Accurate prediction of drug-target interactions (DTIs) plays a pivotal role in drug discovery, facilitating optimization of lead compounds, drug repurposing and elucidation of drug side effects. However, traditional DTI prediction methods are often limited by incomplete biological data and insufficient representation of protein features. In this study, we proposed KG-CNNDTI, a novel knowledge graph-enhanced framework for DTI prediction, which integrates heterogeneous biological information to improve model generalizability and predictive performance. The proposed model utilized protein embeddings derived from a biomedical knowledge graph via the Node2Vec algorithm, which were further enriched with contextualized sequence representations obtained from ProteinBERT. For compound representation, multiple molecular fingerprint schemes alongside the Uni-Mol pre-trained model were evaluated. The fused representations served as inputs to both classical machine learning models and a convolutional neural network-based predictor. Experimental evaluations across benchmark datasets demonstrated that KG-CNNDTI achieved superior performance compared to state-of-the-art methods, particularly in terms of Precision, Recall, F1-Score and area under the precision-recall curve (AUPR). Ablation analysis highlighted the substantial contribution of knowledge graph-derived features. Moreover, KG-CNNDTI was employed for virtual screening of natural products against Alzheimer's disease, resulting in 40 candidate compounds. 5 were supported by literature evidence, among which 3 were further validated in vitro assays.
Alzheimer Disease/drug therapy* ; Biological Products/therapeutic use* ; Humans ; Neural Networks, Computer ; Machine Learning ; Drug Discovery/methods* ; Algorithms ; Drug Evaluation, Preclinical/methods*

Body weight abnormalities, including overweight, obesity, and underweight, have become a dual public health challenge in Chinese adults: overweight and obesity lead to a variety of chronic complications, while underweight increases the risks of malnutrition, sarcopenia, and organ dysfunction. To systematically address these issues, multidisciplinary experts in endocrinology, sports science, nutrition, and psychiatry from various regions have held multiple weight management seminars. Based on the latest epidemiological data and clinical evidence, they expanded the guideline to include assessment and intervention strategies for underweight, in addition to the core content of obesity management. This guideline outlines the etiological mechanisms, evaluation methods, and multidimensional management strategies for overweight and obesity, covering key areas such as diagnosis and assessment, medical nutrition therapy, exercise prescription, pharmacological intervention, and psychological support. It is intended to provide a scientific and standardized approach to weight management across the adult population, aiming to curb the rising prevalence of obesity, mitigate complications associated with abnormal body weight, and improve nutritional status and overall quality of life.

Activity cliffs(ACs)are generally defined as pairs of similar compounds that only differ by a minor structural modification but exhibit a large difference in their binding affinity for a given target.ACs offer crucial insights that aid medicinal chemists in optimizing molecular structures.Nonetheless,they also form a major source of prediction error in structure-activity relationship(SAR)models.To date,several studies have demonstrated that deep neural networks based on molecular images or graphs might need to be improved further in predicting the potency of ACs.In this paper,we integrated the triplet loss in face recognition with pre-training strategy to develop a prediction model ACtriplet,tailored for ACs.Through extensive comparison with multiple baseline models on 30 benchmark datasets,the results showed that ACtriplet was significantly better than those deep learning(DL)models without pre-training.In addition,we explored the effect of pre-training on data representation.Finally,the case study demonstrated that our model's interpretability module could explain the prediction results reasonably.In the dilemma that the amount of data could not be increased rapidly,this innovative framework would better make use of the existing data,which would propel the potential of DL in the early stage of drug discovery and optimization.

Negative logarithm of the acid dissociation constant(pKa)significantly influences the absorption,dis-tribution,metabolism,excretion,and toxicity(ADMET)properties of molecules and is a crucial indicator in drug research.Given the rapid and accurate characteristics of computational methods,their role in predicting drug properties is increasingly important.Although many pKa prediction models currently exist,they often focus on enhancing model precision while neglecting interpretability.In this study,we present GraFpKa,a pKa prediction model using graph neural networks(GNNs)and molecular finger-prints.The results show that our acidic and basic models achieved mean absolute errors(MAEs)of 0.621 and 0.402,respectively,on the test set,demonstrating good predictive performance.Notably,to improve interpretability,GraFpKa also incorporates Integrated Gradients(IGs),providing a clearer visual description of the atoms significantly affecting the pKa values.The high reliability and interpretability of GraFpKa ensure accurate pKa predictions while also facilitating a deeper understanding of the relation-ship between molecular structure and pKa values,making it a valuable tool in the field of pKa prediction.

This study aimed to comprehensively examine the association of gallstones, cholecystectomy, and cancer risk. Multivariable logistic regressions were performed to estimate the observational associations of gallstones and cholecystectomy with cancer risk, using data from a nationwide cohort involving 239 799 participants. General and gender-specific two-sample Mendelian randomization (MR) analysis was further conducted to assess the causalities of the observed associations. Observationally, a history of gallstones without cholecystectomy was associated with a high risk of stomach cancer (adjusted odds ratio (aOR)=2.54, 95% confidence interval (CI) 1.50-4.28), liver and bile duct cancer (aOR=2.46, 95% CI 1.17-5.16), kidney cancer (aOR=2.04, 95% CI 1.05-3.94), and bladder cancer (aOR=2.23, 95% CI 1.01-5.13) in the general population, as well as cervical cancer (aOR=1.69, 95% CI 1.12-2.56) in women. Moreover, cholecystectomy was associated with high odds of stomach cancer (aOR=2.41, 95% CI 1.29-4.49), colorectal cancer (aOR=1.83, 95% CI 1.18-2.85), and cancer of liver and bile duct (aOR=2.58, 95% CI 1.11-6.02). MR analysis only supported the causal effect of gallstones on stomach, liver and bile duct, kidney, and bladder cancer. This study added evidence to the causal effect of gallstones on stomach, liver and bile duct, kidney, and bladder cancer, highlighting the importance of cancer screening in individuals with gallstones.
Humans ; Mendelian Randomization Analysis ; Gallstones/complications* ; Female ; Male ; Cholecystectomy/statistics & numerical data* ; Middle Aged ; Risk Factors ; Aged ; Adult ; Neoplasms/etiology* ; Stomach Neoplasms/epidemiology*

Negative logarithm of the acid dissociation constant (pK _a) significantly influences the absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of molecules and is a crucial indicator in drug research. Given the rapid and accurate characteristics of computational methods, their role in predicting drug properties is increasingly important. Although many pK _a prediction models currently exist, they often focus on enhancing model precision while neglecting interpretability. In this study, we present GraFpK _a, a pK _a prediction model using graph neural networks (GNNs) and molecular fingerprints. The results show that our acidic and basic models achieved mean absolute errors (MAEs) of 0.621 and 0.402, respectively, on the test set, demonstrating good predictive performance. Notably, to improve interpretability, GraFpK _a also incorporates Integrated Gradients (IGs), providing a clearer visual description of the atoms significantly affecting the pK _a values. The high reliability and interpretability of GraFpK _a ensure accurate pK _a predictions while also facilitating a deeper understanding of the relationship between molecular structure and pK _a values, making it a valuable tool in the field of pK _a prediction.

Activity cliffs (ACs) are generally defined as pairs of similar compounds that only differ by a minor structural modification but exhibit a large difference in their binding affinity for a given target. ACs offer crucial insights that aid medicinal chemists in optimizing molecular structures. Nonetheless, they also form a major source of prediction error in structure-activity relationship (SAR) models. To date, several studies have demonstrated that deep neural networks based on molecular images or graphs might need to be improved further in predicting the potency of ACs. In this paper, we integrated the triplet loss in face recognition with pre-training strategy to develop a prediction model ACtriplet, tailored for ACs. Through extensive comparison with multiple baseline models on 30 benchmark datasets, the results showed that ACtriplet was significantly better than those deep learning (DL) models without pre-training. In addition, we explored the effect of pre-training on data representation. Finally, the case study demonstrated that our model's interpretability module could explain the prediction results reasonably. In the dilemma that the amount of data could not be increased rapidly, this innovative framework would better make use of the existing data, which would propel the potential of DL in the early stage of drug discovery and optimization.

ObjectivesTo investigate the molecular epidemiological characteristics of HIV-1 infection among men who had sex with men (MSM) in Taizhou City, Zhejiang Province, and to provide a scientific reference for acquired immune deficiency syndrome prevention and control efforts. MethodsThe research subjects were all newly reported MSM population in Taizhou City from 2020 to 2023. Blood samples without antiviral therapy were collected. The HIV-1 pol gene was amplified and sequenced, and the sequences were submitted to the Stanford University drug resistance database to identify the mutation sites and drug resistance. MEGA 11.0 software was used to analyze the nucleic acid sequences, construct phylogenetic tree, and calculate genetic distance of gene sequences. The molecular transmission network diagram of HIV-1 was constructed using Cytoscape_v3.10.1, and the influencing factors of network entry were analyzed by logistic regression. ResultsA total of 363 newly reported HIV-infected MSM patients were included, with a median age [M (P₂₅, P₇₅)] of 34 (26,47) years old. The majority had an educational level of junior high school or below (55.65%). A total of eight subtypes were found, mainly CRF07_BC and CRF01_AE. The primary drug resistance rate was 10.47% (38/363). The optimal molecular network gene distance was 0.019, with a network access rate of 42.70% (155/363), and a total of 47 molecular clusters were formed. Multivariate logistic analyses showed that compared with the CRF01_AE subtype, the clustering risk of CRF07_BC subtype was higher (OR=1.916, 95%CI: 1.191‒3.109), cases with drug resistance had a higher risk of cluster formation than those without drug resistance (OR=2.011, 95%CI: 1.006‒4.080), and recent infected patients had a lower risk of entering the largest molecular cluster than long-term infected patients (OR=0.376, 95%CI: 0.137‒0.928). ConclusionThe newly diagnosed infections among the MSM population are active in Taizhou City, Zhejiang Province, with a high level of primary drug resistance. Individuals carrying drug-resistant strains are more likely to cluster. Drug resistance monitoring should be strengthened to prevent further spread of drug-resistant strains in the network.

Objective To investigate the status and influencing factors of type 2 diabetes mellitus(T2DM)patients' fear of complications,and to provide a reference for formulating targeted intervention measures.Methods From April to November 2024,370 patients with T2DM in 2 tertiary general hospitals in Daqing City were selected by convenience sampling method.General data questionnaire,Fear of Complications Questionnaire,Self-Perceived Burden Scale,Psychological Capital Questionnaire,Mishel Uncertainty in Illness Scale and Family Apgar Index Questionnaire were used for investigation.Univariate analysis and binary Logistic regression were performed to analyze the influencing factors.Results A total of 364 valid questionnaires were collected,with an effective recovery rate of 98.38％.The score of Fear of Complications Questionnaire was(23.47±7.47),and the incidence of fear of complications was 22.25％.Logistic regression analysis showed that medical payment methods,the number of complications,positive psychological capital and family care were the influencing factors of FoC in T2DM patients.Conclusion The fear of complications in T2DM patients is at a moderate level.Nursing staff should pay attention to the early assessment of patients' fear of complications,promptly identify and take effective measures to reduce the level of patients' fear of complications,improve their quality of life.

Objective:To investigate the efficacy and safety of allopurinol in stable coronary heart disease patients with asymptomatic hyperuricemia (AH).Methods:Sixty stable coronary heart disease with AH patients admitted to the Changsha Third Hospital from January 2022 to December 2023 were selected. Patients were randomly divided into an observation group (allopurinol treatment group) and a control group (placebo group). The results of tablet exercise tests, treatment efficacy, blood uric acid levels, liver and kidney function indicators, and incidence of adverse events were compare and analyzed between two groups of patients.Results:The observation group had better exercise termination time, maximum ST descent time, and ST recovery time than the control group (all P<0.05). The total effective rate of the observation group after 6 weeks of treatment was significantly higher than that of the control group, and the difference was statistically significant (χ 2=5.455, P=0.02). There was no statistically significant difference in blood uric acid levels and liver and kidney function indicators between the two groups before treatment (all P>0.05). After treatment, both groups showed significant improvement in blood uric acid levels and liver and kidney function indicators compared to before treatment (all P<0.05). The levels of blood uric acid and liver and kidney function indicators in the observation group were significantly better than those in the control group after treatment (all P<0.05). The incidence of adverse events in the observation group during treatment was lower than that in the control group (χ 2=5.192, P=0.023). Conclusions:Allopurinol has a certain therapeutic effect on stable coronary heart disease with asymptomatic hyperuricemia, which helps patients enhance their physical activity and reduce the incidence of cardiovascular events.