Predictive Tool Research Articles

Development and validation of a clinical prediction model for glioma grade using machine learning.

Histopathological evaluation is currently the gold standard for grading gliomas; however, this technique is invasive. This study aimed to develop and validate a diagnostic prediction model for glioma by employing multiple machine learning algorithms to identify risk factors associated with high-grade glioma, facilitating the prediction of glioma grading. Data from 1114 eligible glioma patients were obtained from The Cancer Genome Atlas (TCGA) database, which was divided into a training set (n= 781) and a test set (n= 333). Fifty machine learning algorithms were employed, and the optimal algorithm was selected to construct a prediction model. The performance of the machine learning prediction model was compared to the clinical prediction model in terms of discrimination, calibration, and clinical validity to assess the performance of the prediction model. The area under the curve (AUC) values of the machine learning prediction models (training set: 0.870 vs. 0.740, test set: 0.863 vs. 0.718) were significantly improved from the clinical prediction models. Furthermore, significant improvement in discrimination was observed for the Integrated Discrimination Improvement (IDI) (training set: 0.230, test set: 0.270) and Net Reclassification Index (NRI) (training set: 0.170, test set: 0.170) from the clinical prognostic model. Both models showed a high goodness of fit and an increased net benefit. A strong prediction accuracy model can be developed using machine learning algorithms to screen for high-grade glioma risk predictors, which can serve as a non-invasive prediction tool for preoperative diagnostic grading of glioma.

Five-Axis Finish Milling Machining for an Inconel 718 Alloy Monolithic Blisk

Blisks subjected to rough machining for channel creation must undergo finishing processes, and such processes must achieve the required tolerance limits. A high-quality surface finish and predictable long tool life are critical for the finish milling of blisks. Accordingly, the aim of this study was to optimize parameters for the finish machining of an Inconel 718 alloy monolithic blisk. Ball-cone mills were used to machine the blade surface at a constant depth. A sensory tool holder was used to collect cutting force signals during machining, and a digital microscope was used to examine tool wear. The surface texture measuring instrument was used to measure blisk blade surface roughness to evaluate processing quality. This study manipulated two cutting parameters, namely cutting speed and feed per tooth, and investigated their effects. The relationship between cutting conditions and machining efficiency was analyzed. According to the experimental results, we identified a set of optimal parameters at effective cutting speeds of 46.53 m/min, feed per tooth of 0.1 mm/tooth, and depth of cut of 0.05 mm for marginal tool wear and fast cutting speeds. Then the corresponding tool life was estimated by using the derived parameters.

Bipartite networks represent causality better than simple networks: evidence, algorithms, and applications.

A network, whose nodes are genes and whose directed edges represent positive or negative influences of a regulatory gene and its targets, is often used as a representation of causality. To infer a network, researchers often develop a machine learning model and then evaluate the model based on its match with experimentally verified "gold standard" edges. The desired result of such a model is a network that may extend the gold standard edges. Since networks are a form of visual representation, one can compare their utility with architectural or machine blueprints. Blueprints are clearly useful because they provide precise guidance to builders in construction. If the primary role of gene regulatory networks is to characterize causality, then such networks should be good tools of prediction because prediction is the actionable benefit of knowing causality. But are they? In this paper, we compare prediction quality based on "gold standard" regulatory edges from previous experimental work with non-linear models inferred from time series data across four different species. We show that the same non-linear machine learning models have better predictive performance, with improvements from 5.3% to 25.3% in terms of the reduction in the root mean square error (RMSE) compared with the same models based on the gold standard edges. Having established that networks fail to characterize causality properly, we suggest that causality research should focus on four goals: (i) predictive accuracy; (ii) a parsimonious enumeration of predictive regulatory genes for each target gene g; (iii) the identification of disjoint sets of predictive regulatory genes for each target g of roughly equal accuracy; and (iv) the construction of a bipartite network (whose node types are genes and models) representation of causality. We provide algorithms for all goals.

Determination and Correlation of Finger Print Pattern and Blood Grouping in Diabetes Mellitus: An Analytical Study

BACKGROUND: To Determine and correlate between fingerprint patterns and blood group in type II diabetes mellitus. RESULTS: One hundred individuals with type II diabetes mellitus and One hundred normal individuals as controls were selected for study. Fingerprints were obtained by dactyloscopy. Various fingerprint patterns like arches, whorls, loops were studied and correlated with the respective blood group based on ABO system among the volunteer controls and diabetics. After obtaining results, the analysis was made using Chi square test and One way ANOVA which showed, the most predominant fingerprint pattern among the 100 diabetic volunteers is “Arch pattern” and the most predominant blood group was “O” positive among the diabetic volunteers. Also correlating both finger print and blood group “Whorl pattern” with “B” blood group showed higher predominance among the healthy volunteers. CONCLUSIONS: This is the first study correlating finger print pattern with blood grouping in subjects of Diabetes mellitus. Our study findings showed that O positive blood group with arch pattern type of finger print was the most common findings among diabetes. Thus correlating fingerprint pattern and blood grouping act as a early predictor tool for diagnosing Type II Diabetes mellitus.

A Novel Investigation for Early Sex Determination in Alive Adult European Seabass (Dicentrarchus labrax) Using cyp19a1a, dmrt1a, and dmrt1b Genes Expression in Tail Fin tissues.

This study is the first investigation for using sex-related gene expression in tail fin tissues of seabass as early sex determination without killing the fish. The European seabass (Dicentrarchus labrax) is gonochoristic and lacks distinguishable sex chromosomes, so, sex determination is referred to molecular actions for some sex-related genes on autosomal chromosomes which are well known such as cyp19a1a, dmrt1a, and dmrt1b genes which play crucial role in gonads development and sex differentiation. cyp19a1a is expressed highly in females for ovarian development and dmrt1a and dmrt1b are for testis development in males. In this study, we evaluated the difference in the gene expression levels of studied genes by qPCR in tail fins and gonads. We then performed discriminant analysis (DA) using morphometric traits and studied gene expression parameters as predictor tools for fish sex. The results revealed that cyp19a1a gene expression was significantly higher in future females' gonads and tail fins (p ≥ 0.05). Statistically, cyp19a1a gene expression was the best parameter to discriminate sex even the hit rate of any other variable by itself could not correctly classify 100% of the fish sex except when it was used in combination with cyp19a1a. In contrast, Dmrt1a gene expression was higher in males than females but there were difficulties in analyzing dmrt1a and dmrt1b expressions in the tail because levels were low. So, it could be used in future research to differentiate and determine the sex of adult fish using the cyp19a1a gene expression marker without killing or sacrificing fish.

Exploiting genomic tools for genetic dissection and improving the resistance to Fusarium stalk rot in tropical maize.

A stable genomic region conferring FSR resistance at ~250 Mb on chromosome 1 was identified by GWAS. Genomic prediction has the potential to improve FSR resistance. Fusarium stalk rot (FSR) is a global destructive disease in maize; the efficiency of phenotypic selection for improving FSR resistance was low. Novel genomic tools of genome-wide association study (GWAS) and genomic prediction (GP) provide an opportunity for genetic dissection and improving FSR resistance. In this study, GWAS and GP analyses were performed on 562 tropical maize inbred lines consisting of two populations. In total, 15 SNPs significantly associated with FSR resistance were identified across two populations and the combinedPOP consisting of all 562 inbred lines, with the P-values ranging from 1.99 × 10-7 to 8.27 × 10-13, and the phenotypic variance explained (PVE) values ranging from 0.94 to 8.30%. The genetic effects of the 15 favorable alleles ranged from -4.29 to -14.21% of the FSR severity. One stable genomic region at ~ 250Mb on chromosome 1 was detected across all populations, and the PVE values of the SNPs detected in this region ranged from 2.16 to 5.18%. Prediction accuracies of FSR severity estimated with the genome-wide SNPs were moderate and ranged from 0.29 to 0.51. By incorporating genotype-by-environment interaction, prediction accuracies were improved between 0.36 and 0.55 in different breeding scenarios. Considering both the genome coverage and the threshold of the P-value of SNPs to select a subset of molecular markers further improved the prediction accuracies. These findings extend the knowledge of exploiting genomic tools for genetic dissection and improving FSR resistance in tropical maize.

Involvement of FAM170B-AS1, hsa-miR-1202, and hsa-miR-146a-5p in breast cancer.

FAM170B-AS1 is usually expressed low in all organs except for testicular tissues. No study was performed to explore its role in breast cancer (BC). Contradictory results were reported about hsa-miR-1202 and hsa-miR-146a-5p in BC. The present study aimed to explore the involvement of FAM170B-AS1 in BC using bioinformatics predictive tools, followed by a practical validation besides exploring the impact of hsa-miR-1202 and hsa-miR-146a-5p in BC. This study enrolled 96 female patients with BC, 30 patients with benign breast diseases (BBD), and 25 control subjects. The expressions of circulating FAM170B-AS1, hsa-miR-1202, and hsa-miR-146a-5p were quantified using qRT-PCR. These ncRNAs' associations, predictive, and diagnostic roles in BC were statistically tested. The underlying miRNA/mRNA targets of FAM170B-AS1 in BC were bioinformatically predicted followed by confirmation based on the GEPIA and TCGA databases. The expression of FAM170B-AS1 was upregulated in sera of BC patients and hsa-miR-1202 was upregulated in sera of BBD and BC patients while that of hsa-miR-146a-5p was downregulated in BC. These FAM170B-AS1 was significantly associated with BC when compared to BBD. FAM170B-AS1 and hsa-miR-1202 were statistically associated with the BC's stage, grade, and LN metastasis. FAM170B-AS1 and hsa-miR-146a-5p gave the highest specificity and sensitivity for BC. KRAS and EGFR were predicted to be targeted by FAM170B-AS1 through interaction with hsa-miR-143-3p and hsa-miR-7-5p, respectively. Based on the TCGA database, cancer patients having mutations in FAM170B show good overall survival. The present study reported that for the first time, FAM170B-AS1 may be a potential risk factor, predictive, and diagnostic marker for BC. In addition, FAM170B-AS1 might be involved in BC by interacting with hsa-miR-143-3p/KRAS and hsa-miR-7-5p/EGFR through enhancement or repression that may present a new therapeutic option for BC.

Identification of Novel Stemness-based Subtypes and Construction of a Prognostic Risk Model for Patients with Lung Squamous Cell Carcinoma.

Although cancer stem cells (CSCs) contribute to tumorigenesis, progression, and drug resistance, stemness-based classification and prognostic signatures of lung squamous cell carcinoma (LUSC) remain unclarified. This study attempted to identify stemness-based subtypes and develop a prognostic risk model for LUSC. Based on RNA-seq data from The Cancer Genome Atlas (TCGA), Gene-Expression Omnibus (GEO) and Progenitor Cell Biology Consortium (PCBC), mRNA expression-based stemness index (mRNAsi) was calculated by one-class logistic regression (OCLR) algorithm. A weighted gene coexpression network (WGCNA) was employed to identify stemness subtypes. Differences in mutation, clinical characteristics, immune cell infiltration, and antitumor therapy responses were determined. We constructed a prognostic risk model, followed by validations in GEO cohort, pan-cancer and immunotherapy datasets. LUSC patients with subtype C2 had a better prognosis, manifested by higher mRNAsi, higher tumor protein 53 (TP53) and Titin (TTN) mutation frequencies, lower immune scores and decreased immune checkpoints. Patients with subtype C2 were more sensitive to Imatinib, Pyrimethamine, and Paclitaxel therapy, whereas those with subtype C1 were more sensitive to Sunitinib, Saracatinib, and Dasatinib. Moreover, we constructed stemness-based signatures using seven genes (BMI1, CCDC51, CTNS, EIF1AX, FAM43A, THBD, and TRIM68) and found high-risk patients had a poorer prognosis in the TCGA cohort. Similar results were found in the GEO cohort. We verified the good performance of risk scores in prognosis prediction and therapy responses. The stemness-based subtypes shed novel insights into the potential roles of LUSC-stemness in tumor heterogeneity, and our prognostic signatures offer a promising tool for prognosis prediction and guide therapeutic decisions in LUSC.

Predictive modeling for breast cancer based on machine learning algorithms and features selection methods

Breast cancer is one of the leading causes of death among women worldwide. However, early prediction of breast cancer plays a crucial role. Therefore, strong needs exist for automatic accurate early prediction of breast cancer. In this paper, machine learning (ML) classifiers combined with features selection methods are used to build an intelligent tool for breast cancer prediction. The Wisconsin diagnostic breast cancer (WDBC) dataset is used to train and test the model. Classification algorithms, including support vector machine (SVM), light gradient boosting machine (LightGBM), random forest (RF), logistic regression (LR), k-nearest neighbors (k-NN), and naïve Bayes, were employed. Performance measures for each of them were obtained, namely: accuracy, precision, recall, F-score, Kappa, Matthews correlation coefficient (MCC), and time. The results indicate that without feature selection, LightGBM achieves the highest accuracy at 95%. With minimum redundancy maximum relevance (mRMR) feature selection (15 features), LightGBM outperforms other classifiers, achieving an accuracy of 98%. For Pearson correlation coefficient feature selection (15 features), LightGBM also excels with a 95% accuracy rate. Lasso feature selection (5 features) produces varied results across classifiers, with logistic regression achieving the highest accuracy at 96%. These findings underscore the importance of feature selection in refining model performance and in improving detection for breast cancer.

Predictive analysis of terrorist activities in Thailand's Southern provinces: a deep learning approach

Terrorist activities have been on the rise globally, with Thailand experiencing significant challenges, particularly in its three southern border provinces. This study offers a comprehensive analysis aiming to predict forthcoming terrorist events in these provinces. We employed historical data, categorized into nine groups based on military expert recommendations, to train our prediction model. This research tested the prediction capabilities of various methodologies, including decision trees, naïve Bayesian learning techniques, and deep learning artificial neural networks. Notably, the deep neural network emerged as the superior predictive tool, achieving an impressive accuracy of 98.21% and a root mean square error (RMSE) of 0.59%. The primary anticipated events include bombings, shootings, assaults, and acts of vandalism. Our findings also revealed that Pattani Province was the most affected, accounting for 45% of incidents. Specific districts, such as Panare and Yarang, exhibited high crime rates of 40% and 36.84%, respectively. Yala Province, particularly Bannang Sata District, was identified as the hotspot for shooting incidents, with a rate of 34%.

Generative adversarial networks (GANs): Introduction, Taxonomy, Variants, Limitations, and Applications

AbstractThe growing demand for applications based on Generative Adversarial Networks (GANs) has prompted substantial study and analysis in a variety of fields. GAN models have applications in NLP, architectural design, text-to-image, image-to-image, 3D object production, audio-to-image, and prediction. This technique is an important tool for both production and prediction, notably in identifying falsely created pictures, particularly in the context of face forgeries, to ensure visual integrity and security. GANs are critical in determining visual credibility in social media by identifying and assessing forgeries. As the field progresses, a variety of GAN variations arise, along with the development of diverse assessment techniques for assessing model efficacy and scope. The article provides a complete and exhaustive overview of the most recent advances in GAN model designs, the efficacy and breadth of GAN variations, GAN limits and potential solutions, and the blooming ecosystem of upcoming GAN tool domains. Additionally, it investigates key measures like as Inception Score (IS) and Fréchet Inception Distance (FID) as critical benchmarks for improving GAN performance in contrast to existing approaches.

Prediction of nitrate leaching from soil amended with biosolids by machine learning algorithms

This study focused on employing machine learning algorithms to forecast nitrate leaching from soils treated with biochar and vermicompost derived from sugarcane bagasse. input variables including bulk density, porosity, organic carbon, nitrogen, phosphorus, anion exchange capacity, cation exchange capacity, pH, and electrical conductivity, while nitrate leaching was the target variable for prediction. A comparative analysis of machine learning models indicated that Random Forest Regression outperformed linear regression in the prediction of nitrate leaching. Additionally, among the input variables, anion exchange capacity, cation exchange capacity, bulk density, and EC showed the most significant influence in utilizing these models as predictive tools for nitrate leaching from soils treated with slow-release fertilizers.

Predicting nuclear G-quadruplex RNA-binding proteins with roles in transcription and phase separation

RNA-binding proteins are central for many biological processes and their characterization has demonstrated a broad range of functions as well as a wide spectrum of target structures. RNA G-quadruplexes are important regulatory elements occurring in both coding and non-coding transcripts, yet our knowledge of their structure-based interactions is at present limited. Here, using theoretical predictions and experimental approaches, we show that many chromatin-binding proteins bind to RNA G-quadruplexes, and we classify them based on their RNA G-quadruplex-binding potential. Combining experimental identification of nuclear RNA G-quadruplex-binding proteins with computational approaches, we build a prediction tool that assigns probability score for a nuclear protein to bind RNA G-quadruplexes. We show that predicted G-quadruplex RNA-binding proteins exhibit a high degree of protein disorder and hydrophilicity and suggest involvement in both transcription and phase-separation into membrane-less organelles. Finally, we present the G4-Folded/UNfolded Nuclear Interaction Explorer System (G4-FUNNIES) for estimating RNA G4-binding propensities at http://service.tartaglialab.com/new_submission/G4FUNNIES.

The European GO-VIKING project on flow-induced vibrations: overview and current status

Abstract The interaction between cooling fluid and solid structures (rods, tubes) in nuclear power plants may lead to flow-induced vibrations (FIV), causing material fatigue, fretting wear, and eventually loss of component integrity. This can cause further safety issues as well as substantial standstill costs due to longer or unplanned outages. With the growing computational power, the application of modern 3D numerical simulation tools for the accurate prediction of FIV phenomena is rapidly increasing. In 2022, the GO-VIKING (Gathering expertise On Vibration ImpaKt In Nuclear power Generation) project received a grant within the Horizon Europe research and innovation funding program. Sixteen European and two US partners started their collaboration in the field of FIV experiments and analysis. Over four years, the GO-VIKING project investigates FIV phenomena occurring in nuclear reactor cores and steam generators under single- and two-phase flow conditions. The project’s main objectives are to expand the expertise in the field of FIV through generation of new experimental and high-resolution numerical data; development, improvement, and validation of fluid-structure interaction (FSI) methods for FIV evaluation; training stakeholders in the application of these methods; and synthesizing guidelines for the prediction and assessment of FIV phenomena in nuclear reactors. This paper provides an overview of the GO-VIKING objectives, scientific program, as well as of the main scientific achievements in the first project year.

Machine Learning to Predict Outcomes of Endovascular Intervention for Patients With PAD

Endovascular intervention for peripheral artery disease (PAD) carries nonnegligible perioperative risks; however, outcome prediction tools are limited. To develop machine learning (ML) algorithms that can predict outcomes following endovascular intervention for PAD. This prognostic study included patients who underwent endovascular intervention for PAD between January 1, 2004, and July 5, 2023, with 1 year of follow-up. Data were obtained from the Vascular Quality Initiative (VQI), a multicenter registry containing data from vascular surgeons and interventionalists at more than 1000 academic and community hospitals. From an initial cohort of 262 242 patients, 26 565 were excluded due to treatment for acute limb ischemia (n = 14 642) or aneurysmal disease (n = 3456), unreported symptom status (n = 4401) or procedure type (n = 2319), or concurrent bypass (n = 1747). Data were split into training (70%) and test (30%) sets. A total of 112 predictive features (75 preoperative [demographic and clinical], 24 intraoperative [procedural], and 13 postoperative [in-hospital course and complications]) from the index hospitalization were identified. Using 10-fold cross-validation, 6 ML models were trained using preoperative features to predict 1-year major adverse limb event (MALE; composite of thrombectomy or thrombolysis, surgical reintervention, or major amputation) or death. The primary model evaluation metric was area under the receiver operating characteristic curve (AUROC). After selecting the best performing algorithm, additional models were built using intraoperative and postoperative data. Overall, 235 677 patients who underwent endovascular intervention for PAD were included (mean [SD] age, 68.4 [11.1] years; 94 979 [40.3%] female) and 71 683 (30.4%) developed 1-year MALE or death. The best preoperative prediction model was extreme gradient boosting (XGBoost), achieving the following performance metrics: AUROC, 0.94 (95% CI, 0.93-0.95); accuracy, 0.86 (95% CI, 0.85-0.87); sensitivity, 0.87; specificity, 0.85; positive predictive value, 0.85; and negative predictive value, 0.87. In comparison, logistic regression had an AUROC of 0.67 (95% CI, 0.65-0.69). The XGBoost model maintained excellent performance at the intraoperative and postoperative stages, with AUROCs of 0.94 (95% CI, 0.93-0.95) and 0.98 (95% CI, 0.97-0.99), respectively. In this prognostic study, ML models were developed that accurately predicted outcomes following endovascular intervention for PAD, which performed better than logistic regression. These algorithms have potential for important utility in guiding perioperative risk-mitigation strategies to prevent adverse outcomes following endovascular intervention for PAD.

Assessing the toxicity of pesticides exposure on hepatic miRNA-target gene alterations in rat liver tissues via molecular and integrated network bioinformatics analysis

The prevalent use of pesticides, including pirimiphos-methyl (PPM) and bifenthrin (BF), poses a serious health risk, particularly to workers who encounter these chemicals daily. Despite the recognized hepatotoxic effects, the specific molecular mechanisms, especially those involving miRNAs in liver damage caused by PPM and BF, are not fully elucidated. Prior studies have not exhaustively analyzed the hepatic miRNA-target gene dynamics following exposure to these pesticides; thus, this research aims to fill that gap through an extensive miRNA analysis to discern their regulation in PPM or BF-induced hepatic toxicity. In this study, male Sprague-Dawley rats were exposed to BF or PPM for 28 days through oral gavage, simulating the chronic exposure faced by humans. We conducted a thorough assessment of the hepatotoxicity induced by PPM and BF, employing multiple evaluation levels, including histological analysis, liver enzyme measurements, and real-time PCR to detect changes in hepatic miRNA-target gene expressions. Additionally, we utilized DIANA-miRPath prediction tools to delineate the functional implications of these hepatic miRNA target genes. Our findings reveal a significant modulation in the expression of rno-miR-155–5p and rno-miR-122–5p, along with their target genes, following PPM and BF treatment. In contrast, rno-miR-21–5p levels remained unaltered. These observations suggest potential utility of these specific hepatic miRNAs as biomarkers for liver injury resulting from pesticide exposure. Subsequent GO enrichment analysis linked target genes to functions like molecular activity, protein binding, and cellular processes. Additionally, KEGG pathway analysis showed these genes, influenced by varied miRNA expressions, play significant roles in metabolic and signaling pathways In conclusion, this study enhances our comprehension of the biological roles of miRNAs in hepatic toxicity induced by PPM and BF. The insights gained here not only shed light on molecular mechanisms but also open avenues for considering these miRNAs as potential diagnostic biomarkers in conditions of pesticide-induced hepatotoxicity, thereby guiding future therapeutic strategies.

Situating the phosphonated calixarene-cytochrome C association by molecular dynamics simulations.

Protein-calixarenes binding plays an increasingly central role in many applications, spanning from molecular recognition to drug delivery strategies and protein inhibition. These ligands obey a specific bio-supramolecular chemistry, which can be revealed by computational approaches, such as molecular dynamics simulations. In this paper, we rely on all-atom, explicit-solvent molecular dynamics simulations to capture the electrostatically driven association of a phosphonated calix-[4]-arene with cytochome-C, which critically relies on surface-exposed paired lysines. Beyond two binding sites identified in direct agreement with the x-ray structure, the association has a larger structural impact on the protein dynamics. Then, our simulations allow a direct comparison to analogous calixarenes, namely, sulfonato, similarly reported as "molecular glue." Our work can contribute to a robust in silico predictive tool to assess binding sites for any given protein of interest for crystallization, with the specificity of a macromolecular cage whose endo/exo orientation plays a role in the binding.

Integrating Intellectual Consciousness AI based on Ensemble Machine Learning for Price Negotiation in E-commerce using Text and Voice-Based Chatbot

Online shopping has experienced an enormous boost in recent years. With this evolution, the majority of internet shopping's capabilities have been developed, but some functions, such as negotiating with store owners, are still not available. This paper suggests employing a chatbot with a voice assistant to negotiate product prices. Customers can communicate with the chatbot to get assistance in finding a reasonable price for a product. In online purchasing, there is a possibility that the consumers or the product seller's budget may be compromised. In order to assist in purchasing, algorithm has been created in machine learning that uses the forecasting of historical data to avoid compromising situations. However, improper dataset or when irrelevant aspects or at- tributes of the data are used, price prediction might become less accurate. Ecommerce companies do not merely depend on price prediction tools due to the significant financial losses brought on even by a single inaccurate price prediction. Additionally, few models fail to perform well when the data saturates or when an attribute becomes inaccessible after the period for which the model's prediction was reliant. By controlling these alterations, the accuracy and dependability are preserved in the model pro- posed in this study.

Prodromal symptoms of rheumatoid arthritis in a primary care database: variation by ethnicity and socioeconomic status.

To assess whether prodromal symptoms of rheumatoid arthritis (RA), as recorded in the Clinical Practice Research Datalink Aurum (CPRD) database of English primary care records, differ by ethnicity and socioeconomic status. A cross-sectional study to determine the coding of common symptoms (≥0.1% in the sample) in the 24 months preceding RA diagnosis in CPRD Aurum, recorded between January 1st 2004 to May 1st 2022. Eligible cases were adults with a code for RA diagnosis. For each symptom, a logistic regression was performed with the symptom as dependent variable, and ethnicity and socioeconomic status as independent variables. Results were adjusted for sex, age, BMI, and smoking status. White ethnicity and the highest socioeconomic quintile were comparators. In total, 70115 cases were eligible for inclusion, of which 66.4% female. Twenty-one symptoms were coded in > 0.1% of cases so were included in the analysis. Patients of South Asian ethnicity had higher frequency of codes for several symptoms, with the largest difference by odds ratio being muscle cramps (OR 1.71, 1.44-2.57) and shoulder pain (1.44, 1.25-1.66). Patients of Black ethnicity had higher prevalence of several codes including unintended weight loss (2.02, 1.25-3.28) and ankle pain (1.51, 1.02-2.23). Low socioeconomic status was associated with morning stiffness (1.74, 1.08-2.80) and falls (1.37, 2.03-1.82). There are significant differences in coded symptoms between demographic groups, which must be considered in clinical practice in diverse populations and to avoid algorithmic bias in prediction tools derived from routinely collected healthcare data.

Untargeted NMR-based metabolomics analysis of kidney allograft perfusates identifies a signature of delayed graft function.

Kidney transplantation (KTx) necessarily conveys an ischemia/reperfusion (I/R) process, which impacts on allograft outcomes. Delayed graft function (DGF) is defined as a non-decrease of serum creatinine by at least 10% daily on 3 consecutive days during the first 7days post-KTx. DGF significantly conditions both short- and long-term graft outcomes. Still there is a lack of DGF predictive biomarkers. This study aimed to explore the potential of kidney graft perfusate metabolomics to predict DGF occurrence. 49 human perfusates from grafts categorized upon donor type [donation after brain death (DBD)/donation after circulatory death (DCD)] and DGF occurrence and 19 perfusates from a murine model classified upon death type (DBD/DCD) were collected and analyzed by NMR-based metabolomics. The multivariate analysis of the murine data highlighted significant differences between perfusate metabolomes of DBD versus DCD. These differences were similarly observed in the human perfusates. After correcting for the type of donor, multivariate analysis of human data demonstrated a metabolomics signature that could be correlated with DGF occurrence. The metabolome of kidney grafts is influenced by the donor's type in both human and pre-clinical studies and could be correlated with DGF in the human DBD cohort. Thus, metabolomic analysis of perfusate applied prior to KTx may represent a new predictive tool for clinicians in a more personalized management of DGF. Moreover, our data paves the way to better understand the impact of donor's types on the biochemical events occurring between death and the hypothermic storage.