What is principal component analysis?

Methyl jasmonate enabled maintained the postharvest flavor quality of ginger (Zingiber officinale roscoe) by reducing the loss of terpene volatile compounds

Ginger, as a globally vital medicinal and food homologous crop, plays an irreplaceable role in human diet and healthcare. However, during the storage of ginger, the decline of physical properties and degradation of volatile flavor quality have emerged as an industrial concern that severely restricts the market value of the product. MeJA plays an essential role in extending fruit shelf life and regulate the synthesis of volatiles in horticultural products, yet its application in ginger remains unreported. This study investigated whether MeJA could delay the deterioration of external quality and the loss of volatile compounds, thereby maintaining the flavor quality of ginger during storage. The results demonstrated that MeJA retarded weight loss, moisture reduction, texture softening, and color darkening in ginger rhizomes during storage. In addition, dynamic profiles of volatile compounds in the postharvest stage of ginger rhizomes were characterized via HS-SPME/GC-MS methodology. A total of 67 volatile components were identified and quantified precisely, which were divided into terpenes, alcohols, esters, aldehydes, ketones, and others. Terpenes represented by zingiberene, farnesene, β-sesquiphellandrene, α-curcumene, (E)-β-farnesene, and β-elemene, was the most abundant classification of compounds in ginger, comprising approximately 70 % of the total content. Compared with the control group, MeJA reduced the loss rate of total quantity and total content of volatiles, while effectively slowed the loss of various volatiles, especially after 35d of storage. Furthermore, 30 characteristic components with an odor activity values (OAVs) ≥ 1 were identified, predominantly exhibiting spicy, green, floral, fatty, and fruity fragrances. It is noteworthy that the most prominent scent of ginger is the spicy aroma, which can be significantly up-regulated by MeJA. Moreover, MeJA treatment was found to enhance the expression levels of terpene-related genes in ginger. This study clarified the patterns of variation in physical properties, volatile compounds, and aroma intensity during the storage of ginger, providing a theoretical basis for mitigating the deterioration of flavor quality in ginger rhizomes during postharvest storage. This research holds significant importance for promoting the comprehensive utilization and high-quality development of ginger.

Analysis of Key Differential Metabolites in Intervertebral Disc Degeneration Based on Untargeted Metabolomics

Background

Intervertebral disc degeneration disease (IVDD) is a prevalent orthopedic condition that causes chronic lower back pain, imposing a substantial economic burden on patients and society. Despite its high incidence, the pathophysiological mechanisms of IVDD remain incompletely understood.

Objective

This study aimed to identify metabolomic alterations in IVDD patients and explore the key metabolic pathways and metabolites involved in its pathogenesis.

Methods

Serum samples from 20 IVDD patients and 20 healthy controls were analyzed using ultra-high-performance liquid chromatography-mass spectrometry (UHPLC-MS). The identified metabolites were mapped to metabolic pathways using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database.

Results

Significant alterations were observed in metabolites such as 2-methyl-1,3-cyclohexadiene, stearoyl sphingomyelin, methylcysteine, L-methionine, and cis, cis-muconic acid. These metabolites were involved in pathways including glycine, serine, and threonine metabolism, cyanoamino acid metabolism, and the citrate cycle (TCA cycle).

Conclusion

The identified metabolic alterations provide insights into the pathogenesis of IVDD and suggest potential therapeutic targets for future investigation.

An automatic control system based on machine vision and deep learning for car windscreen clean

Raindrops on the windscreen significantly impact a driver's visibility during driving, affecting safe driving. Maintaining a clear windscreen is crucial for drivers to mitigate accident risks in rainy conditions. A real-time rain detection system and an innovative wiper control method are introduced based on machine vision and deep learning. An all-weather raindrop detection model is constructed using a convolutional neural network (CNN) architecture, utilising an improved YOLOv8 model. The all-weather model achieved a precision rate of 0.89, a recall rate of 0.83, and a detection speed of 63 fps, meeting the system's real-time requirements. The raindrop area ratio is computed through target detection, which facilitates the assessment of rainfall begins and ends, as well as intensity variations. When the raindrop area ratio exceeds the wiper activation threshold, the wiper starts, and when the area ratio approaches zero, the wiper stops. The wiper control method can automatically adjust the detection frequency and the wiper operating speed according to changes in rainfall intensity. The wiper activation threshold can be adjusted to make the wiper operation more in line with the driver's habits.

Computational Elucidation of Human β-Defensin-2 as a Dual Inhibitor of MMP-9 and PKC-βII for Diabetic Wound Management

Diabetic wounds (DWs) are the most devastating complication, resulting in significant mortality and morbidity in diabetic patients. Although the pathophysiology of DWs is multifaceted, evidence has revealed that prolonged inflammation with infections, extracellular matrix (ECM) degradation, and unnecessary NETosis impair DW healing. This theoretical problem highlights the necessity of developing a novel strategy focused on targeting the "specific" molecular modalities of DWs. The primary culprits, matrix metalloproteinase (MMP)-9 and protein kinase C (PKC)-βII, are responsible for impaired angiogenesis, NETosis, and ECM degradation. Thus, interest in identifying selective inhibitors for the effective management of DW has increased. The current study exemplified human β-defensin-2 (HBD-2), a biological macromolecule that functions as a dual inhibitor of MMP-9 and PKC-βII, via protein-protein docking and molecular dynamics simulation studies. Overall, the data analysis revealed that HBD-2 possesses strong binding affinity and stability against MMP-9 and PKC-βII, suggesting that HBD-2 may be an ideal therapeutic for the accelerated healing of DW. Our findings suggest HBD-2's potential as an innovative therapeutic for accelerated DW healing, offering valuable insights into its molecular mechanisms. However, in vitro and in vivo studies are required to bridge the gap between computational modeling and clinical application.

PPARγ modulator predictor (PGMP_v1): chemical space exploration and computational insights for enhanced type 2 diabetes mellitus management

Peroxisome proliferator-activated receptor gamma (PPARγ) plays a critical role in adipocyte differentiation and enhances insulin sensitivity. In contemporary drug discovery, in silico design strategies offer significant advantages by revealing essential structural insights for lead optimization. The study is guided by two main objectives: (i) a ligand-based approach to explore the chemical space of PPARγ modulators followed by molecular docking ensembles (MDEs) to investigate ligand-binding interactions, (ii) the development of a supervised ML model for a large dataset of compounds targeting PPARγ. Additionally, the combination of chemical space networks with ML models enables the rapid screening and prediction of PPARγ modulators. These modeling analyses will assist medicinal chemists in designing more potent PPARγ modulators. To further enhance accessibility for the scientific community, we developed an online tool, "PGMP_v1," aimed at prospective screening for PPARγ modulators. The tool "PGMP_v1" is available at the provided link https://github.com/Amincheminfom/PGMP_v1 . The integration of these computational methods has uncovered crucial structural motifs that are essential for PPARγ activity, advancing the development of more effective modulators in the future.

In silico study to identify novel NEK7 inhibitors from natural sources by a combination strategy

Cancer poses a significant global health challenge and significantly contributes to mortality. NEK7, related to the NIMA protein kinase family, plays a crucial role in spindle assembly and cell division. The dysregulation of NEK7 is closely linked to the onset and progression of various cancers, especially colon and breast cancer, making it a promising target for cancer therapy. Nevertheless, the shortage of high-quality NEK7 inhibitors highlights the need for new therapeutic strategies. In this study, we utilized a multidisciplinary approach, including virtual screening, molecular docking, pharmacokinetics, molecular dynamics simulations (MDs), and MM/PBSA calculations, to evaluate natural compounds as NEK7 inhibitors comprehensively. Through various docking strategies, we identified three natural compounds: (-)-balanol, digallic acid, and scutellarin. Molecular docking revealed significant interactions at residues such as GLU112 and ALA114, with docking scores of -15.054, -13.059, and -11.547 kcal/mol, respectively, highlighting their potential as NEK7 inhibitors. MDs confirmed the stability of these compounds at the NEK7-binding site. Hydrogen bond analysis during simulations revealed consistent interactions, supporting their strong binding capacity. MM/PBSA analysis identified other crucial amino acids contributing to binding affinity, including ILE20, VAL28, ILE75, LEU93, ALA94, LYS143, PHE148, LEU160, and THR161, crucial for stabilizing the complex. This research demonstrated that these compounds exceeded dabrafenib in binding energy, according to MM/PBSA calculations, underscoring their effectiveness as NEK7 inhibitors. ADME/T predictions showed lower oral toxicity for these compounds, suggesting their potential for further development. This study highlights the promise of these natural compounds as bases for creating more potent derivatives with significant biological activities, paving the way for future experimental validation.

Machine learning-based identification and validation of amino acid metabolism related genes as novel biomarkers in chronic kidney disease

Objectives

Chronic kidney disease (CKD) is a progressive illness with a high rate of morbidity and mortality with no proven therapy. Alterations of amino acid(AA) metabolism are associated with the incidence and progression of CKD. To characterize the potential value of AA metabolism related genes in the diagnosis and progression of CKD.

Methods

We filtered the key genes associated with AA metabolism based on the least absolute shrinkage and selection operator (LASSO) and SVM algorithm. Then, we constructed logistic regression models and evaluated the accuracy and specificity by nomogram analysis and DCA. Also, we mapped the ROC curves.Meanwhile, in order to determine the underlying mechanism and relevant biological features of CKD, we conducted differential analysis between high and low risk subgroups in CKD. Moreover,we employed ssGSEA algorithm to evaluate the infiltration abundance of immune cells and calculated the correlation among the immune cells with the key genes. Finally,we validated the expression and clinical relevance of amino acid metabolism key genes via cultured cells and clinical data. A total of six key genes related to amino acid metabolism were identified, including ALDH18A1, CENPF, CSAD, CTH, CYP27B1, HBB.

Results

All six genes exhibited promising diagnostic capabilities (AUC:0.7 to 0.9). Immune cells such as Activated CD4+ T cells, Regulatory T cells, Immature B cells and MDSC,etc.infiltrated differentially in the high and low risk groups of CKD. There were correlations between immune cells abundance and the expression of key genes. All key genes correlated significantly with markers of kidney injury, such as eGFR and serum creatinine. The expression of ALDH18A1, CENPF were increased while CSAD, CTH and CYP27B1 were decreased in HK-2 cells cultured with indole sulfate.

Conclusions

Our study identified key genes involved in amino acid metabolism associated with immune cells infiltration and renal function in CKD, which may be potential biomarkers for the diagnosis and prognosis of CKD.

Gastric Cancer Models Developed via GelMA 3D Bioprinting Accurately Mimic Cancer Hallmarks, Tumor Microenvironment Features, and Drug Responses

Current in vitro models for gastric cancer research, such as 2D cell cultures and organoid systems, often fail to replicate the complex extracellular matrix (ECM) found in vivo. For the first time, this study utilizes a gelatin methacryloyl (GelMA) hydrogel, a biomimetic ECM-like material, in 3D bioprinting to construct a physiologically relevant gastric cancer model. GelMA's tunable mechanical properties allow for the precise manipulation of cellular behavior within physiological ranges. Genetic and phenotypic analyses indicate that the 3D bioprinted GelMA (3Db) model accurately mimics the clinical tumor characteristics and reproduces key cancer hallmarks, such as cell proliferation, invasion, migration, angiogenesis, and the Warburg effect. Comparisons of gene expression and drug responses between the 3Db model and patient-derived xenograft models, both constructed from primary gastric cancer cells, validate the model's clinical relevance. The ability of the 3Db model to closely simulate in vivo conditions highlights its crucial role in identifying treatment targets and predicting patient-specific responses, showcasing its potential in high-throughput drug screening and clinical applications. This study is the first to report the pivotal role of GelMA-based 3D bioprinting in advancing gastric cancer research and regenerative medicine.

Identification of mitophagy-related key genes and their correlation with immune cell infiltration in acute myocardial infarction via bioinformatics analysis

Background

Acute myocardial infarction (AMI), a subset of acute coronary syndrome, remains the major cause of mortality worldwide. Mitochondrial dysfunction is critically involved in AMI progression, and mitophagy plays a vital role in eliminating damaged mitochondria. This study aimed to explore mitophagy-related biomarkers and their potential molecular basis in AMI.

Methods

AMI datasets (GSE24519 and GSE34198) from the Gene Expression Omnibus database were combined and the batch effects were removed. Differentially expressed genes (DEGs) in AMI were selected, intersected with mitophagy-related genes for mitophagy-related DEGs (MRDEGs), and then subjected to enrichment analyses. Next, the MRDEGs were screened using machine learning methods (logistic regression analysis, RandomForest, least absolute shrinkage and selection operator) to construct a diagnostic risk model and select the key genes in AMI. The diagnostic efficacy of the model was evaluated using a nomogram. Moreover, the infiltration patterns of different immune cells in two risk groups were compared. We also explored the interactions between the key genes themselves or with miRNAs/transcription factors (TFs) and drug compounds and visualized the protein structure of the key genes. Finally, we explored and validated the expression of key genes in plasma samples of patients with an AMI and healthy individuals.

Results

We screened 28 MRDEGs in AMI. Based on machine learning methods, 12 key genes were screened for the diagnostic risk model, including AGPS, CA2, CAT, LTA4H, MYO9B, PRDX6, PYGB, SIRT3, TFEB, TOM1, UBA52, and UBB. The nomogram further revealed the accuracy of the model for AMI diagnosis. Moreover, we found a lower abundance of immune cells such as gamma delta T and natural killer cells in the high-risk group, and the expression of key genes showed a significant correlation with immune infiltration levels in both groups. Finally, 64 miRNA-mRNA pairs, 75 TF-mRNA pairs, 119 RNA-binding protein-mRNA pairs, and 32 drug-mRNA pairs were obtained in the interaction networks.

Conclusions

In total, 12 key MRDEGs were identified and a risk model was constructed for AMI diagnosis. The findings of this study might provide novel biomarkers for improving the detection of AMI.

Machine learning can predict surgical indication: new clustering model from a large adult spine deformity database

PURPOSE

The choice of the best management for Adult Spine Deformity (ASD) is challenging. Health-related quality of life (HRQoL), comorbidities, symptoms and spine geometry, along with surgical risk and potential residual disability play a role, and a definite algorithm for patient management is lacking. Machine learning allows to analyse complex settings more efficiently than other available statistical tools. Aim of this study was to develop a machine-learning algorithm that, based on baseline data, would be able to predict whether an ASD patient would undergo surgery or not.

METHODS

Retrospective evaluation of prospectively collected data. Demographic data, HRQoL and radiographic parameters were collected. Two clustering methods were performed to differentiate groups of patients with similar characteristics. Three models were then used to identify the most relevant variables for management prediction.

RESULTS

Data from 1319 patients were available. Three clusters were identified: older subjects with sagittal imbalance and high PI, younger patients with greater coronal deformity and no sagittal imbalance, older patients with moderate sagittal imbalance and lower PI. The group of younger patients showed the highest error rate for the prediction (37%), which was lower for the other two groups (20-27%). For all groups, quality of life parameters such as the ODI and the SRS 22 and the Cobb angle of the major curve were the strongest predictors of surgical indication, albeit with different odds ratios in each group.

CONCLUSION

Three clusters could be identified along with the variables that, in each, are most likely to drive the choice of management.

Targeted Analysis, Metabolic Profiling, and Fingerprinting Based on an LC(GC)-MS Approach for the Comprehensive Evaluation of Pesticide Content in Edible Plants

Pesticides are commonly found in plant-based foods, which inevitably reduces food quality and poses significant health risks to consumers. The extensive variety of crops and the wide range of pesticides used means that no single analytical approach can provide clear and comprehensive information on the pesticide-protection status of a crop. Since most pesticide analyses in food rely on chromatographic techniques combined with various MS platforms, this article focuses exclusively on LC-MS and GC-MS system methodologies. In summary, this paper critically reviews analytical modes-specifically, multi reaction monitoring, data-dependent analysis, and data-independent analysis-and scanning regimes, including full scan, MS, MS/MS, suspect screening, and fingerprinting strategies, for pesticide detection in edible plants. The advantages and disadvantages of these methodologies, as well as their complementary applications, are thoroughly examined.

Characterizing phenotype variants of Cercosporidium personatum, causal agent of peanut late leaf spot disease, their morphology, genetics and metabolites

Cercosporidium personatum (CP) causes peanut late leaf spot (LLS) disease with 70% yield losses unless controlled by fungicides. CP grows slowly in culture, exhibiting variable phenotypes. To explain those variations, we analyzed the morphology, genomes, transcriptomes and chemical composition of three morphotypes, herein called RED, TAN, and BROWN. We characterized, for the first time in CP, anthraquinone (AQ) precursors of dothistromin (DOT), including averantin, averufin, norsolorinic acid, versicolorin B, versicolorin A, nidurufin and averufanin. BROWN had the highest AQ and melanin (15 mg/g DW) contents. RED had the highest ergosterol (855 µM FW) and chitin (beta-glucans, 4% DW) contents. RED and TAN had higher resistance to xenobiotics (p ≤ 1.0E-3), including chlorothalonil, tebuconazole and caffeine, compared to CP NRRL 64,463. In RED, TAN, and BROWN, rates of single nucleotide polymorphisms (SNP) (1.4-1.7 nt/kb) and amino acid changes (3k-4k) were higher than in NRRL 64,463. Differential gene expression (p ≤ 1.0E-5) was observed in 47 pathogenicity/virulence genes, 41 carbohydrate-active enzymes (CAZymes), and 23 pigment/mycotoxin biosynthesis genes. We describe the MAT1 locus, and a method to evaluate CP-xenobiotic resistance in 5 days. Chemical profiles indicate each CP morphotype could trigger different immune response in plants, probably hindering development of durable LLS resistance.

An exploration of distinguishing subjective cognitive decline and mild cognitive impairment based on resting-state prefrontal functional connectivity assessed by functional near-infrared spectroscopy

Purpose

Functional near-infrared spectroscopy (fNIRS) has shown feasibility in evaluating cognitive function and brain functional connectivity (FC). Therefore, this fNIRS study aimed to develop a screening method for subjective cognitive decline (SCD) and mild cognitive impairment (MCI) based on resting-state prefrontal FC and neuropsychological tests via machine learning.

Methods

Functional connectivity data measured by fNIRS were collected from 55 normal controls (NCs), 80 SCD individuals, and 111 MCI individuals. Differences in FC were analyzed among the groups. FC strength and neuropsychological test scores were extracted as features to build classification and predictive models through machine learning. Model performance was assessed based on accuracy, specificity, sensitivity, and area under the curve (AUC) with 95% confidence interval (CI) values.

Results

Statistical analysis revealed a trend toward compensatory enhanced prefrontal FC in SCD and MCI individuals. The models showed a satisfactory ability to differentiate among the three groups, especially those employing linear discriminant analysis, logistic regression, and support vector machine. Accuracies of 94.9% for MCI vs. NC, 79.4% for MCI vs. SCD, and 77.0% for SCD vs. NC were achieved, and the highest AUC values were 97.5% (95% CI: 95.0%-100.0%) for MCI vs. NC, 83.7% (95% CI: 77.5%-89.8%) for MCI vs. SCD, and 80.6% (95% CI: 72.7%-88.4%) for SCD vs. NC.

Conclusion

The developed screening method based on resting-state prefrontal FC measured by fNIRS and machine learning may help predict early-stage cognitive impairment.

Multi-omic biomarker panel in pancreatic cyst fluid and serum predicts patients at a high risk of pancreatic cancer development

Integration of multi-omic data for the purposes of biomarker discovery can provide novel and robust panels across multiple biological compartments. Appropriate analytical methods are key to ensuring accurate and meaningful outputs in the multi-omic setting. Here, we extensively profile the proteome and transcriptome of patient pancreatic cyst fluid (PCF) (n = 32) and serum (n = 68), before integrating matched omic and biofluid data, to identify biomarkers of pancreatic cancer risk. Differential expression analysis, feature reduction, multi-omic data integration, unsupervised hierarchical clustering, principal component analysis, spearman correlations and leave-one-out cross-validation were performed using RStudio and CombiROC software. An 11-feature multi-omic panel in PCF [PIGR, S100A8, REG1A, LGALS3, TCN1, LCN2, PRSS8, MUC6, SNORA66, miR-216a-5p, miR-216b-5p] generated an AUC = 0.806. A 13-feature multi-omic panel in serum [SHROOM3, IGHV3-72, IGJ, IGHA1, PPBP, APOD, SFN, IGHG1, miR-197-5p, miR-6741-5p, miR-3180, miR-3180-3p, miR-6782-5p] produced an AUC = 0.824. Integration of the strongest performing biomarkers generated a 10-feature cross-biofluid multi-omic panel [S100A8, LGALS3, SNORA66, miR-216b-5p, IGHV3-72, IGJ, IGHA1, PPBP, miR-3180, miR-3180-3p] with an AUC = 0.970. Multi-omic profiling provides an abundance of potential biomarkers. Integration of data from different omic compartments, and across biofluids, produced a biomarker panel that performs with high accuracy, showing promise for the risk stratification of patients with pancreatic cystic lesions.

Assessing the toxicological impact of PET-MPs exposure on IVDD: Insights from network toxicology and molecular docking

Polyethylene terephthalate microplastics (PET-MPs) have emerged as a significant environmental concern due to their persistence and potential health hazards. Their role in degenerative diseases, particularly intervertebral disc degeneration (IVDD), remains poorly understood, highlighting the need for systematic evaluation of their molecular toxicity. In this study, network toxicology and molecular docking approaches were applied to investigate the toxicological mechanisms of PET-MPs-induced IVDD. Comprehensive analyses of GEO, ChEMBL, STITCH, GeneCards, and OMIM databases identified 46 potential targets associated with PET-MPs exposure, which were further refined to seven core targets, including AKT1, CASP3, and SRC, using STRING and Cytoscape software. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses revealed that PET-MPs influence immune-related pathways, such as Ras signaling, apoptosis, VEGF receptor signaling, and neutrophil extracellular trap (NET) formation. Molecular docking analysis confirmed strong binding affinities of PET-MPs to these core targets, suggesting its potential to disrupt key cellular processes. These findings indicate that PET-MPs may accelerate IVDD progression by modulating apoptosis, extracellular matrix (ECM) metabolism, angiogenesis, and immune responses. This study provides valuable insights into the molecular mechanisms underlying PET-MPs-induced IVDD and highlights the utility of network toxicology in evaluating the toxicity of emerging environmental pollutants, offering a theoretical foundation for understanding the health risks of PET-MPs and guiding strategies to mitigate their impact on degenerative diseases.

Global assessment of leukemia care quality: insights from the quality of care index (QCI) from 1990 to 2021

Background

While advancements in leukemia care have been made, the global quality of care remains a concern. This study utilizes a modified quality of care index (QCI) to assess the global status of leukemia care.

Methods

We analyzed data from the global burden of disease (GBD) study spanning 1990-2021. The QCI was constructed using principal component analysis, based on the weighted variances of key indicators. We compared the original QCI with our modified version, analyzed QCI trends across different age groups and leukemia subtypes, identified key influencing factors using linear mixed models (LMM), and used spatial autocorrelation analysis to verify the autocorrelation of the socio-demographic index (SDI) region. Then we employed the bayesian age-period-cohort (BAPC) model to predict future QCI trends.

Findings

Between 1990 and 2021, both the age-standardized incidence rate (ASIR) and age-standardized death rate (ASDR) for leukemia exhibited a consistent decline. Our modified QCI method outperformed the original approach, particularly when the variance explained by the first principal component was below 80%, demonstrating higher correlation with the healthcare access and quality index (HAQI) (Pearson r = 0.91 vs. 0.89) and improved explanatory power (R2 = 0.82 vs. 0.79). Over past three decades, QCI was highest in San Marino (97.72%) and lowest in Fiji (3.51%), with significant regional variations across SDI levels (F = 133.40, p < 2e-16). High-SDI regions had the highest QCI (78.50%; 95% confidence interval: 77.20%, 79.70%). QCI trends varied by age, peaking at 94.49% in the 15-19 age group in 2021 and declining to 0.44% in the 75-79 age group. LMM analysis identified sex, age, year, SDI region, and leukemia subtype as significant QCI determinants. Spatial autocorrelation analysis confirmed positive autocorrelation within SDI regions (Global Moran's I = 0.87, p < 2e-16). Projections suggest a generally fluctuating upward trend in QCI for leukemia, reaching 79.58% by 2046.

Interpretation

The QCI serves as an effective metric for evaluating the quality of leukemia care. Our findings reveal a strong association between leukemia QCI and regional economic and educational development. Age is a critical factor, with an aging population contributing to a potential decline in QCI. These results underscore the urgent need for targeted interventions to enhance health services for older adults and to improve care quality in economically disadvantaged regions.

Funding

This study was supported by the National Natural Science Foundation of China (General Program) (No. 82370176) and the Key Research and Development Program of Hubei Province (No. CZKYXM2023036JZ).

Identification and validation of cuproptosis-related genes for diagnosis and therapy in nonalcoholic fatty liver disease

In recent years, nonalcoholic fatty liver disease (NAFLD) has become a more serious public health issue worldwide. This study strived to investigate the molecular mechanism of pathogenesis of NAFLD and explore promising diagnostic and therapeutic targets for NAFLD. Raw data from GSE130970 were downloaded from the Gene Expression Omnibus database. We used the dataset to analyze the expression levels of cuproptosis-related genes in NAFLD patients and healthy controls to identify the differentially expressed cuproptosis-related genes (DECRGs). The relationship and potential mechanism between DECRGs and clinicopathological factors were examined by enrichment analysis and two consensus clustering methods. We screened key DECRGs based on Random Forest (RF), and then verified the key DECRGs in NAFLD patients, high-fat diet (HFD)-fed mice, and palmitic acid-induced AML12 cells. ROC analysis showed good diagnostic function of DECRGs in normal and NAFLD liver tissue. Two consensus clusters indicated the important role of cuproptosis in the development of NAFLD. We screened for key DECRGs (DLD, DLAT) based on RF and found a close relationship between the DECRGs and clinicopathological factors. We collected clinical blood samples to verify the differences in gene expression levels by qPCR. In addition, we further verified the expression levels of DLD and DLAT in HFD mice and AML12 cells, which showed the same results. This study provides a novel perspective on the pathogenesis of NAFLD. We identified two cuproptosis-related genes that are closely related to NAFLD. These genes may play a significant role in the molecular pathogenesis of NAFLD, which may be useful to make progress in the diagnosis and treatment of NAFLD.

Machine learning methods to discover hidden patterns in well-being and resilience for healthy aging

BACKGROUND

A whole person approach to healthy aging can provide insight into social factors that may be critical. Digital technologies, such as mobile health (mHealth) applications, hold promise to provide novel insights for healthy aging and the ability to collect data between clinical care visits. Machine learning/artificial intelligence methods have the potential to uncover insights into healthy aging. Nurses and nurse informaticians have a unique lens to shape the future use of this technology.

METHODS

The purpose of this research was to apply machine learning methods to MyStrengths+MyHealth de-identified data (N = 988) for adults 45 years of age and older. An exploratory data analysis process guided this work.

RESULTS

Overall (n = 988), the average Strength was 66.1% (SD = 5.1), average Challenges 66.5% (SD = 7.5), and average Needs 60.06% (SD = 3.1). There was a significant difference between Strengths and Needs (p < 0.001), between Challenges and Needs (p < 0.001), and no significant differences between average Strengths and Challenges. Four concept groups were identified from the data (Thinking, Moving, Emotions, and Sleeping). The Thinking group had the most statistically significant challenges (11) associated with having at least one Thinking Challenge and the highest average Strengths (66.5%) and Needs (83.6%) compared to the other groups.

CONCLUSION

This retrospective analysis applied machine learning methods to de-identified whole person health resilience data from the MSMH application. Adults 45 and older had many Strengths despite numerous Challenges and Needs. The Thinking group had the highest Strengths, Challenges, and Needs, which aligns with the literature and highlights the co-occurring health challenges experienced by this group. Machine learning methods applied to consumer health data identify unique insights applicable to specific conditions (e.g., cognitive) and healthy aging. The next steps involve testing personalized interventions with nurses leading artificial intelligence integration into clinical care.

Protein Binding Site Representation in Latent Space

Interpretability and reliability of deep learning models are important for computer-based drug discovery. Aiming to understand feature perception by such a model, we investigate a graph neural network for affinity prediction of protein-ligand complexes. We assess a latent representation of ligand binding sites and investigate underlying geometric structure in this latent space and its relation to protein function. We introduce an automated computational pipeline for dimensionality reduction, clustering, hypothesis testing, and visualization of latent space. The results indicate that the learned protein latent space is inherently structured and not randomly distributed. Several of the identified protein binding site clusters in latent space correspond to functional protein families. Ligand size was found to be a determinant of cluster geometry. The computational pipeline proved applicable to latent space analysis and interpretation and can be adapted to work for different datasets and deep learning models.

Loss of Hoxa5 function affects Hox gene expression in different biological contexts

Hoxa5 plays numerous roles in development, but its downstream molecular effects are mostly unknown. We applied bulk RNA-seq assays to characterize the transcriptional impact of the loss of Hoxa5 gene function in seven different biological contexts, including developing respiratory and musculoskeletal tissues that present phenotypes in Hoxa5 mouse mutants. This global analysis revealed few common transcriptional changes, suggesting that HOXA5 acts mainly via the regulation of context-specific effectors. However, Hox genes themselves appeared as potentially conserved targets of HOXA5 across tissues. Notably, a trend toward reduced expression of HoxA genes was observed in Hoxa5 null mutants in several tissue contexts. Comparative analysis of epigenetic marks along the HoxA cluster in lung tissue from two different Hoxa5 mutant mouse lines revealed limited effect of either mutation indicating that Hoxa5 gene targeting did not significantly perturb the chromatin landscape of the surrounding HoxA cluster. Combined with the shared impact of the two Hoxa5 mutant alleles on phenotype and Hox expression, these data argue against the contribution of local cis effects to Hoxa5 mutant phenotypes and support the notion that the HOXA5 protein acts in trans in the control of Hox gene expression.

Data-driven discovery of the interplay between genetic and environmental factors in bacterial growth

A complex interplay of genetic and environmental factors influences bacterial growth. Understanding these interactions is crucial for insights into complex living systems. This study employs a data-driven approach to uncover the principles governing bacterial growth changes due to genetic and environmental variation. A pilot survey is conducted across 115 Escherichia coli strains and 135 synthetic media comprising 45 chemicals, generating 13,944 growth profiles. Machine learning analyzes this dataset to predict the chemicals' priorities for bacterial growth. The primary gene-chemical networks are structured hierarchically, with glucose playing a pivotal role. Offset in bacterial growth changes is frequently observed across 1,445,840 combinations of strains and media, with its magnitude correlating to individual alterations in strains or media. This counterbalance in the gene-chemical interplay is supposed to be a general feature beneficial for bacterial population growth.

Detecting cardiovascular diseases using unsupervised machine learning clustering based on electronic medical records

BACKGROUND

Electronic medical records (EMR)-trained machine learning models have the potential in CVD risk prediction by integrating a range of medical data from patients, facilitate timely diagnosis and classification of CVDs. We tested the hypothesis that unsupervised ML approach utilizing EMR could be used to develop a new model for detecting prevalent CVD in clinical settings.

METHODS

We included 155,894 patients (aged ≥ 18 years) discharged between January 2014 and July 2022, from Xuhui Hospital, Shanghai, China, including 64,916 CVD cases and 90,979 non-CVD cases. K-means clustering was used to generate the clustering models with k = 2, 4, and 8 as predetermined number of clusters k = 2, 4, and 8. Bayesian theorem was used to estimate the models' predictive accuracy.

RESULTS

The overall predictive accuracy of the 2-, 4-, and 8-classification clustering models in the training set was 0.856, 0.8634, and 0.8506, respectively. Similarly, the predictive accuracy of the 2-, 4-, and 8-classification clustering models in the testing set was 0.8598, 0.8659, and 0.8525, respectively. After reducing from 19 dimensions to 2 dimensions by principal component analysis, significant separation was observed for CVD cases and non-CVD cases in both training and testing sets.

CONCLUSION

Our findings indicate that the utilization of EMR data can support the development of a robust model for CVD detection through an unsupervised ML approach. Further investigation using longitudinal design is needed to refine the model for its applications in clinical settings.

Four functional genotoxic marker genes (Bax, Btg2, Ccng1, and Cdkn1a) discriminate genotoxic hepatocarcinogens from non-genotoxic hepatocarcinogens and non-genotoxic non-hepatocarcinogens in rat public toxicogenomics data, Open TG-GATEs

BACKGROUND

Previously, Japanese Environmental Mutagen and Genome Society/Mammalian Mutagenicity Study Group/Toxicogenomics Study Group (JEMS/MMS toxicogenomic study group) proposed 12 genotoxic marker genes (Aen, Bax, Btg2, Ccnf, Ccng1, Cdkn1a, Gdf15, Lrp1, Mbd1, Phlda3, Plk2, and Tubb4b) to discriminate genotoxic hepatocarcinogens (GTHCs) from non-genotoxic hepatocarcinogens (NGTHCs) and non-genotoxic non-hepatocarcinogens (NGTNHCs) in mouse and rat liver using qPCR and RNA-Seq and confirmed in public rat toxicogenomics data, Open TG-GATEs, by principal component analysis (PCA). On the other hand, the U.S. Environmental Protection Agency (US EPA) suggested seven genotoxic marker genes (Bax, Btg2, Ccng1, Cgrrf1, Cdkn1a, Mgmt, and Tmem47) with Open TG-GATEs data. Four genes (Bax, Btg2, Ccng1, and Cdkn1a) were common in these two studies. In the present study, we examined the performance of these four genes in Open TG-GATEs data using PCA.

RESULTS

The study's findings are of paramount significance, as these four genes proved to be highly effective in distinguishing five typical GTHCs (2-acetylaminofluorene, aflatoxin B1, 2-nitrofluorene, N-nitrosodiethylamine and N-nitrosomorpholine) from seven typical NGTHCs (clofibrate, ethanol, fenofibrate, gemfibrozil, hexachlorobenzene, phenobarbital, and WY-14643) and 11 NGTNHCs (allyl alcohol, aspirin, caffeine, chlorpheniramine, chlorpropamide, dexamethasone, diazepam, indomethacin, phenylbutazone, theophylline, and tolbutamide) by PCA at 24 h after a single administration with 100% accuracy. These four genes also effectively distinguished two typical GTHCs (2-acetylaminofluorene and N-nitrosodiethylamine) from seven NGTHCs and ten NGTNHCs by PCA on 29 days after 28 days-repeated administrations, with a similar or even better performance compared to the previous 12 genes. Furthermore, the study's analysis revealed that the three intermediate GTHC/NGTHCs (methapyrilene, monocrotaline, and thioacetamide, which were negative in the Salmonella test but positive in the in vivo rat liver test) were located in the intermediate region between typical GTHCs and typical NGTHCs by PCA.

CONCLUSIONS

The present results unequivocally demonstrate the availability of four genotoxic marker genes ((Bax, Btg2, Ccng1, and Cdkn1a) and PCA in discriminating GTHCs from NGTHCs and NGTNHCs in Open TG-GATEs. These findings strongly support our recommendation that future rat liver in vivo toxicogenomics tests prioritize these four genotoxic marker genes, as they have proven to be highly effective in discriminating between different types of hepatocarcinogens.

Application of machine learning for mass spectrometry-based multi-omics in thyroid diseases

Thyroid diseases, including functional and neoplastic diseases, bring a huge burden to people's health. Therefore, a timely and accurate diagnosis is necessary. Mass spectrometry (MS) based multi-omics has become an effective strategy to reveal the complex biological mechanisms of thyroid diseases. The exponential growth of biomedical data has promoted the applications of machine learning (ML) techniques to address new challenges in biology and clinical research. In this review, we presented the detailed review of applications of ML for MS-based multi-omics in thyroid disease. It is primarily divided into two sections. In the first section, MS-based multi-omics, primarily proteomics and metabolomics, and their applications in clinical diseases are briefly discussed. In the second section, several commonly used unsupervised learning and supervised algorithms, such as principal component analysis, hierarchical clustering, random forest, and support vector machines are addressed, and the integration of ML techniques with MS-based multi-omics data and its application in thyroid disease diagnosis is explored.

Fuzzy-Based Identification of Transition Cells to Infer Cell Trajectory for Single-Cell Transcriptomics

With the continuous evolution of single-cell RNA sequencing technology, it has become feasible to reconstruct cell development processes using computational methods. Trajectory inference is a crucial downstream analytical task that provides valuable insights into understanding cell cycle and differentiation. During cell development, cells exhibit both stable and transition states, which makes it challenging to accurately identify these cells. To address this challenge, we propose a novel single-cell trajectory inference method using fuzzy clustering, named scFCTI. By introducing fuzzy clustering and quantifying cell uncertainty, scFCTI can identify transition cells within unstable cell states. Moreover, scFCTI can obtain refined cell classification by characterizing different cell stages, which gain more accurate single-cell trajectory reconstruction containing transition paths. To validate the effectiveness of scFCTI, we conduct experiments on five real datasets and four different structure simulation datasets, comparing them with several state-of-the-art trajectory inference methods. The results demonstrate that scFCTI outperforms these methods by successfully identifying unstable cell clusters and obtaining more accurate cell paths with transition states. Especially the experimental results demonstrate that scFCTI can reconstruct the cell trajectory more precisely.

Small molecule inhibits BACE1 activity by a dual mechanism confirmed by simulations-based study

Alzheimer's disease (AD) is a progressive and largely incurable neurodegenerative disorder that affects millions of people worldwide. It is characterised by the accumulation of amyloid-beta plaques and neurofibrillary tangles in the brain. It is commenced by cleavage of amyloid precursor protein (APP) by β-secretase, β-site amyloid precursor protein cleaving enzyme (BACE1; also called Asp2, memapsin 2). Therefore, BACE1 is a prime target for developing therapeutics against AD. In this study, we have identified a small molecule that potentially inhibits the activity of BACE1 by interacting with the active site residues. Also, the flap region seems to be involved in enhancing the stability of the small molecule at the active site. We have used Umibecestat (CNP-520) as a positive control. Our in silico results show that the identified molecule has a much better orientation at the active site of BACE1 than Umibecestat and inhibits by blocking the active site and modulating flap dynamics. We have utilised virtual high-throughput screening assay, ADME profiling, and blood-brain-barrier crossing ability to narrow down potential leads. The two shortlisted molecules were then subjected to atomistic molecular dynamics simulations study. Overall, our study proposes a much better inhibitor and a rational molecule for lead development against AD.

Complex Thrombo-Inflammatory Responses versus Outcomes of Non-COVID-19 Community-Acquired Pneumonia and COVID-19

INTRODUCTION

The thrombo-inflammatory response and outcomes of community-acquired pneumonia (CAP) due to various organisms (non-COVID-19 CAP) versus CAP due to a single virus, SARS-CoV-2 (i.e., COVID-19) may differ.

METHODS

Adults hospitalized with non-COVID-19 CAP (December 1, 2021-June 15, 2023) or COVID-19 (March 2, 2020-June 15, 2023) in Canada. We compared non-COVID-19 CAP and COVID-19 baseline, thrombo-inflammatory response, and mortality. We measured plasma cytokine and coagulation factor levels in a sample of patients, did hierarchical clustering, and compared cytokine and coagulation factor levels.

RESULTS

In 2,485 patients (non-COVID-19 CAP, n = 719; COVID-19 patients, n = 2,157), non-COVID-19 CAP patients had significantly lower 28-day mortality (CAP vs. COVID-19 waves 1 and 2; 10% vs. 18% and 16%, respectively), intensive care unit admission (CAP vs. all waves; 15% vs. 39%, 37%, 33%, and 24%, respectively), invasive ventilation (CAP vs. waves 1, 2, and 3 patients; 11% vs. 25%, 20%, and 16%), vasopressor use (CAP 12% vs. 23%, 21%, and 18%), and renal replacement therapy use (CAP 3% vs. Omicron 7%). Complexity of hierarchical clustering aligned directly with mortality: COVID-19 wave 1 and 2 patients had six clusters at admission and higher mortality than non-COVID-19 CAP and Omicron that had three clusters at admission. Pooling all COVID-19 waves increased complexity with seven clusters on admission.

CONCLUSION

Complex thrombo-inflammatory responses aligned with mortality of CAP. At a fundamental level, the human thrombo-inflammatory response to a brand new virus was "confused" whereas humans had eons of time to develop a more concise efficient thrombo-inflammatory host response to CAP.

Association of a composite trait for anthropometrics, adiposity and energy expenditure with cardiometabolic diseases: An age-stratified cohort and genetic risk score analysis

AIM

Various anthropometric measures capture distinct as well as overlapping characteristics of an individual's body composition. To characterize independent body composition measures, we aimed to reduce easily-obtainable individual measures reflecting adiposity, anthropometrics and energy expenditure into fewer independent constructs, and to assess their potential sex- and age-specific relation with cardiometabolic diseases.

METHODS

Analyses were performed within European ancestry participants from UK Biobank (N = 418,963, mean age 58.0 years, 56% women). Principal components (PC) analyses were used for the dimension reduction of 11 measures of adiposity, anthropometrics and energy expenditure. PCs were studied in relation to incident type 2 diabetes mellitus (T2D) and coronary artery disease (CAD). Multivariable-adjusted Cox regression analyses, adjusted for confounding factors, were performed in all and stratified by age. Genome-wide association studies were performed in half of the cohort (N = 156,295) to identify genetic variants as instrumental variables. Genetic risk score analyses were performed in the other half of the cohort stratified by age of disease onset (N = 156,295).

RESULTS

We identified two PCs, of which PC1 reflected lower overall adiposity (negatively correlated with all adiposity aspects) and PC2 reflected more central adiposity (mainly correlated with higher waist-hip ratio, but with lower total body fat) and increased height, collectively capturing 87.8% of the total variance. Similar to that observed in the multivariable-adjusted regression analyses, we found associations between the PC1 genetic risk score and lower risks of CAD and T2D [CAD cases <50 years, odds ratio: 0.91 (95% confidence interval 0.87, 0.94) per SD; T2D cases <50 years, odds ratio: 0.76 (0.72, 0.81)], which attenuated with higher age (p-values 8.13E-4 and 2.41E-6, respectively). No associations were found for PC2.

CONCLUSIONS

The consistently observed weaker associations of the composite traits with cardiometabolic disease suggests the need for age-specific cardiometabolic disease prevention strategies.

A review on description dynamics and conformational changes of proteins using combination of principal component analysis and molecular dynamics simulation

Understanding how proteins behave dynamically and undergo conformational changes is essential to comprehending their biological roles. This review article examines the potent tool of using Molecular Dynamics simulations in conjunction with Principal Component Analysis (PCA) to explore protein dynamics. Molecular dynamics data can be made easier to read by removing prominent patterns through the use of PCA, a sophisticated dimensionality reduction approach. Researchers can obtain critical insights into the fundamental principles governing protein function by using PCA on MD simulation data. We provide a systematic approach to PCA that includes data collection, input coordinate selection, and result interpretation. Protein collective movements and fundamental dynamics are made visible by PCA, which makes it possible to identify conformational substates that are crucial to function. By means of principal component analysis, scientists are able to observe and measure large-scale movements, like hinge bending and domain motions, as well as pinpoint areas of protein structural stiffness and flexibility. Moreover, PCA allows temporal separation, distinguishing slower global motions from faster local changes. A strong foundation for researching protein dynamics is provided by the combination of PCA and Molecular Dynamics simulations, which have applications in drug development and enhance our comprehension of intricate biological systems.

Comparative transcriptomic analysis revealed potential mechanisms regulating the hypertrophy of goose pectoral muscles

Pectoral muscle development is an important economic trait. According to the different essence, muscle development can be divided into 2 processes: embryonic muscle fiber generation and postnatal muscle fiber hypertrophy, and postnatal muscle fiber hypertrophy has a greater impact on muscle development than the number of muscle fibers formed during the embryonic phase in poultry. However, the underlying mechanisms regulating the hypertrophy of goose pectoral muscles have not been elucidated. Therefore, the purpose of the present study was to conduct transcriptome sequencing in pectoral muscles of both Landes (LD) and Sichuan White (SW) geese at 6, 10, and 30 weeks of age to reveal the molecular mechanisms regulating pectoral muscle hypertrophy through intra-breed and inter-breed bioinformatics analyses. Phenotypically, the pectoral muscle weight/index of LD and SW geese increased from 6 to 30 weeks of age, and except for the pectoral muscle index at 10 weeks of age (P = 0.962), at the same age, the pectoral muscle weight/index of LD geese were significantly higher than that of SW geese (P < 0.05). In transcriptional regulation, intra-breed bioinformatics analysis identified 3331 genes whose expression levels were opposite to the trend of pectoral muscle hypertrophy both in LD and SW geese, and the 3331 genes were mainly enriched into abundant KEGG pathways related to lipid metabolism, proliferation/apoptosis, and immune response. Moreover, 23 genes (including SLC2A10, TNFRSF1A, PRKAA1, SLC27A4, ITGB2, THY1, RHOA, MYL10, ACTB, PRKCB, PIK3R2, RAC2, DMD, LATS2, YAP1, WWTR1, SMAD7, CTGF, FGF1, AXIN2, GLI2, ID2, and CCND2) who were enriched in 6 crosstalk pathways named viral myocarditis, insulin resistance, sphingolipid signaling pathway, hippo signaling pathway, chemokine signaling pathway, and leukocyte transendothelial migration were identified as the key candidate genes regulating the hypertrophy of goose pectoral muscles. In inter-breed bioinformatics analysis, abundant different expression genes (DEGs) related to lipid metabolism, immune response, and proliferation/apoptosis were identified between LD and SW geese too, and compared with SW geese, the expression level of MYL10 in LD geese was lower, while the expression levels of GLI2/CTGF/SMAD7 in LD geese were higher. These results suggested that the hypertrophy of goose pectoral muscles might be achieved through more lipid deposition and less leukocyte infiltration to promote the proliferation of cells within the muscles, and the low expression of MYL10 and high expressions of GLI2/CTGF/SMAD7 might the keys to induce the pectoral muscle hypertrophy of LD geese from 6 to 30 weeks of age over that of SW geese. All data the present study obtained will provide new insights into the molecular mechanisms regulating the hypertrophy of goose pectoral muscles.

Integrative analysis of homologous recombination repair patterns unveils prognostic signatures and immunotherapeutic insights in breast cancer

Globally, breast cancer (BC) is the leading cause of female death and morbidity. Homologous recombination repair (HRR) is critical in BC. However, the prognostic role and immunotherapy response of HRR in BC remains to be clarified. Firstly, we identified HRR types in BC samples from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) dataset (GSE42568) based on 65 HRR genes (HRRGs). A differentially expressed gene (DEG) list for different HRR types was generated. Then, the influences of gene sets composed of these DEGs on biological pathways and BC prognosis were explored. Next, we identified gene clusters based on gene sets composed of DEGs. Genes associated with prognosis for DEGs were identified using univariate Cox regression. Finally, the HRR score was constructed based on genes associated with prognosis. We analyzed how HRR score correlates with tumor mutation burden (TMB), immune cell infiltration (ICI), and immunotherapy response. Three HRR clusters were discovered. HRR subtype A demonstrated decreased infiltration and a high number of immunosuppressive cells with a poor prognosis. DEGs among various HRR types were predominantly enriched in cell cycle and genomic stability-related pathways. The prognostic model based on sixteen DEGs accurately predicted BC prognosis. The HRRGs were differentially expressed in three DEG clusters. TMB, ICI, and immunotherapy responses differed significantly between the high and low HRR groups (HSG, LSG). The HSG was distinguished by a high degree of ICI and low TMB. LSG had a better response to anti-PD-1 or anti-PD-1 and anti-CTLA4 combination therapy. This work revealed that HRR patterns would contribute to predicting prognosis and immunotherapy response in BC, which may benefit patients.

Prediction of Post Traumatic Epilepsy Using MR-Based Imaging Markers

Post-traumatic epilepsy (PTE) is a debilitating neurological disorder that develops after traumatic brain injury (TBI). Despite the high prevalence of PTE, current methods for predicting its occurrence remain limited. In this study, we aimed to identify imaging-based markers for the prediction of PTE using machine learning. Specifically, we examined three imaging features: Lesion volumes, resting-state fMRI-based measures of functional connectivity, and amplitude of low-frequency fluctuation (ALFF). We employed three machine-learning methods, namely, kernel support vector machine (KSVM), random forest, and an artificial neural network (NN), to develop predictive models. Our results showed that the KSVM classifier, with all three feature types as input, achieved the best prediction accuracy of 0.78 AUC (area under the receiver operating characteristic (ROC) curve) using nested cross-validation. Furthermore, we performed voxel-wise and lobe-wise group difference analyses to investigate the specific brain regions and features that the model found to be most helpful in distinguishing PTE from non-PTE populations. Our statistical analysis uncovered significant differences in bilateral temporal lobes and cerebellum between PTE and non-PTE groups. Overall, our findings demonstrate the complementary prognostic value of MR-based markers in PTE prediction and provide new insights into the underlying structural and functional alterations associated with PTE.

Interspecific Association and Environmental Interpretation of Dominant Species in Shrub Layer of Pinus massoniana Community on Chinese Islands

Understanding the factors driving species coexistence and competition in the shrub layer of semi-natural forests is crucial for effective forest management and conservation. However, there is limited knowledge about the interspecific associations of the main species in the shrub layer of Pinus massoniana communities in the semi-natural forest of Sandu Gulf, Ningde, Fujian Province, China. Therefore, this study aimed to investigate the influence of the abiotic environment on plant communities within the semi-natural forest of P. massoniana on the islands of Sandu Gulf. By exploring these interspecific associations, we sought to provide a more accurate understanding of the causes and processes of species coexistence and competition. The ultimate goal of this project was to offer a reference basis for optimizing the shrub layer structure in P. massoniana (plantation) forests. We found that (1) Heptapleurum heptaphyllum was the most dominant species in the shrub layer, while Smilax china demonstrated the broadest range of environmental adaptability and correspondingly broader niche than other species. (2) Our analysis revealed a predominance of positive associations among the dominant species in the shrub layer. However, the overall association was not significant, with relatively small positive and negative associations between species pairs. The significant test rate was low, and the NRI exhibited a non-significant aggregation. These findings suggest that the plant community in the shrub layer has not yet reached its most stable stage. (3) We also observed that the distribution of dominant species in the shrub layer was primarily affected by factors such as total potassium, pH, available potassium, and light (canopy density). (4) Soil pH value decreased gradually as sampling points moved inward away from the coastline, and island isolation, temperature, and precipitation indirectly affected the species' importance in the shrub layer. Considering the intense competition among the understory species, it is crucial for conservation efforts to prioritize species pairs with reduced ecological niche overlap or significant positive associations. This approach will effectively reduce competition and contribute to the maintenance of stability in the shrub layer. This study provides a theoretical basis for environmental and vegetation restoration, optimizing the species composition of island plantation forests, rationalizing plant composition, and implementing effective operation and management practices for local P. massoniana forests.

Significance of pyroptosis-related genes in the diagnosis and classification of diabetic kidney disease

OBJECTIVE

This study aimed to identify the potential biomarkers associated with pyroptosis in diabetic kidney disease (DKD).

METHODS

Three datasets from the Gene Expression Omnibus (GEO) were downloaded and merged into an integrated dataset. Differentially expressed genes (DEGs) were filtered and intersected with pyroptosis-related genes (PRGs). Pyroptosis-related DEGs (PRDEGs) were obtained and analyzed using functional enrichment analysis. Random forest, Least Absolute Shrinkage and Selection Operator, and logistic regression analyses were used to select the features of PRDEGs. These feature genes were used to build a diagnostic prediction model, identify the subtypes of the disease, and analyze their interactions with transcription factors (TFs)/miRNAs/drugs and small molecules. We conducted a comparative analysis of immune cell infiltration at different risk levels of pyroptosis. qRT-PCR was used to validate the expression of the feature genes.

RESULTS

A total of 25 PRDEGs were obtained. These genes were coenriched in biological processes and pathways, such as the regulation of inflammatory responses. Five key genes (CASP1, CITED2, HTRA1, PTGS2, S100A12) were identified and verified using qRT-PCR. The diagnostic model based on key genes has a good diagnostic prediction ability. Five key genes interacted with TFs and miRNAs in 67 and 80 pairs, respectively, and interacted with 113 types of drugs or molecules. Immune infiltration of samples with different pyroptosis risk levels showed significant differences. Thus, CASP1, CITED2, HTRA1, PTGS2 and S100A12 are potential DKD biomarkers.

CONCLUSION

Genes that regulate pyroptosis can be used as predictors of DKD. Early diagnosis of DKD can aid in its effective treatment.

Identification of Renal Ischemia-Reperfusion Injury Subtypes and Predictive Model for Graft Loss after Kidney Transplantation Based on Programmed Cell Death-Related Genes

Introduction

Ischemia-reperfusion injury (IRI) is detrimental to kidney transplants and may contribute to poor long-term outcomes of transplantation. Programmed cell death (PCD), a regulated cell death form triggered by IRI, is often indicative of an unfavorable prognosis following transplantation. However, given the intricate pathophysiology of IRI and the considerable variability in clinical conditions during kidney transplantation, the specific patterns of cell death within renal tissues remain ambiguous. Consequently, accurately predicting the outcomes for transplanted kidneys continues to be a formidable challenge.

Methods

Eight Gene Expression Omnibus datasets of biopsied transplanted kidney samples post-IRI and 1,548 PCD-related genes derived from 18 PCD patterns were collected in our study. Consensus clustering was performed to identify distinct IRI subtypes based on PCD features (IRI PCD subtypes). Differential enrichment analysis of cell death, metabolic signatures, and immune infiltration across these subtypes was evaluated. Three machine learning algorithms were used to identify PCD patterns related to prognosis. Genes associated with graft loss were screened for each PCD type. A predictive model for graft loss was constructed using 101 combinations of 10 machine learning algorithms.

Results

Four IRI subtypes were identified: PCD-A, PCD-B, PCD-C, and PCD-D. PCD-A, characterized by high enrichment of multiple cell death patterns, significant metabolic paralysis, and immune infiltration, showed the poorest prognosis among the four subtypes. While PCD-D involved the least kind of cell death patterns with the features of extensive activation of metabolic pathways and the lowest immune infiltration, correlating with the best prognosis in the four subtypes. Using various machine learning algorithms, 10 cell death patterns and 42 PCD-related genes were identified as positively correlated with graft loss. The predictive model demonstrated high sensitivity and specificity, with area under the curve values for 0.5-, 1-, 2-, 3-, and 4-year graft survival at 0.888, 0.91, 0.926, 0.923, and 0.923, respectively.

Conclusion

Our study explored the comprehensive features of PCD patterns in transplanted kidney samples post-IRI. The prediction model shows great promise in forecasting graft loss and could aid in risk stratification in patients following kidney transplantation.

Machine Learning-Assisted Surface-Enhanced Raman Spectroscopy Detection for Environmental Applications: A Review

Surface-enhanced Raman spectroscopy (SERS) has gained significant attention for its ability to detect environmental contaminants with high sensitivity and specificity. The cost-effectiveness and potential portability of the technique further enhance its appeal for widespread application. However, challenges such as the management of voluminous quantities of high-dimensional data, its capacity to detect low-concentration targets in the presence of environmental interferents, and the navigation of the complex relationships arising from overlapping spectral peaks have emerged. In response, there is a growing trend toward the use of machine learning (ML) approaches that encompass multivariate tools for effective SERS data analysis. This comprehensive review delves into the detailed steps needed to be considered when applying ML techniques for SERS analysis. Additionally, we explored a range of environmental applications where different ML tools were integrated with SERS for the detection of pathogens and (in)organic pollutants in environmental samples. We sought to comprehend the intricate considerations and benefits associated with ML in these contexts. Additionally, the review explores the future potential of synergizing SERS with ML for real-world applications.

Identification and validation of a novel defined stress granule-related gene signature for predicting the prognosis of ovarian cancer via bioinformatics analysis

Ovarian cancer (OC) is a malignant gynecological cancer with an extremely poor prognosis. Stress granules (SGs) are non-membrane organelles that respond to stressors; however, the correlation between SG-related genes and the prognosis of OC remains unclear. This systematic analysis aimed to determine the expression levels of SG-related genes between high- and low-risk groups of patients with OC and to explore the prognostic value of these genes. RNA-sequencing data and clinical information from GSE18520 and GSE14407 in the Gene Expression Omnibus (GEO) and ovarian plasmacytoma adenocarcinoma in The Cancer Genome Atlas (TCGA) were downloaded. SG-related genes were obtained from GeneCards, the Molecular Signatures Database, and the literature. First, 13 SG-related genes were identified in the prognostic model using least absolute shrinkage and selection operator (LASSO) Cox regression. The prognostic value of each SG-related gene for survival and its relationship with clinical characteristics were evaluated. Next, we performed a functional enrichment analysis of SG-related genes. The protein-protein interactions (PPI) of SG-related genes were visualized using Cytoscape with STRING. According to the median risk score from the LASSO Cox regression, a 13-gene signature was created. All patients with OC in TCGA cohort and GEO datasets were classified into high- and low-risk groups. Five SG-related genes were differentially expressed between the high- and low-risk OC groups in the GEO datasets. The 13 SG-related genes were related to several important oncogenic pathways (TNF-α signaling, PI3K-AKT-mTOR signaling, and WNT-β-catenin signaling) and several cellular components (cytoplasmic stress granule, cytoplasmic ribonucleoprotein granule, and ribonucleoprotein granule). The PPI network identified 11 hub genes with the strongest interactions with ELAVL1. These findings indicate that SG-related genes (DNAJA1, ELAVL1, FBL, GRB7, MOV10, PABPC3, PCBP2, PFN1, RFC4, SYNCRIP, USP10, ZFP36, and ZFP36L1) can be used to predict OC prognosis.

Genome-Wide Association Study and Phenotype Prediction of Reproductive Traits in Large White Pigs

In a study involving 385 Large White pigs, a genome-wide association study (GWAS) was conducted to investigate reproductive traits, specifically the number of healthy litters (NHs) and the number of weaned litters (NWs). Several SNP loci, including ALGA0098819, ALGA0037969, and H3GA0032302, were significantly associated with these traits. In the combined-parity analysis, candidate genes, such as BLVRA, STK17A, PSMA2, and C7orf25, were identified. GO and KEGG pathway enrichment analyses revealed that these genes are involved in key biological processes, including organic synthesis, the regulation of sperm activity, spermatogenesis, and meiosis. In the by-parity analysis, the PLCXD3 gene was significantly associated with the NW trait in the second and fourth parities, while RNASEH1, PYM1, and SEPTIN9 were linked to cell proliferation, DNA repair, and metabolism, suggesting their potential role in regulating reproductive traits. These findings provide new molecular markers for the genetic study of reproductive traits in Large White pigs. For the phenotypic prediction of NH and NW traits, several machine learning models (GBDT, RF, LightGBM, and Adaboost.R2), as well as traditional models (GBLUP, BRR, and BL), were evaluated using SNP data in varying proportions. After PCA processing, the GBDT model achieved the highest PCC for NH (0.141), while LightGBM reached the highest PCC for NW (0.146). The MAE, MSE, and RMSE results showed that the traditional models exhibited stable error rates, while the machine learning models performed comparatively better across the different SNP ratios. Overall, PCA processing provided some improvement in the predictive performance of all of the models, though the overall increase in accuracy was limited.

Constitutive expression of cucumber CsACS2 in Arabidopsis Thaliana disrupts anther dehiscence through ethylene signaling and DNA methylation pathways

KEY MESSAGE

Constitutive expression of cucumber CsACS2 in Arabidopsis disrupts anther dehiscence and male fertility via ethylene signaling and DNA methylation, revealing new avenues for enhancing crop reproductive traits. The cucumber gene CsACS2, encoding ACC (1-aminocyclopropane-1-carboxylic acid) synthase, plays a pivotal role in ethylene biosynthesis and sex determination. This study investigates the effects of constitutive CsACS2 expression in Arabidopsis thaliana on anther development and male fertility. Transgenic Arabidopsis plants overexpressing CsACS2 exhibited male sterility due to inhibited anther dehiscence, which was linked to suppressed secondary cell wall thickening. RNA-Seq analysis revealed upregulation of ethylene signaling pathway genes and downregulation of secondary cell wall biosynthesis genes, with gene set enrichment analysis indicating the involvement of DNA methylation. Rescue experiments demonstrated that silver nitrate (AgNO₃) effectively restored fertility, while 5-azacytidine (5-az) partially restored it, highlighting the roles of ethylene signaling and DNA methylation in this process. Constitutive CsACS2 expression in Arabidopsis disrupts anther development through ethylene signaling and DNA methylation pathways, providing new insights into the role of ethylene in plant reproductive development and potential applications in crop improvement.

Homeotic and nonhomeotic patterns in the tetrapod vertebral formula

Vertebrate development and phylogeny are intimately connected through the vertebral formula, the numerical distribution of vertebrae along the body axis into different categories such as neck and chest. A key window into this relationship is through the conserved Hox gene clusters. Hox gene expression boundaries align with vertebral boundaries, and their manipulation in model organisms often results in the transformation of one vertebral type into its neighbor, a homeotic transformation. If the variety in the vertebrate body plan is produced by homeotic shifts, then the number of adjacent vertebrae will be inversely related when making interspecies comparisons since the gain in one vertebra is due to the loss in its neighbor. To date, such a pattern across species consistent with homeotic transitions has only been found in the thoracolumbar vertebral count of mammals. To further investigate potential homeotic relationships in other vertebrate classes and along the entire body axis, we compiled a comprehensive dataset of complete tetrapod vertebral formulas and systematically searched for patterns by analyzing combinations of vertebrae. We uncovered mammalian homeotic patterns and found balances between distal vertebrae not anticipated by a Hox-vertebral homeotic relationship, including one that emerged during the progression from theropods to birds. We also identified correlations between vertebral counts and intergenic distances in the HoxB gene cluster which do not align with the common picture of a colinear relationship between Hox expression and vertebral categories. This quantitative approach revises our expectations for the diversity of a Hox-mediated vertebrate body plan.

Direct reprogramming of fibroblasts into spiral ganglion neurons by defined transcription factors

Degeneration of the cochlear spiral ganglion neurons (SGNs) is one of the major causes of sensorineural hearing loss and significantly impacts the outcomes of cochlear implantation. Functional regeneration of SGNs holds great promise for treating sensorineural hearing loss. In this study, we systematically screened 33 transcriptional regulators implicated in neuronal and SGN fate. Using gene expression array and principal component analyses, we identified a sequential combination of Ascl1, Pou4f1 and Myt1l (APM) in promoting functional reprogramming of SGNs. The neurons induced by APM expressed mature neuronal and SGN lineage-specific markers, displayed mature SGN-like electrophysiological characteristics and exhibited single-cell transcriptomes resembling the endogenous SGNs. Thus, transcription factors APM may serve as novel candidates for direct reprogramming of SGNs and hearing recovery due to SGN damages.

Magnetic Resonance Imaging Texture Analysis Based on Intraosseous and Extraosseous Lesions to Predict Prognosis in Patients with Osteosarcoma

Objectives: To construct an optimal magnetic resonance imaging (MRI) texture model to evaluate histological patterns and predict prognosis in patients with osteosarcoma (OS). Methods: Thirty-four patients underwent pretreatment MRI and were diagnosed as having OS by surgical resection or biopsy between September 2008 and June 2018. Histological patterns and 3-year survival were recorded. Manual segmentation was performed in intraosseous, extraosseous, and entire lesions on T1-weighted, T2-weighted, and contrast-enhanced T1-weighted images to extract texture features and perform principal component analysis. A support vector machine algorithm with 3-fold cross-validation was used to construct and validate the models. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate diagnostic performance in evaluating histological patterns and 3-year survival. Results: Eight patients were chondroblastic and the remaining twenty-six patients were non-chondroblastic patterns. Twenty-seven patients were 3-year survivors, and the remaining seven patients were non-survivors. In discriminating chondroblastic from non-chondroblastic patterns, the model from extraosseous lesions on the T2-weighted images showed the highest diagnostic performance (AUCs of 0.94 and 0.89 in the training and validation sets). The model from intraosseous lesions on the T1-weighted images showed the highest diagnostic performance in discriminating 3-year non-survivors from survivors (AUCs of 0.99 and 0.88 in the training and validation sets) with a sensitivity, specificity, positive predictive value, and negative predictive value of 85.7%, 92.6%, 75.0%, and 96.2%, respectively. Conclusions: The texture models of extraosseous lesions on T2-weighted images can discriminate the chondroblastic pattern from non-chondroblastic patterns, while the texture models of intraosseous lesions on T1-weighted images can discriminate 3-year non-survivors from survivors.

Integrated bioinformatics analysis and in vivo validation of potential immune-related genes linked to diabetic nephropathy

Background

Diabetic nephropathy (DN) is a common microvascular complication of diabetes mellitus and the main cause of chronic renal failure. This study explored the potential immunomodulation-related genes (IRGs) in DN using bioinformatics.

Methods

IRGs were identified using GeneCards, and differentially expressed genes were identified using the GSE99339, GSE96804, and GSE30122 datasets. We conducted enrichment analyses using Gene Ontology, gene set enrichment analysis (GSEA), and Kyoto Encyclopedia of Genes and Genomes to identify the associated signaling pathways. Prognostic models were constructed using Least Absolute Shrinkage and Selection Operator regression. The predictive power of IRGs was evaluated using receiver operating characteristic (ROC) curves. Furthermore, we utilized ssGSEA to determine the relative abundance of immune cell infiltration. The expression of five significant IRGs was further validated using immunohistochemistry (IHC) in individuals with DN and real-time PCR (RT-PCR) in animal experiments.

Results

In total, 17 immunomodulation-related differentially expressed genes were identified, which were enriched in immune-associated pathways and inflammation. GSEA unveiled substantial enrichments in metabolic irregularities and the structural composition of the extracellular matrix. ROC analysis results revealed that the diagnostic efficacy of IFNAR2 and CASP3 was comparatively high. Notably, we identified potential IRGs for DN, including CASP3, LGALS9, and SST, using IHC and RT-PCR.

Conclusions

CASP3, LGALS9, and SST are potential IRGs in patients with DN. Our findings may offer a theoretical basis for developing more focused and innovative immunotherapy approaches for patients with DN.

Comparative analysis of White and African American groups reveals unique lipid and inflammatory features of diabetes

Importance

African Americans have a higher prevalence of Type 2 Diabetes (T2D) compared to White groups. T2D is a health disparity clinically characterized by dysregulation of lipids and chronic inflammation. However, how the relationships among biological and sociological predictors of T2D drive this disparity remains to be addressed.

Objective

To determine characteristic plasma lipids and systemic inflammatory biomarkers contributing to diabetes presentation between White and African American groups.

Design

We performed a cross-sectional retrospective cohort study using pre-existing demographic and clinical data from two diverse studies: Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) and AllofUs. From HANDLS (N=40), we used information from wave 1 (2004). From AllofUs (N=17,339), we used data from the Registered Tier Dataset v7, available in the AllofUs researcher workbench.

Setting

HANDLS is a population-based cohort study involving 3720 participants in the Baltimore area supported by the Intramural Research Program of the National Institute on Aging. HANDLS is a longitudinal study designed to understand the sources of persistent health disparities in overall longevity and chronic disease in White and African American individuals. The AllofUs study is an NIH funded multicenter study consisting of patient-level data from 331,382 individuals from 35 hospitals in the United States aimed at sampling one million or more people living in the United States to provide a collection of broadly accessible data.

Participants

The HANDLS subcohort participants (N=40) were divided into four groups equally distributed by race, sex, and diabetes status. Groups were also matched by age, body mass index, and poverty status. The analysis pipeline consisted of evaluating the significance of the variables race and disease status using the 2-way ANOVA test and post-ANOVA comparisons using Fisher LSD test, reporting unadjusted p-values. Additionally, unsupervised (PCA) and supervised (OPLS-DA) clustering analysis was performed to determine putative biological drivers of variability and main immunological and metabolic features characterizing diabetes in White and African American groups from HANDLS. Major clinical findings were validated in a large cohort of White and African American groups with T2D in the AllofUS research study (N=17,339). AllofUs groups were of similar range in age and BMI as HANDLS. Furthermore, a linear regression model was built adjusting for age and BMI to determine differences in clinical findings between White and African American groups with T2D.

Main Outcomes and Measures

Primary outcomes using a HANDLS subcohort (N=40) were clinical parameters related to diabetes, plasma lipids determined by lipidomics and measured by mass spectrometry, and cytokine profiling using a customized panel of 52 cytokines and growth factors measured by Luminex. Outcomes evaluated in the AllofUs study (N=17,339) were clinical: cholesterol to HDL ratio, triglycerides, fasting glucose, insulin, and hemoglobin A1C.

Results

In the HANDLS subcohort, White individuals with diabetes had elevated cholesterol to HDL ratio (mean difference -1.869, p =0.0053 ) , high-sensitivity C-reactive protein (mean difference -9.135, p =0.0040), and clusters of systemic triglycerides measured by lipidomics, compared to White individuals without diabetes. These clinical markers of dyslipidemia (cholesterol to HDL ratio and triglycerides) and inflammation (hs-CRP) were not significantly elevated in diabetes in African Americans from the HANDLS subcohort. These results persisted even when controlling for statin use. Diabetes in White individuals in the HANDLS cohort was characterized by a marked elevation in plasma lipids, while an inflammatory status characterized by Th17-cytokines was predominant in the African American group from the HANDLS subcohort. We validated the key findings of elevated triglycerides and cholesterol to HDL ratio in White individuals with T2D in a sample (N=17,339) of the AllofUs study.

Conclusions and Relevance

Our results show that diabetes can manifest with healthy lipid profiles, particularly in these cohorts of African Americans. This study suggests that Th17-inflammation associated with diabetes is characteristic of African Americans, while a more classic inflammation is distinctive of White individuals from HANDLS cohort. Further, clinical markers of dyslipidemia seem to characterize diabetes presentation only in White groups, and not in African Americans.

Exploring structural variances in monatomic metallic glasses using machine learning and molecular dynamics simulation

CONTEXT

BCC and FCC metals have different glass-forming abilities (GFA) and exhibit different characteristics during the glass transition. However, the structural origin of their different GFAs is still not clear. Here, we explored the structures of eight monatomic metallic glasses by combining molecular dynamics (MD) simulations and machine learning (ML). Our findings reveal that, despite their common long-range disordered atomic structure, metallic glasses can be further classified into two distinct categories indicating an underlying structural order within the disorder. Using machine learning, we found that BCC liquids can sample more diverse glass states than FCC liquids. Furthermore, glasses formed from BCC metals (GFFBs) exhibit a higher degree of disorder than glasses formed from FCC metals (GFFFs). These findings highlight the inherent differences between GFFFs and GFFBs, which help explain the different glass-forming abilities of FCC and BCC metals. Additionally, our results demonstrate the promising potential of computer vision and ML methods in exploring material structures.

METHOD

Classical molecular dynamics simulations were employed to generate configurations of GFFBs and GFFFs, and the simulations were performed using the LAMMPS code. Inter-atomic interactions were described using a classical embedded atom model (EAM) potential. The initial configuration of the model consists of 32,000 atoms in a three-dimensional (3D) cubic box with periodic boundary conditions applied in all three directions. For machine learning, we utilized an unsupervised machine learning method along with MobileNetV2 for classifying glass structures. Image entropy and image distances were used to measure the structural differences of the metallic glasses.

Classification models and SAR analysis of anaplastic lymphoma kinase (ALK) inhibitors using machine learning algorithms with two data division methods

Anaplastic lymphoma kinase (ALK) plays a critical role in the development of various cancers. In this study, the dataset of 1810 collected inhibitors were divided into a training set and a test set by the self-organizing map (SOM) and random method, respectively. We developed 32 classification models using Support Vector Machines (SVM), Decision Trees (DT), Random Forests (RF), and Extreme Gradient Boosting (XGBoost) to distinguish between highly and weakly active ALK inhibitors, with the inhibitors represented by MACCS and ECFP4 fingerprints. Model 7D which was built by the RF algorithm using training set 1/test set 1 divided by the SOM method, provided the best performance with a prediction accuracy of 90.97% and a Matthews correlation coefficient (MCC) value of 0.79 on the test set. We clustered the 1810 inhibitors into 10 subsets by K-Means algorithm to find out the structural characteristics of highly active ALK inhibitors. The main scaffolds of highly active ALK inhibitors were also analyzed based on ECFP4 fingerprints. It was found that some substructures have a significant effect on high activity, such as 2,4-diarylaminopyrimidine analogues, pyrrolo[2,1-f][1,2,4]triazin, indolo[2,3-b]quinoline-11-one, benzo[d]imidazol and pyrrolo[2,3-b]pyridine. In addition, the subsets were summarized into several clusters, among which four clusters showed a significant relationship with ALK inhibitory activity. Finally, Shapley additive explanations (SHAP) was also used to explain the influence of modeling features on model prediction results. The SHAP results indicated that our models can well reflect the structural features of ALK inhibitors.

Moving beyond species: fungal function in house dust provides novel targets for potential indicators of mold growth in homes

BACKGROUND

Increased risk of asthma and other respiratory diseases is associated with exposures to microbial communities growing in damp and moldy indoor environments. The exact causal mechanisms remain unknown, and occupant health effects have not been consistently associated with any species-based mold measurement methods. We need new quantitative methods to identify homes with potentially harmful fungal growth that are not dependent upon species. The goal of this study was to identify genes consistently associated with fungal growth and associated function under damp conditions for use as potential indicators of mold in homes regardless of fungal species present. A de novo metatranscriptomic analysis was performed using house dust from across the US, incubated at 50%, 85%, or 100% equilibrium relative humidity (ERH) for 1 week.

RESULTS

Gene expression was a function of moisture (adonis2 p < 0.001), with fungal metabolic activity increasing with an increase in moisture condition (Kruskal-Wallis p = 0.003). Genes associated with fungal growth such as sporulation (n = 264), hyphal growth (n = 62), and secondary metabolism (n = 124) were significantly upregulated at elevated ERH conditions when compared to the low 50% ERH (FDR-adjusted p ≤ 0.001, log2FC ≥ 2), indicating that fungal function is influenced by damp conditions. A total of 67 genes were identified as consistently associated with the elevated 85% or 100% ERH conditions and included fungal developmental regulators and secondary metabolite genes such as brlA (log2FC = 7.39, upregulated at 100% compared to 85%) and stcC (log2FC = 8.78, upregulated at 85% compared to 50%).

CONCLUSIONS

Our results demonstrate that moisture conditions more strongly influence gene expression of indoor fungal communities compared to species presence. Identifying genes indicative of microbial growth under damp conditions will help develop robust monitoring techniques for indoor microbial exposures and improve understanding of how dampness and mold are linked to disease. Video Abstract.

A molecular video-derived foundation model for scientific drug discovery

Accurate molecular representation of compounds is a fundamental challenge for prediction of drug targets and molecular properties. In this study, we present a molecular video-based foundation model, named VideoMol, pretrained on 120 million frames of 2 million unlabeled drug-like and bioactive molecules. VideoMol renders each molecule as a video with 60-frame and designs three self-supervised learning strategies on molecular videos to capture molecular representation. We show high performance of VideoMol in predicting molecular targets and properties across 43 drug discovery benchmark datasets. VideoMol achieves high accuracy in identifying antiviral molecules against common diverse disease-specific drug targets (i.e., BACE1 and EP4). Drugs screened by VideoMol show better binding affinity than molecular docking, revealing the effectiveness in understanding the three-dimensional structure of molecules. We further illustrate interpretability of VideoMol using key chemical substructures.

Digital identification and adulteration analysis of Codonopsis Radix and Stellariae Radix based on the "digital identity" of chemical compositions

Introduction

Under the background of digitalization of traditional Chinese medicine (TCM), this study aimed to realize the digital identification and adulteration analysis of Codonopsis Radix (CR) and Stellariae Radix (SR) based on chemical analysis.

Methods

This study combined digitalization concepts and chemical analysis and conducted a chemical analysis of CR and SR from different batches based on UHPLC-QTOF-MSE. Furthermore, the shared ions were extracted from different batches of CR and SR as their "ion characterization" after digital quantization. Then, the data matrices of unique ions of CR relative to SR and SR relative to CR were screened out, and the top-N ions were outputted as the "digital identities" of CR and SR, sorted by ionic strength. Finally, the above "digital identities" of CR and SR were used as benchmarks for matching positive samples and market samples to provide feedback on the matching credibility (MC) for identification and adulteration analysis.

Results

The results showed that based on the "digital identities" of CR and SR, the digital identification of CR, SR, and positive samples can be realized at the individual level of TCM efficiently and accurately, even if 3% of SR in the mixed samples can still be identified efficiently and accurately. Moreover, 1 of the 12 batches of market samples was identified as an adulterated sample.

Conclusion

It proved that the identification and adulteration analysis of two herbs can be realized efficiently and quickly through the "digital identities" of chemical compositions. It has important reference significance for developing the digital identification of CR and SR at the individual level of Chinese medicine based on the "digital identity" of chemical compositions, which was beneficial to the construction of digital quality control of CR and SR.

Automated mapping of electronic data capture fields to SDTM

OBJECTIVE

The goal of this work is to reduce the amount of manual work required to go from data capture to regulatory submission. It will be shown that the use of Siamese networks will allow for the generation of embeddings that can be used by traditional machine learning classifiers to perform the classification at much higher levels of accuracy than standard approaches.

METHODS

Siamese networks are a method for training data embeddings such that data within the same class are closer with respect to a given distance metric than they are to data points in another class. Because they are designed to learn similarity within pairs of data points, they work well in situations where the number of classes is relatively large compared to the number of training samples. In this work, we will show that embeddings generated via a Siamese network from metadata associated with electronic data capture forms can be used to predict the associated SDTM field.

RESULTS

With a relatively simple network coupled with a basic classification algorithm, the proposed method can achieve accuracies greater than 90%, which is significantly higher than what has been achieved with traditional methods, with many of the inaccurate mappings due to a lack of training data. In many cases, there is a 15% increase in accuracy vs. more traditional methods.

CONCLUSION

Leveraging Siamese networks, it is possible to generate embeddings that efficiently represent data fields in a lower dimensional space. This allows the creation of a system that can automatically map between data schemas at high levels of accuracy. Such systems represent the first step in automating one of the many labor-intensive data management tasks associated with clinical trials.