BiMPADR: A Deep Learning Framework for Predicting Adverse Drug Reactions in New Drugs.

Detecting the unintended adverse reactions of drugs (ADRs) is a crucial concern in pharmacological research. The experimental validation of drug-ADR associations often entails expensive and time-consuming investigations. Thus, a computational model to predict ADRs from known associations is essential for enhanced efficiency and cost-effectiveness. Here, we propose BiMPADR, a novel model that integrates drug gene expression into adverse reaction features using a message passing neural network on a bipartite graph of drugs and adverse reactions, leveraging publicly available data. By combining the computed adverse reaction features with the structural fingerprints of drugs, we predict the association between drugs and adverse reactions. Our models obtained high AUC (area under the receiver operating characteristic curve) values ranging from 0.861 to 0.907 in an external drug validation dataset under differential experiment conditions. The case study on multiple BET inhibitors also demonstrated the high accuracy of our predictions, and our model's exploration of potential adverse reactions for HWD-870 has contributed to its research and development for market approval. In summary, our method would provide a promising tool for ADR prediction and drug safety assessment in drug discovery and development.

Constructing a strongly interacting Pea-Cod binary protein system by introducing metal cations toward enhanced gelling properties.

Developing prospective plant-animal binary protein systems with desirable nutritional and rheological properties stands as a significant and challenging pursuit within the food industry. Our understanding of the effect of adding salt on the aggregation behavior of food proteins is currently based on single model protein systems, however, this knowledge is rather limited following binary protein systems. Herein, various ionic strength settings are used to mitigate the repulsive forces between pea-cod mixed proteins during the thermal process, which further benefits the construction of a strengthened gel network. Transmission electron microscopy (TEM) and dynamic light scattering (DLS) collectively demonstrated that larger heat-induced protein aggregates were formed, which increased in size with higher ionic strength. In the presence of 2.5 mM CaCl2 and 50 mM NaCl, the disulfide bonds significantly increased from 19.3 to 27.53 and 30.5 µM/g, respectively. Notably, similar aggregation behavior could be found when introducing 2.5 mM CaCl2 or 25 mM NaCl, due to the enhanced aggregation tendency by specific binding of Ca2+ to proteins. With relevance to the strengthened cross-links between protein molecules, salt endowed composite gels with preferable gelling properties, evidenced by increased storage modulus. Additionally, the gelling temperature of mixed proteins decreased below 50 °C at elevated ionic strength. Simultaneously, the proportion of network proteins in composite gels increased remarkably from 82.05 % to 93.61 % and 92.31 % upon adding 5.0 mM CaCl2 and 100 mM NaCl, respectively. The findings provide a valuable foundation for designing economically viable and health-oriented plant-animal binary protein systems.

Heat-induced agglomeration of water-soluble cod proteins toward gelled structures.

Cod proteins (CPs) have potential applications in designing desirable gel-based products, and this study aimed to unravel their heat-induced aggregation pattern and further probe the roles in protein gels. SDS-PAGE analysis indicated that high-precipitation-coefficient aggregates (HPCAs) of CPs aggregates were composed of considerable polymers of myosin heavy chains and actin, and their low-precipitation-coefficient aggregates (LPCAs) contained myosin light chains and tropomyosin. Studies from correlation analysis between the structure and aggregation kinetics revealed that the generation of ß-sheet and SS bonds were responsible for their spontaneous thermal aggregation induced by heating temperature and protein concentration, respectively. Additionally, as protein denaturation ratio increased, more and larger HPCAs were formed, which was evidenced driving the network formation of protein gels and resulting in higher storage modulus (G') values. These novel findings may be applicable to other animal proteins for better tailoring the manufacturing of muscle gel-based products.

From bonds to interactions: comprehensive molecular characterization via polarizable bond-dipole approach.

Accurately characterizing molecular interactions stands as a pivotal requirement for ensuring the reliability of molecular dynamics simulations. In line with our bond-dipole-based interaction model proposed by Gao et al. [X.-C. Gao, Q. Hao and C.-S. Wang, J. Chem. Theory Comput., 2017, 13, 2730-2741.], we have implemented an efficient and concise approach to compute electrostatic potential. This methodology capitalizes on the polarizable nature of chemical bond dipoles, resulting in a model of remarkable simplicity. In this study, we have revised the polarizable bond-dipole-based force field (PBFF) through the meticulous curation of quantum chemical data sets. These data sets encompass a comprehensive collection of 40 000 conformations, including those of water, methylamine, methanol, and N-methylacetamide. Additionally, we incorporate 520 hydrogen-bonded dimers into our data sets. In pursuit of enhanced accuracy in molecular dynamics simulations and a more faithful representation of potential energy landscapes, we undertook the re-optimization of the nonbonded parameters within the PBFF framework. Concurrently, we intricately fine-tuned the bonded parameters. The results of our comprehensive evaluation denote that this newly optimized force field method adeptly and efficiently computes structural characteristics, harmonic frequencies, and interaction energies. Overall, this study provides further validation for the applicability of PBFF in molecular dynamics simulations.

Correlation between serum lipid levels and endocrine resistance in patients with ER-positive breast cancer.

Lipid metabolism may be involved in the development of endocrine drug resistance in ER-positive (ER+) breast cancer (BC). This study aimed to investigate the relationship between serum lipid levels, risk stratification of dyslipidemia, and endocrine resistance. We collected the data from 166 ER + breast cancer patients who received endocrine therapy (ET). 73 of 166 patients (44.0%)developed endocrine resistance. Univariate and multivariate COX regression were conducted to explore the potential factors affecting endocrine resistance in BC. The clinical T stage, mean serum lipid levels in ET progression-free-survival (total cholesterol, triglycerides, low-density lipoprotein cholesterol, apolipoprotein A, and triglycerides/high-density lipoprotein cholesterol) were correlated with endocrine resistance (R = 0.214, P = .006; R = 0.268, P < .001; R = 0.182, P = .019;R = 0.197, P = .011; R = 0.211, P = .006; R = 0.159, P < .041). Clinical stage, triglycerides (TG) in endocrine therapy progression-free-survival (ePFS) and low-density lipoprotein cholesterol (LDL-C) in ePFS were independent predictors of endocrine resistance (P < .05; OR = 1.406, CI 1.108-1.783, P < .05; OR = 1.309, CI 1.026-1.669, P < .05, respectively). Moreover, in clinical stage III, the ePFS was worse in patients with in the high-risk and extremely high-risk group the median ePFS time was 8.0 months (95% CI: 1.140-14.860, P < .05). Clinical stage, TG in ePFS and LDL-C in ePFS may act as a new predictive biomarker for endocrine resistance in BC. The lipid levels of BC patients should be closely monitored throughout the treatment process, and patients with dyslipidemia should receive treatment immediately.

Synergistic Antimicrobial Hybrid Bio-Surface Formed by Self-Assembled BSA Nanoarchitectures with Chitosan Oligosaccharide.

Innovation in green, convenient, and sustainable antimicrobial packaging materials for food is an inevitable trend to address global food waste challenges caused by microbial contamination. In this study, we developed a biogenic, hydrophobic, and antimicrobial protein network coating for food packaging. Experimental results show that disulfide bond breakage can induce the self-assembly of bovine albumin (BSA) into protein networks driven by hydrophobic interactions, and chitosan oligosaccharide (COS) with antimicrobial activity can be stably bound in this network by electrostatic interactions. The inherent antimicrobial activity of COS and the numerous hydrophobic regions on the surface of the BSA-network give the BSA@COS-network significant in vitro antimicrobial ability. More importantly, the BSA@COS-network coating can prolong the onset of spoilage of strawberries in various packaging materials by nearly 3-fold in storage. This study shows how surface functionalization via protein self-assembly is integrated with the biological functioning of natural antibacterial activity for advanced food packaging applications.

Human embryonic stem cells secrete macrophage migration inhibitory factor: A novel finding.

Macrophage migration inhibitory factor (MIF) is expressed in a variety of cells and participates in important biological mechanisms. However, few studies have reported whether MIF is expressed in human Embryonic stem cells (ESCs) and its effect on human ESCs. Two human ESCs cell lines, H1 and H9 were used. The expression of MIF and its receptors CD74, CD44, CXCR2, CXCR4 and CXCR7 were detected by an immunofluorescence assay, RT-qPCR and western blotting, respectively. The autocrine level of MIF was measured via enzyme-linked immunosorbent assay. The interaction between MIF and its main receptor was investigated by co-immunoprecipitation and confocal immunofluorescence microscopy. Finally, the effect of MIF on the proliferation and survival of human ESCs was preliminarily explored by incubating cells with exogenous MIF, MIF competitive ligand CXCL12 and MIF classic inhibitor ISO-1. We reported that MIF was highly expressed in H1 and H9 human ESCs. MIF was positively expressed in the cytoplasm, cell membrane and culture medium. Several surprising results emerge. The autosecreted concentration of MIF was 22 ng/mL, which was significantly higher than 2 ng/mL-6 ng/mL in normal human serum, and this was independent of cell culture time and cell number. Human ESCs mainly expressed the MIF receptors CXCR2 and CXCR7 rather than the classical receptor CD74. The protein receptor that interacts with MIF on human embryonic stem cells is CXCR7, and no evidence of interaction with CXCR2 was found. We found no evidence that MIF supports the proliferation and survival of human embryonic stem cells. In conclusion, we first found that MIF was highly expressed in human ESCs and at the same time highly expressed in associated receptors, suggesting that MIF mainly acts in an autocrine form in human ESCs.

The effect of ferroptosis - related proteins and histone deacetylases1 on neoadjuvant chemotherapy in breast cancer.

Ferroptosis may improve the efficacy of tumor treatment, according to recent evidences. This study is to explore value of histone deacetylases 1 (HDAC1), ATP binding cassette subfamily B member 1 and ferroptosis-related proteins as potential predictive biomarkers. Eighty-two women who received neoadjuvant chemotherapy (NAC) confirmed breast cancer was included. Immunohistochemistry staining of HDAC1, ATP binding cassette subfamily B member 1 and ferroptosis-related proteins was performed in core needle biopsy and tumor resection tissue. Univariate and multivariate logistic regression were conducted to explore the potential biomarkers for breast cancer undergoing NAC. There was a weak positive correlation of HDAC1 level before and after NAC with imaging outcome (R = 0.390, P < .001). The expression of HDAC1 and glutathione peroxidase 4 before NAC was an independent predictor of imaging efficacy (OR = 7.633, CI 1.831-31.821, P < .001; OR = 0.700, CI 0.505-0.971, P < .05, respectively). HDAC1 and Glutathione peroxidase 4 may act as a new predictive biomarker for NAC in breast cancer. And personalized treatment can be provided based on them.

Soy protein particles with enhanced anti-aggregation behaviors under various heating temperatures, pH, and ionic strengths.

Protein-containing food products are frequently heated during processing to passivate anti-nutritional components. However, heating also contributes to protein aggregation and gelation, which limits its application in protein-based aqueous systems. In this study, heat-stable soy protein particles (SPPs) were fabricated by preheating at 120 °C for 30 min and at 0.5% (w/v) protein concentration. Compared to untreated soy proteins (SPs), SPPs exhibited a higher denaturation ratio, stronger conformational rigidity, compacter colloidal structure, and higher surface charge. The aggregation state of SPs and SPPs at various heating conditions (temperatures, pH, ionic strength, and types) was analyzed by dynamic light scattering, atomic force microscopy, and cryo-scanning electron microscopy. SPPs showed less increase in particle size and greater anti-aggregation ability than SPs. When heated in the presence of salt ions (Na+, Ca2+) or at acidic conditions, both SPs and SPPs formed larger spherical particles, but the size increase rate of SPPs was significantly lower than SPs. These findings provide theoretical information for preparing heat-stable SPPs. Furthermore, the development of SPPs is conducive to designing protein-enriched ingredients for producing innovative foods.

Mechanisms of heat-mediated aggregation behavior of water-soluble cod protein.

Cod proteins (CPs) are considered potential functional ingredients for developing gel-based foods, but present studies on the aggregation behavior of CPs upon heating remain limited. With this respect, the heat-induced aggregation kinetics of CPs at a subunit level was investigated. Based on different centrifugal forces, CPs aggregates were divided into three fractions: large-sized, intermediary-sized, and small-sized aggregates. SDS-PAGE and diagonal SDS-PAGE indicated that myosin heavy chains exhibited a higher affinity with actin to form intermediary-sized and large-sized aggregates; tropomyosin and myosin light chains were hardly engaged in the thermal aggregation and formed small-sized aggregates. The highly-polymerized aggregates adopted considerable transitions of helix-to-sheet in protein structures, whereas the structure of small-sized aggregates featured substantial helix-coil transitions. Furthermore, molecular interactions at different heating stages were revealed. These novel insights might advance our knowledge on the heat-induced aggregation behavior of CPs and provide fundamental information for the application of CPs in gel-based foods.

Controlling the assembly of soy ß-conglycinin to fabricate heat-stable particles for high protein liquid systems.

BACKGROUND: Recently, there is a growing interest in developing protein-fortified liquid systems, which are formulated to provide special nutrient combinations to those with special dietary needs. The fabrication of heat-stable protein for protein-fortified liquid systems relies heavily on precise control of the edible protein-building process. RESULTS: Results suggested that heat-stable 7S protein particles (7SPPs) could be obtained by preheating at 100 °C for an extended time, whereas 7S proteins with better gelling properties were discovered after preheating at lower temperatures. According to the findings of the protein conformational and morphological characterization, the 7SPPs showed rather stable tertiary and secondary structures as well as size distributions, which might be responsible for their heat stability. Additionally, during the reheating test, suspensions of 7SPPs showed no signs of gelation and had a low viscosity even though the protein content was as high as 120 mg mL-1 . However, 7S proteins with improved gelling properties were found to show rising aggregate size, higher susceptibility and larger conformational structure changing rates upon reheating treatment. CONCLUSION: Soy ß-conglycinin (7S) proteins with tunable heat stability were successfully prepared by preheating 10 mg mL-1 protein dispersions at various temperatures (80-120 °C) and durations (15-120 min). These findings provide fundamental insights for developing 7S-based protein-fortified systems. © 2023 Society of Chemical Industry.

LncAABR07053481 inhibits bone marrow mesenchymal stem cell apoptosis and promotes repair following steroid-induced avascular necrosis.

The osteonecrotic area of steroid-induced avascular necrosis of the femoral head (SANFH) is a hypoxic microenvironment that leads to apoptosis of transplanted bone marrow mesenchymal stem cells (BMSCs). However, the underlying mechanism remains unclear. Here, we explore the mechanism of hypoxic-induced apoptosis of BMSCs, and use the mechanism to improve the transplantation efficacy of BMSCs. Our results show that the long non-coding RNA AABR07053481 (LncAABR07053481) is downregulated in BMSCs and closely related to the degree of hypoxia. Overexpression of LncAABR07053481 could increase the survival rate of BMSCs. Further exploration of the downstream target gene indicates that LncAABR07053481 acts as a molecular "sponge" of miR-664-2-5p to relieve the silencing effect of miR-664-2-5p on the target gene Notch1. Importantly, the survival rate of BMSCs overexpressing LncAABR07053481 is significantly improved after transplantation, and the repair effect of BMSCs in the osteonecrotic area is also improved. This study reveal the mechanism by which LncAABR07053481 inhibits hypoxia-induced apoptosis of BMSCs by regulating the miR-664-2-5p/Notch1 pathway and its therapeutic effect on SANFH.

Machine learning to predict futile recanalization of large vessel occlusion before and after endovascular thrombectomy.

Background and purpose: Futile recanalization occurs when the endovascular thrombectomy (EVT) is a technical success but fails to achieve a favorable outcome. This study aimed to use machine learning (ML) algorithms to develop a pre-EVT model and a post-EVT model to predict the risk of futile recanalization and to provide meaningful insights to assess the prognostic factors associated with futile recanalization. Methods: Consecutive acute ischemic stroke patients with large vessel occlusion (LVO) undergoing EVT at the National Advanced Stroke Center of Nanjing First Hospital (China) between April 2017 and May 2021 were analyzed. The baseline characteristics and peri-interventional characteristics were assessed using four ML algorithms. The predictive performance was evaluated by the area under curve (AUC) of receiver operating characteristic and calibration curve. In addition, the SHapley Additive exPlanations (SHAP) approach and partial dependence plot were introduced to understand the relative importance and the influence of a single feature. Results: A total of 312 patients were included in this study. Of the four ML models that include baseline characteristics, the "Early" XGBoost had a better performance {AUC, 0.790 [95% confidence intervals (CI), 0.677-0.903]; Brier, 0.191}. Subsequent inclusion of peri-interventional characteristics into the "Early" XGBoost showed that the "Late" XGBoost performed better [AUC, 0.910 (95% CI, 0.837-0.984); Brier, 0.123]. NIHSS after 24 h, age, groin to recanalization, and the number of passages were the critical prognostic factors associated with futile recanalization, and the SHAP approach shows that NIHSS after 24 h ranks first in relative importance. Conclusions: The "Early" XGBoost and the "Late" XGBoost allowed us to predict futile recanalization before and after EVT accurately. Our study suggests that including peri-interventional characteristics may lead to superior predictive performance compared to a model based on baseline characteristics only. In addition, NIHSS after 24 h was the most important prognostic factor for futile recanalization.

[Anti-herpes simplex virus type â of tectorigenin derivative and effect on Toll-like receptors in vitro].

The study investigated the inhibitory effect and mechanism of tectorigenin derivative(SGY) against herpes simplex virus type Ⅰ(HSV-1) by in vitro experiments. The cytotoxicity of SGY and positive drug acyclovir(ACV) on African green monkey kidney(Vero) cells and mouse microglia(BV-2) cells was detected by cell counting kit-8(CCK-8) method, and the maximum non-toxic concentration and median toxic concentration(TC_(50)) of the drugs were calculated. After Vero cells were infected with HSV-1, the virulence was determined by cytopathologic effects(CPE) to calculate viral titers. The inhibitory effect of the tested drugs on HSV-1-induced cytopathy in Vero cells was measured, and their modes of action were initially explored by virus adsorption, replication and inactivation. The effects of the drugs on viral load of BV-2 cells 24 h after HSV-1 infection and the Toll-like receptor(TLR) mRNA expression were detected by real-time fluorescence quantitative PCR(RT-qPCR). The maximum non-toxic concentrations of SGY against Vero and BV-2 cells were 382.804 µg·mL~(-1) and 251.78 µg·mL~(-1), respectively, and TC_(50) was 1 749.98 µg·mL~(-1) and 2 977.50 µg·mL~(-1), respectively. In Vero cell model, the half maximal inhibitory concentration(IC_(50)) of SGY against HSV-1 was 54.49 µg·mL~(-1), and the selection index(SI) was 32.12, with the mode of action of significantly inhibiting replication and directly inactivating HSV-1. RT-qPCR results showed that SGY markedly reduced the viral load in cells. The virus model group had significantly increased relative expression of TLR2, TLR3 and tumor necrosis factor receptor-associated factor 3(TRAF3) and reduced relative expression of TLR9 as compared with normal group, and after SGY intervention, the expression of TLR2, TLR3 and TRAF3 was decreased to different degrees and that of TLR9 was enhanced. The expression of inflammatory factors inducible nitric oxide synthase(iNOS), tumor necrosis factor-α(TNF-α), and interleukin-1ß(IL-1ß) was remarkably increased in virus model group as compared with that in normal group, and the levels of these inflammatory factors dropped after SGY intervention. In conclusion, SGY significantly inhibited and directly inactivated HSV-1 in vitro. In addition, it modulated the expression of TLR2, TLR3 and TLR9 related pathways, and suppressed the increase of inflammatory factor levels.

Hypoxic condition induced H3K27me3 modification of the LncRNA Tmem235 promoter thus supporting apoptosis of BMSCs.

Bone marrow mesenchymal stem cells (BMSCs) have strong regenerative potential and show good application prospects for treating clinical diseases. However, in the process of BMSC transplantation for treating ischemic and hypoxic diseases, BMSCs have high rates of apoptosis in the hypoxic microenvironment of transplantation, which significantly affects the transplantation efficacy. Our previous studies have confirmed the key role of long non-coding RNA Tmem235 (LncRNA Tmem235) in the process of hypoxia-induced BMSC apoptosis and its downstream regulatory mechanism, but the upstream mechanism by which hypoxia regulates LncRNA Tmem235 expression to induce BMSC apoptosis is still unclear. Under hypoxic conditions, we found that the level of LncRNA Tmem235 promoter histone H3 lysine 27 trimethylation modification (H3K27me3) was significantly increased by CHIP-qPCR. Moreover, H3K27me3 cooperated with LncRNA Tmem235 promoter DNA methylation to inhibit the expression of LncRNA Tmem235 and promote apoptosis of BMSCs. To study the mechanism of hypoxia-induced modification of LncRNA Tmem235 promoter H3K27me3 in the hypoxia model of BMSCs, we detected the expression of H3K27 methylase and histone demethylase and found that only histone methylase enhancer of zeste homolog 2 (EZH2) expression was significantly upregulated. Knockdown of EZH2 significantly decreased the level of H3K27me3 modification in the LncRNA Tmem235 promoter. The EZH2 promoter region contains a hypoxia-responsive element (HRE) that interacts with hypoxia-inducible factor-1alpha (HIF-1α), which is overexpressed under hypoxic conditions, thereby promoting its overexpression. In summary, hypoxia promotes the modification of the LncRNA Tmem235 promoter H3K27me3 through the HIF-1α/EZH2 signaling axis, inhibits the expression of LncRNA Tmem235, and leads to hypoxic apoptosis of BMSCs. Our findings improve the regulatory mechanism of LncRNA Tmem235 during hypoxic apoptosis of BMSCs and provide a more complete theoretical pathway for targeting LncRNA to inhibit hypoxic apoptosis of BMSCs.

[Protective effect of Forsythiae Fructus extract on mice with herpes simplex encephalitis].

This study aims to explore the protective effect of Forsythiae Fructus extract(FFE) against herpes simplex virus encephalitis(HSE) in mice. To be specific, life extension rate of mice, viral load in mouse brain, levels of tumor necrosis factor-α(TNF-α), interleukin-1ß(IL-1ß), and interferon-α(IFN-α), and nitric oxide(NO) content in mouse brain were determined. Mice were classified into normal group, model group, acyclovir(ACV) group, and high-dose, medium-dose, and low-dose(100, 50, 25 mg·kg~(-1), respectively) FFE groups. HSE was induced in mice in corresponding groups. Then, the life extension rate was compared among groups. Viral load in brain was detected by real-time fluorescent quantitative PCR, the changes of TNF-α, IL-1ß, and IFN-α in brain by enzyme-linked immunosorbent assay(ELISA), NO content in brain with nitrate reduction method, and pathological changes by hematoxylin-eosin(HE) staining. The result showed that the life extension rate in the high-dose, medium-dose, and low-dose FFE groups was 27.93%, 19.94%, and 10.66%, respectively, and the difference between the high-dose group and the model group was statistically significant(P<0.05). FFE decreased the viral load in brains of HSE mice. The levels of TNF-α, IL-1ß, and IFN-α in ACV group and high-dose and medium-dose FFE groups were lower than those in the model group(P<0.01,P<0.05), and NO content in the three FFE groups was lower than that in the model group(P<0.01). In conclusion, FFE can improve the survival rate of HSE mice, reduce the load of herpes simplex virus type Ⅰ(HSV-1) in the brains of HSE mice, decrease the levels of inflammatory factors and NO content, and alleviate inflammation and pathological damage, thereby protecting the central nervous system.

A Dynamic Nomogram to Identify Patients at High Risk of Poor Outcome in Stroke Patients with Chronic Kidney Disease.

Background and Purpose: Predicting poor outcome for stroke patients with chronic kidney disease (CKD) in clinical practice is difficult. There are no tools available to use for predicting poor outcome in these patients. We aimed to construct and validate a dynamic nomogram to identify CKD-stroke patients at high risk of a 3-month poor outcome. Patients and Methods: We used data for 502 CKD patients who had an acute ischemic stroke, from Nanjing First Hospital, between September 2014 and September 2020, to train the nomogram. An additional 108 patients enrolled from October 2020 to May 2021 were used for temporal external validation. The performance of the nomogram was evaluated by the area under the receiver operating characteristics curve (AUC) and a calibration plot. The clinical utility of the nomogram was measured by decision curve analysis (DCA) and the clinical impact curve (CIC). Results: The median age of the cohort was 79 (70-84) years. Age, urea, premorbid modified Rankin Scale (mRS), National Institutes of Health Stroke Scale (NIHSS) on admission, hemiplegia, mechanical thrombectomy, early neurological deterioration, and respiratory infection were used as predictors of 3-month poor outcome to develop the nomogram. In the training set, the AUC of the dynamic nomogram was 0.873 and the calibration plot showed good predictive ability, and both DCA and CIC indicated the excellent clinical usefulness and applicability of the nomogram. In the external validation set, the AUC was 0.875 and the calibration plot also showed good agreement. Conclusion: This is the first dynamic nomogram constructed for CKD-stroke patients to precisely and expediently identify patients with a high risk of 3-month poor outcome. The outstanding performance and great clinical predictive utility demonstrated the ability of the dynamic nomogram to help clinicians to deploy preventive interventions.

Machine Learning-Assisted Preoperative Diagnosis of Infection Stones in Urolithiasis Patients.

Purpose: The decision-making of how to treat urinary infection stones was complicated by the difficulty in preoperative diagnosis of these stones. Hence, we developed machine learning (ML) models that can be leveraged to discriminate between infection and noninfection stones in urolithiasis patients before treatment. Materials and Methods: We enrolled 462 patients with urinary stones and randomly stratified them into training (80%) and testing sets (20%). ML models were constructed using five algorithms (decision tree, random forest classifier [RFC], extreme gradient boosting, categorical boosting, and adaptive boosting) and 15 preoperative variables and were compared with conventional logistic regression (LR) analysis. Performance measurement was the area under the receiver operating characteristic curve (AUC) in the testing set. We also analyzed the importance of 15 features on the prediction of infection stones in each ML model. Results: Sixty-two (13.4%) patients with infection stones were included in the study. On the testing set, all the five ML models demonstrated strong discrimination (AUC: 0.892-0.951). The RFC model was chosen as the final model [AUC: 0.951 (95% confidence interval, CI, 0.934-0.968); sensitivity: 0.906; specificity: 0.924], significantly outperforming the traditional LR model [AUC: 0.873 (95% CI 0.843-0.904)]. Gender, urine white blood cell counts, and urine pH level were the top 3 important features. Conclusion: Our RFC model was the first model for the preoperative identification of infection stones with superior predictive performance. This novel model could be useful for risk assessment and decision support for infection stones.

Abnormality of Functional Connections in the Resting State Brains of Schizophrenics.

To explore the change of brain connectivity in schizophrenics (SCZ), the resting-state EEG source functional connections of SCZ and healthy control (HC) were investigated in this paper. Different band single-layer networks, multilayer networks, and improved multilayer networks were constructed and their topological attributes were extracted. The topological attributes of SCZ and HC were automatically distinguished using ensemble learning methods called Ensemble Learning based on Trees and Soft voting method, and the effectiveness of different network construction methods was compared based on the classification accuracy. The results showed that the classification accuracy was 89.38% for α band network, 82.5% for multilayer network, and 86.88% for improved multilayer network. Comparing patients with SCZ to those with Alzheimer's disease (AD), the classification accuracy of improved multilayer network was the highest, which was 88.12%. The power spectrum in the α band of SCZ was significantly lower than HC, whereas there was no significant difference between SCZ and AD. This indicated that the improved multilayer network can effectively distinguish SCZ and other groups not only when their power spectrum was significantly different. The results also suggested that the improved multilayer topological attributes were regarded as biological markers in the clinical diagnosis of patients with schizophrenia and even other mental disorders.