simple and multiple linear regression models on a 1:1 propensity score-matched cohort (n = 5828). Model performance was compared using differences in root mean square error (∆RMSE) with bootstrapped 95% confidence intervals (CIs). The full cohort contained a representative distribution of AIS grades (45% grade A, 13% grade B, 18% grade C, and 24% grade D), and the propensity score-matched cohort characteristics were well balanced. The RF algorithm demonstrated the highest validation accuracy (81.7%). Predictive models showed no significant differences between models using true versus imputed AIS grades, with 95% CIs for ∆RMSE of -0.60 to 0.47 for simple regression and -0.63 to 0.46 for multiple regression models. The coefficients of AIS grades also did not significantly differ between models with true versus imputed values. A data-driven approach to imputation resulted in a robust method for imputing admission AIS grades that demonstrated clinical validity in the SCIMS database. This approach extends the utility of this longitudinal database and may provide a framework for other SCI databases.

Mediterranean Diet Modulates Gene Expression of Cholesterol Efflux Receptors in High-Risk Cardiovascular Patients.

In this study, we investigated gene expression related to cholesterol efflux receptors in individuals at high cardiovascular risk undergoing Mediterranean dietary interventions. Through transcriptomic analysis, we examined samples from two randomized controlled trials: PREDIMED and PREDIMED-Plus, with 151 and 89 elderly adults, respectively. Blood cells were isolated at baseline and after a 12-month intervention. In the PREDIMED trial, participants followed different Mediterranean diets: one supplemented with extra-virgin olive oil (traditional Mediterranean diet enriched with extra-virgin olive oil [MedDiet-EVOO]), another with nuts (MedDiet enriched with nuts MedDiet-Nuts [MedDiet-Nuts]), and a low-fat control diet. The PREDIMED-Plus trial compared an energy-reduced Mediterranean diet (Er-MedDiet) with physical activity to an ad libitum Mediterranean diet. Over time, mild but significant upregulation of genes like ATP binding cassette subfamily A member 1 (ABCA1), retinoid X receptor alpha (RXRA), retinoid X receptor beta (RXRB), and Nuclear Receptor Subfamily 1 Group H Member 3 (NR1H3) was observed in response to MedDiet-EVOO, MedDiet-Nuts, and Er-MedDiet. Notably, RXRA expression was higher in both MedDiet-EVOO and MedDiet-Nuts compared to the control diet. Differences in gene expression, particularly RXRA, ATP binding cassette subfamily G member 1 (ABCG1), NR1H3, and Peroxisome Proliferator Activated Receptor Delta (PPARD), were evident between MedDiet-Nuts and the control diet. In the PREDIMED-Plus trial, no significant differences in gene expression were found between dietary groups. Principal component analysis (PCA) and linear discriminant analysis (LDA) showed overlapping gene expression profiles across different Mediterranean diet interventions. In conclusion, our study highlights the cardiovascular health benefits of long-term adherence to a Mediterranean diet, both normocaloric and hypocaloric, primarily reflected by mild upregulation of cholesterol efflux-related genes-specifically involving RXRA, RXRB, ABCA1, ABCG1, Nuclear Receptor Subfamily 1 Group H Member 2(NR1H2), and PPARD-among elderly adults at high cardiovascular risk. This suggests a potential mechanism by which these diets may exert cardiovascular protective effects.

Identifying the Geographical Origin of Wolfberry Using Near-Infrared Spectroscopy and Stacking-Orthogonal Linear Discriminant Analysis

The geographical origin identification of wolfberry is key to ensuring its medicinal and edible quality. To accurately identify the geographical origin, the Stacking-Orthogonal Linear Discriminant Analysis (OLDA) algorithm was proposed by combining OLDA with the Stacking ensemble learning framework. In this study, Savitzky–Golay (SG) + Multiplicative Scatter Correction (MSC) served as the optimal preprocessing method. Four classifiers—K-Nearest Neighbors (KNN), Decision Tree, Support Vector Machine (SVM), and Naive Bayes—were used to explore 12 stacked combinations on 400 samples from five regions in Gansu: Zhangye, Yumen, Wuwei, Baiyin, and Dunhuang. When Principal Component Analysis (PCA), PCA + Linear Discriminant Analysis (LDA), and OLDA were used for feature extraction, Stacking-OLDA achieved the highest average identification accuracy of 99%. The overall accuracy of stacked combinations was generally higher than that of single-classifier models. This study also assessed the role of different classifiers in different combinations, finding that Stacking-OLDA combined with KNN as the meta-classifier achieved the highest accuracy. Experimental results demonstrate that Stacking-OLDA has excellent classification performance, providing an effective approach for the accurate classification of wolfberry origins and offering an innovative solution for quality control in the food industry.