Natural product-based drug combinations (NPDCs) present distinctive advantages in treating complex diseases. While high-throughput screening (HTS) and conventional computational methods have partially accelerated synergistic drug combination discovery, their applications remain constrained by experimental data fragmentation, high costs, and extensive combinatorial space. Recent developments in artificial intelligence (AI), encompassing traditional machine learning and deep learning algorithms, have been extensively applied in NPDC identification. Through the integration of multi-source heterogeneous data and autonomous feature extraction, prediction accuracy has markedly improved, offering a robust technical approach for novel NPDC discovery. This review comprehensively examines recent advances in AI-driven NPDC prediction, presents relevant data resources and algorithmic frameworks, and evaluates current limitations and future prospects. AI methodologies are anticipated to substantially expedite NPDC discovery and inform experimental validation.
Artificial Intelligence ; Biological Products/chemistry* ; Humans ; Drug Combinations ; Drug Discovery/methods* ; Machine Learning ; Algorithms

Increasing evidence showed that histone deacetylase 6 (HDAC6) dysfunction is directly associated with the onset and progression of various diseases, especially cancers, making the development of HDAC6-targeted anti-tumor agents a research hotspot. In this study, artificial intelligence (AI) technology and molecular simulation strategies were fully integrated to construct an efficient and precise drug screening pipeline, which combined Voting strategy based on compound-protein interaction (CPI) prediction models, cascade molecular docking, and molecular dynamic (MD) simulations. The biological potential of the screened compounds was further evaluated through enzymatic and cellular activity assays. Among the identified compounds, Cmpd.18 exhibited more potent HDAC6 enzyme inhibitory activity (IC₅₀ = 5.41 nM) than that of tubastatin A (TubA) (IC₅₀ = 15.11 nM), along with a favorable subtype selectivity profile (selectivity index ≈ 117.23 for HDAC1), which was further verified by the Western blot analysis. Additionally, Cmpd.18 induced G2/M phase arrest and promoted apoptosis in HCT-116 cells, exerting desirable antiproliferative activity (IC₅₀ = 2.59 μM). Furthermore, based on long-term MD simulation trajectory, the key residues facilitating Cmpd.18's binding were identified by decomposition free energy analysis, thereby elucidating its binding mechanism. Moreover, the representative conformation analysis also indicated that Cmpd.18 could stably bind to the active pocket in an effective conformation, thus demonstrating the potential for in-depth research of the 2-(2-phenoxyethyl)pyridazin-3(2H)-one scaffold.

Purpose: Explore the causal relationship between the gut microbiota and erectile dysfunction (ED) at phylum, class, order, family, and genus levels, and identify specific pathogenic bacteria that may be associated with the onset and progression of ED. Materials and Methods: The genetic variation data of 196 human gut microbiota incorporated in our study came from the human gut microbiome Genome Wide Association Studies (GWAS) dataset released by the MiBioGen Consortium. The GWAS statistics for ED were extracted from one study by Bovijn et al., which included 223,805 participants of European ancestry, of whom 6,175 were diagnosed with ED. Subsequently, Mendelian randomization (MR) analysis was carried out to explore whether a causal relationship exists between the gut microbiota and ED. Additionally, bidirectional MR analysis was performed to examine the directionality of the causal relationship. Results: Through MR analysis, we found that family Lachnospiraceae (odds ratio [OR]: 1.27, 95% confidence interval [CI]: 1.05–1.52, p=0.01) and its subclass genus LachnospiraceaeNC2004 group (OR: 1.17, 95% CI: 1.01–1.37, p=0.04) are associated with a higher risk of ED. In addition, genus Oscillibacter (OR: 1.17, 95% CI: 1.02–1.35, p=0.03), genus Senegalimassilia (OR: 1.32, 95% CI: 1.06–1.64, p=0.01) and genus Tyzzerella3 (OR: 1.14, 95% CI: 1.02–1.27, p=0.02) also increase the risk of ED. In contrast, the inverse variance weighted estimate of genus RuminococcaceaeUCG013 (OR: 0.77, 95% CI: 0.61–0.96, p=0.02) suggests that it has a protective effect against the occurrence of ED. Conclusions This study preliminarily identified 6 bacterial taxa that may have a causal relationship with ED, including family Lachnospiraceae, genus Lachnospiraceae NC2004 group, Oscillibacter, Senegalimassilia, Tyzzerella 3 and Ruminococcaceae UCG013. These identified important bacterial taxa may serve as candidates for microbiome intervention in future ED clinical trials.

Objective:To investigate the predictive value for major adverse cardiovascular events (MACE) occurring within 1 year in patients with old myocardial infarction(OMI) using characteristics of myocardial scar derive from dual-energy CT (DECT) post-processing technique.Methods:OMI patients who received coronary CT angiography following dual-energy CT with late iodine enhancement (LIE-DECT) in the Second Affiliated Hospital of Nantong University from November 2019 to October 2022 were continuously included, and the images of all enrolled patients were reconstructed using 40 keV monoenergetic plus (Mono+) map, LIE (representing myocardial scar) was quantified on left ventricular short-axis images, including the LIE segments, the LIE score, and the LIE degree. All enrolled patients were followed up for MACE, defined as hospitalization for heart failure, malignant arrhythmia, and cardiac death. Regression analysis was used to investigate the relationship between the quantified value of myocardial scar and the occurrence of MACE, and receiver operating characteristic curve (ROC) was used to evaluate the efficacy of quantified value of myocardial scar in predicting MACE. The area under the curve (AUC) was compared using the DeLong test.Results:Finally, 231 patients with OMI were included, and MACE occurred in 37 cases (16.0%) within 1 year after LIE-DECT examination. The LIE segments 5 (4, 7), the LIE score 27 (13, 49) and the LIE degree 9.4%(7.5%, 15.5%) in the MACE group were higher than those in the non-MACE group 3 (2, 5), 9 (6, 15) and 6.7%(6.3%, 7.9%) (all P<0.001). Multivariable logistic regression analysis showed that after adjusting for confounders, the LIE segments ( OR=2.118, P<0.001), the LIE score ( OR=3.168, P<0.001), and the LIE degree ( OR=3.092, P<0.001) remained risk factors for the development of MACE. On ROC analysis, AUC of LIE segments, LIE score and LIE degree were 0.715, 0.822 and 0.806 (all P<0.001), with sensitivities of 81.1%, 86.5%, and 91.9%, and specificities of 53.6%, 69.6%, and 60.8%, respectively. DeLong′s test showed that the predictive efficacy of LIE score and LIE degree was higher than that of LIE segments ( Z=2.63, P=0.008; Z=1.96, P=0.049), and there were no significant differences in the predictive efficacy of LIE score and LIE degree ( Z=0.60, P=0.551). Conclusion:The LIE segments, the LIE score and the LIE degree detected by LIE-DECT 40 keV Mono+maps are risk factors for the occurrence of MACE in patients with OMI and have good efficacy in predicting the occurrence of MACE, which can be used as important indicators for assessing the clinical prognosis of OMI.

Objective:To compare the image quality, radiation dose, and total iodine content of abdominal computed tomography (CT) enhancement scanning of overweight and obese patients with different scanning protocols, and to explore the optimal keV image serial for abdominal CT.Methods:A total of 90 overweight or obese patients [24 kg/m 2 ≤ body mass index (BMI) < 28 kg/m 2 or BMI ≥ 28 kg/m 2] were divied into groups A, B and C, with 30 patients in each group. Group A used Gemstone spectral imaging (GSI) mode and contrast medium with 320 mg I/ml, group B used low tube voltage mode (100 kVp) and contrast medium with 370 mg I/ml, and group C used conventional tube voltage mode (120 kVp) and contrast medium with 370 mg I/ml. Monochromatic energy images at 50-70 keV (5 keV interval) were reconstructed for the arterial and portal vein phases of group A. Radiation dose and total iodine content were recorded and calculated for the 3 groups. The region of interest was placed on the organ, blood vessel, and erector spinae muscle at same level. The CT values and image noise values were measured, and the contrast-to-noise ratio (CNR) was calculated. All images were scored subjectively in double-blinded by two radiologists. One-way analysis of variance or Kruskal-Wallis H test were used to compare The CT values, CNRs, and subjective scores of each subgroup image in group A, group B and group C, and the radiation doses and total iodine contents in 3 groups were compared. The optimal keV value for group A was selected. Results:At 50-60 keV, the CT values and CNRs of arterial and portal vein phases in group A were higher ( P < 0.05) than or similar to those in groups B and C ( P > 0.05), and the subjective scores were lower than those of groups B and C at 50 and 55 keV ( H = 34.47, 41.27, P < 0.05), whereas there was no statistically significant difference at 60 keV ( P > 0.05). At 65 and 70 keV, only the CT value and CNR of the renal cortex in group A at the 65 keV of arterial phase were higher than those in groups B and C ( F = 102.38, 29.47, P < 0.001). The subjective scores were not significantly between groups B and C ( P > 0.05). There were no statistically significant difference between CT values, CNRs, or subjective scores in group B and group C ( P > 0.05). The effective doses in groups A and B were 24.72% and 25.78% lower than those in group C, respectively. Compared to groups B and C, the total iodine content in group A decreased by 12.50% and 13.34%, respectively. Conclusions:GSI model combined with a low-concentration contrast medium in abdominal CT for overweight and obese patients can meet the image quality requirements while reducing patient total iodine content and radiation dose. The optimal keV value of enhanced abdominal CT for double phases was 60 keV.

Colorectal cancer(CRC)is highly heterogeneous,but traditional TNM staging cannot distinguish the heterogeneity of CRC well,which can no longer meet the treatment needs.Integrating the clinical features,molecular genetic changes in cancer tissue,transcriptome and proteome changes,as well as immune matrix characteristics,the consensus molecular subtype(CMS)of CRC is by far the best description of its heterogeneity.This paper first discusses the molecular genetic changes of three types of CRC cancer tissues(chromosomal instability,microsatellite instability,and CpG island methylation phenotype).Then it systematically elaborates on the clinical characteristics,treatment directions,and prognosis evaluation of CRC patients with different CMS subtypes,as well as their relationship with immunotherapy and changes in gut microbiota.With the continuous improvement of sequencing technology and the prospective precision medicine clinical trial exploration,the"multi-molecule multi-drug"treatment model based on CMS typing will become the core of future precision medicine and personalized medicine.

Objective To establish a body composition analysis system based on chest CT,and to observe its value for evaluating content of chest muscle and adipose.Methods T7-T8 layer CT images of 108 pneumonia patients were collected(segmented dataset),and chest CT data of 984 patients were screened from the COVID 19-CT dataset(10 cases were randomly selected as whole test dataset,the remaining 974 cases were selected as layer selection dataset).T7-T8 layer was classified based on convolutional neural network(CNN)derived networks,including ResNet,ResNeXt,MobileNet,ShuffleNet,DenseNet,EfficientNet and ConvNeXt,then the accuracy,precision,recall and specificity were used to evaluate the performance of layer selection dataset.The skeletal muscle(SM),subcutaneous adipose tissue(SAT),intermuscular adipose tissue(IMAT)and visceral adipose tissue(VAT)were segmented using classical fully CNN(FCN)derived network,including FCN,SegNet,UNet,Attention UNet,UNET++,nnUNet,UNeXt and CMUNeXt,then Dice similarity coefficient(DSC),intersection over union(IoU)and 95 Hausdorff distance(HD)were used to evaluate the performance of segmented dataset.The automatic body composition analysis system was constructed based on optimal layer selection network and segmentation network,the mean absolute error(MAE),root mean squared error(RMSE)and standard deviation(SD)of MAE were used to evaluate the performance of automatic system for testing the whole test dataset.Results The accuracy,precision,recall and specificity of DenseNet network for automatically classifying T7-T8 layer from chest CT images was 95.06％,84.83％,92.27％and 95.78％,respectively,which were all higher than those of the other layer selection networks.In segmentation of SM,SAT,IMAT and overall,DSC and IoU of UNet++network were all higher,while 95HD of UNet++network were all lower than those of the other segmentation networks.Using DenseNet as the layer selection network and UNet++as the segmentation network,MAE of the automatic body composition analysis system for predicting SM,SAT,IMAT,VAT and MAE was 27.09,6.95,6.65 and 3.35 cm2,respectively.Conclusion The body composition analysis system based on chest CT could be used to assess content of chest muscle and adipose.Among them,the UNet++network had better segmentation performance in adipose tissue than SM.

Purpose: Explore the causal relationship between the gut microbiota and erectile dysfunction (ED) at phylum, class, order, family, and genus levels, and identify specific pathogenic bacteria that may be associated with the onset and progression of ED. Materials and Methods: The genetic variation data of 196 human gut microbiota incorporated in our study came from the human gut microbiome Genome Wide Association Studies (GWAS) dataset released by the MiBioGen Consortium. The GWAS statistics for ED were extracted from one study by Bovijn et al., which included 223,805 participants of European ancestry, of whom 6,175 were diagnosed with ED. Subsequently, Mendelian randomization (MR) analysis was carried out to explore whether a causal relationship exists between the gut microbiota and ED. Additionally, bidirectional MR analysis was performed to examine the directionality of the causal relationship. Results: Through MR analysis, we found that family Lachnospiraceae (odds ratio [OR]: 1.27, 95% confidence interval [CI]: 1.05–1.52, p=0.01) and its subclass genus LachnospiraceaeNC2004 group (OR: 1.17, 95% CI: 1.01–1.37, p=0.04) are associated with a higher risk of ED. In addition, genus Oscillibacter (OR: 1.17, 95% CI: 1.02–1.35, p=0.03), genus Senegalimassilia (OR: 1.32, 95% CI: 1.06–1.64, p=0.01) and genus Tyzzerella3 (OR: 1.14, 95% CI: 1.02–1.27, p=0.02) also increase the risk of ED. In contrast, the inverse variance weighted estimate of genus RuminococcaceaeUCG013 (OR: 0.77, 95% CI: 0.61–0.96, p=0.02) suggests that it has a protective effect against the occurrence of ED. Conclusions This study preliminarily identified 6 bacterial taxa that may have a causal relationship with ED, including family Lachnospiraceae, genus Lachnospiraceae NC2004 group, Oscillibacter, Senegalimassilia, Tyzzerella 3 and Ruminococcaceae UCG013. These identified important bacterial taxa may serve as candidates for microbiome intervention in future ED clinical trials.

Purpose: Explore the causal relationship between the gut microbiota and erectile dysfunction (ED) at phylum, class, order, family, and genus levels, and identify specific pathogenic bacteria that may be associated with the onset and progression of ED. Materials and Methods: The genetic variation data of 196 human gut microbiota incorporated in our study came from the human gut microbiome Genome Wide Association Studies (GWAS) dataset released by the MiBioGen Consortium. The GWAS statistics for ED were extracted from one study by Bovijn et al., which included 223,805 participants of European ancestry, of whom 6,175 were diagnosed with ED. Subsequently, Mendelian randomization (MR) analysis was carried out to explore whether a causal relationship exists between the gut microbiota and ED. Additionally, bidirectional MR analysis was performed to examine the directionality of the causal relationship. Results: Through MR analysis, we found that family Lachnospiraceae (odds ratio [OR]: 1.27, 95% confidence interval [CI]: 1.05–1.52, p=0.01) and its subclass genus LachnospiraceaeNC2004 group (OR: 1.17, 95% CI: 1.01–1.37, p=0.04) are associated with a higher risk of ED. In addition, genus Oscillibacter (OR: 1.17, 95% CI: 1.02–1.35, p=0.03), genus Senegalimassilia (OR: 1.32, 95% CI: 1.06–1.64, p=0.01) and genus Tyzzerella3 (OR: 1.14, 95% CI: 1.02–1.27, p=0.02) also increase the risk of ED. In contrast, the inverse variance weighted estimate of genus RuminococcaceaeUCG013 (OR: 0.77, 95% CI: 0.61–0.96, p=0.02) suggests that it has a protective effect against the occurrence of ED. Conclusions This study preliminarily identified 6 bacterial taxa that may have a causal relationship with ED, including family Lachnospiraceae, genus Lachnospiraceae NC2004 group, Oscillibacter, Senegalimassilia, Tyzzerella 3 and Ruminococcaceae UCG013. These identified important bacterial taxa may serve as candidates for microbiome intervention in future ED clinical trials.

Objective This research aims to investigate the impact of Orff music therapy on long-term schizophrenic patients in hospitals.Methods The study was a randomized,single-blind controlled trial conducted from April,2023 to September,2023.From April to September 2023,sixty-eight individuals diagnosed with persistent schizophrenia were enrolled and evenly distributed into a pair of cohorts:a treatment group of thirty-four people receiving the intervention,and an equal number forming the control group for comparative purposes.Individuals enrolled in the experimental arm of the study were administered Orff-music therapy alongside routine rehabilitation treatment across a span of two months.For a period of 8 weeks,the control group was given only standard rehabilitation treatment,whereas the research group underwent Orff-music therapy in addition to the standard rehabilitation treatment.Results Before treatment,there were no significant differences in the positive and negative symptoms scale(PANSS),the inpatient psychosis rehabilitation observe scale(IPROS)and the personal and social performance scale(psp)between two groups.After intervention,the PANSS showed that the changes were better in the study group than in the control group in 3 indicators:negative symptoms(-3.20±4.13 vs.-0.17±2.43,P<0.001),general symptoms(-2.79±3.83 vs.-0.17±2.99,P=0.003)and the total scores(-5.88±6.36 vs.0.00±4.08,P<0.001),but not in positive symptoms(P>0.05).The IPROS showed that the performances of patients in the study group were better than the control group in terms of participation in work therapy(-0.82±2.08 vs.0.23±2.10,P=0.041),socialization(-0.59±1.94 vs.0.53±1.69,P=0.014)and ability to live(-0.94±2.50 vs.0.15±1.48,P=0.033),the changes in scores before and after the intervention were significantly different when compared to the control group.The PSP showed that the changes in scores before and after the treatment of the study group was better than the control group in terms of social activity[0(-1,0)vs.0(0,0),P=0.011],and self-care[0(-1,0)vs.0(-0.25,0),P=0.012]were better than the control group.Conclusion For long-term hospitalized patients with chronic schizophrenia.Orff-music therapy can be a powerful tool for alleviating mental issues,fostering social functioning,and enhancing rehabilitation results.