1.Identify drug-drug interactions via deep learning: A real world study.
Jingyang LI ; Yanpeng ZHAO ; Zhenting WANG ; Chunyue LEI ; Lianlian WU ; Yixin ZHANG ; Song HE ; Xiaochen BO ; Jian XIAO
Journal of Pharmaceutical Analysis 2025;15(6):101194-101194
Identifying drug-drug interactions (DDIs) is essential to prevent adverse effects from polypharmacy. Although deep learning has advanced DDI identification, the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits. Here, we developed a Multi-Dimensional Feature Fusion model named MDFF, which integrates one-dimensional simplified molecular input line entry system sequence features, two-dimensional molecular graph features, and three-dimensional geometric features to enhance drug representations for predicting DDIs. MDFF was trained and validated on two DDI datasets, evaluated across three distinct scenarios, and compared with advanced DDI prediction models using accuracy, precision, recall, area under the curve, and F1 score metrics. MDFF achieved state-of-the-art performance across all metrics. Ablation experiments showed that integrating multi-dimensional drug features yielded the best results. More importantly, we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs. Among 12 real-world adverse drug reaction reports, the predictions of 9 reports were supported by relevant evidence. Additionally, MDFF demonstrated the ability to explain adverse DDI mechanisms, providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.
2.Prediction of immunotherapy targets for chronic cerebral hypoperfusion by bioinformatics method.
Mei ZHAO ; Yanpeng XUE ; Qingqing TIAN ; He YANG ; Qing JIANG ; Mengfan YU ; Xin CHEN
Journal of Biomedical Engineering 2025;42(2):382-388
Chronic cerebral hypoperfusion (CCH) plays an important role in the occurrence and development of vascular dementia (VD). Recent studies have indicated that multiple stages of immune-inflammatory response are involved in the process of cerebral ischemia, drawing increasing attention to immune therapies for cerebral ischemia. This study aims to identify potential immune therapeutic targets for CCH using bioinformatics methods from an immunological perspective. We identified a total of 823 differentially expressed genes associated with CCH, and further screened for 9 core immune-related genes, namely RASGRP1, FGF12, SEMA7A, PAK6, EDN3, BPHL, FCGRT, HSPA1B and MLNR. Gene enrichment analysis showed that core genes were mainly involved in biological functions such as cell growth, neural projection extension, and mesenchymal stem cell migration. Biological signaling pathway analysis indicated that core genes were mainly involved in the regulation of T cell receptor, Ras and MAPK signaling pathways. Through LASSO regression, we identified RASGRP1 and BPHL as key immune-related core genes. Additionally, by integrating differential miRNAs and the miRwalk database, we identified miR-216b-5p as a key immune-related miRNA that regulates RASGRP1. In summary, the predicted miR-216b-5p/ RASGRP1 signaling pathway plays a significant role in immune regulation during CCH, which may provide new targets for immune therapy in CCH.
Humans
;
Computational Biology/methods*
;
Brain Ischemia/therapy*
;
Immunotherapy
;
MicroRNAs/genetics*
;
Signal Transduction
;
Dementia, Vascular/genetics*
;
Chronic Disease
3.The impact of two-stage Turnbull-Cutait pull-through coloanal anastomosis on anal function and surgical safety in the treatment of low rectal cancer
Hanxiao ZHAO ; Yuchen GUO ; Liang HE ; Luyao ZHANG ; Jia'nan SUN ; Xuan SUN ; Yinquan ZHAO ; Yanpeng XING ; Yanjun WANG ; Meiling WANG ; Yang GONG ; Quan WANG
Chinese Journal of General Surgery 2025;40(2):101-107
Objective:To explore the anal function and postoperative complications of 2-stage Turnbull-Cutait pull-through coloanal anastomosis (TCA) for low rectal cancer.Methods:Patients undergoing radical rectal cancer resection from Feb 2023 to Nov 2024 in the First Hospital of Jilin University were divided into the TCA surgery group and the low anterior resection combined with prophylactic stoma (LAR) surgery group.Results:Among the 102 patients, there were 50 cases in the TCA group and 52 cases in the LAR group. In the single-arm analysis of the TCA group, the overall complication rate was 44%. The incidence rates of severe LARS at 1 month, 3 months, and 6 months after surgery were 97%, 77%, and 64% respectively. There was no significant difference in the complication rate within 30 days after surgery between the two groups,(44% vs. 38%, χ2=0.135, P>0.05). There was no significant difference in the incidence rate of severe LARS between the TCA group and the LAR group (77% vs. 69%, χ2=0.202, P>0.05), and there was not significant difference in the incidence rate of severe LARS between the two groups at the 6th month after surgery,(64% vs. 48%, χ2=1.132, P>0.05). Conclusion:In patients who underwent TCA surgery, the LARS symptoms gradually decreased over time. Compared with patients undergoing low anterior resection and stoma reversal, there were no significant differences in complications within 30 days after surgery and LARS symptoms within half a year.
4.Identify drug-drug interactions via deep learning:A real world study
Jingyang LI ; Yanpeng ZHAO ; Zhenting WANG ; Chunyue LEI ; Lianlian WU ; Yixin ZHANG ; Song HE ; Xiaochen BO ; Jian XIAO
Journal of Pharmaceutical Analysis 2025;15(6):1249-1263
Identifying drug-drug interactions(DDIs)is essential to prevent adverse effects from polypharmacy.Although deep learning has advanced DDI identification,the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits.Here,we developed a Multi-Dimensional Feature Fusion model named MDFF,which integrates one-dimensional simplified molec-ular input line entry system sequence features,two-dimensional molecular graph features,and three-dimensional geometric features to enhance drug representations for predicting DDIs.MDFF was trained and validated on two DDI datasets,evaluated across three distinct scenarios,and compared with advanced DDI prediction models using accuracy,precision,recall,area under the curve,and F1 score metrics.MDFF achieved state-of-the-art performance across all metrics.Ablation experiments showed that integrating multi-dimensional drug features yielded the best results.More importantly,we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs.Among 12 real-world adverse drug reaction reports,the predictions of 9 reports were supported by relevant evidence.Additionally,MDFF demon-strated the ability to explain adverse DDI mechanisms,providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.
5.The impact of two-stage Turnbull-Cutait pull-through coloanal anastomosis on anal function and surgical safety in the treatment of low rectal cancer
Hanxiao ZHAO ; Yuchen GUO ; Liang HE ; Luyao ZHANG ; Jia'nan SUN ; Xuan SUN ; Yinquan ZHAO ; Yanpeng XING ; Yanjun WANG ; Meiling WANG ; Yang GONG ; Quan WANG
Chinese Journal of General Surgery 2025;40(2):101-107
Objective:To explore the anal function and postoperative complications of 2-stage Turnbull-Cutait pull-through coloanal anastomosis (TCA) for low rectal cancer.Methods:Patients undergoing radical rectal cancer resection from Feb 2023 to Nov 2024 in the First Hospital of Jilin University were divided into the TCA surgery group and the low anterior resection combined with prophylactic stoma (LAR) surgery group.Results:Among the 102 patients, there were 50 cases in the TCA group and 52 cases in the LAR group. In the single-arm analysis of the TCA group, the overall complication rate was 44%. The incidence rates of severe LARS at 1 month, 3 months, and 6 months after surgery were 97%, 77%, and 64% respectively. There was no significant difference in the complication rate within 30 days after surgery between the two groups,(44% vs. 38%, χ2=0.135, P>0.05). There was no significant difference in the incidence rate of severe LARS between the TCA group and the LAR group (77% vs. 69%, χ2=0.202, P>0.05), and there was not significant difference in the incidence rate of severe LARS between the two groups at the 6th month after surgery,(64% vs. 48%, χ2=1.132, P>0.05). Conclusion:In patients who underwent TCA surgery, the LARS symptoms gradually decreased over time. Compared with patients undergoing low anterior resection and stoma reversal, there were no significant differences in complications within 30 days after surgery and LARS symptoms within half a year.
6.Macrogenomics-based study of the mechanism of GeGen QinLian Decoction in ameliorating dysbiosis in a rat model of antibiotic-associated diarrhea
Qiumei TANG ; Xue HAN ; Guangyong YANG ; Rui CHEN ; Wenjia WANG ; Xiaohua TU ; Weiyi TIAN ; Yanpeng CAI ; Chi CHEN ; Guangzhi HE
Acta Laboratorium Animalis Scientia Sinica 2024;32(11):1379-1389
Objective To investigate changes in the intestinal flora and function in rats with antibiotic-associated diarrhea(A AD)treated with GeGen QinLian Decoction(GQD).Methods Sixty male or female SPF-grade SD rats were fed for 7 days and then divided randomly into Control(Con)and modeling groups(1∶5 ratio).Rats in the modeling group received clindamycin 250 mg/kg by gavage once a day for 7 consecutive days.After successful modeling,the rats were divided randomly into model(Mod),high-dose GQD(GQD-H,10.08 g/kg),medium-dose GQD(GQD-M,5.04 g/kg),low-dose GQD(GQD-L,2.52 g/kg),and live Bifidobacterium power(LBP,0.15 g/kg)groups(n=10 rats per group).GQD and LBP were administered by gavage,and the Con and Mod groups were given an equal volume of saline by gavage once a day.Feces were collected after 7 consecutive days of administration for macrogenomics sequencing analysis.Results α diversity and β diversity suggested that intestinal microbial diversity differed between the Mod and GQD-treated groups.GQD increased the abundance of thick-walled bacteria and decreased the abundance of Aspergillus at the phylum level,and increased the relative abundances of the intestinal mucus bacteria Blautia,Bacteroides,Thomasclavelia,and Mediterraneibacter,and decreased the relative abundances of Adlercreutzia,Muribaculum,and Escherichia at the genus level.GQD also up-regulated the amino acid,carbohydrate,and immune disease pathways.Conclusions GQD improves the abundance ratio of beneficial and pathogenic intestinal bacteria in rats with antibiotic-associated diarrhea,which in turn reduces the intestinal inflammatory response and repairs the intestinal immune system.
7.Macrogenomics-based study of the mechanism of GeGen QinLian Decoction in ameliorating dysbiosis in a rat model of antibiotic-associated diarrhea
Qiumei TANG ; Xue HAN ; Guangyong YANG ; Rui CHEN ; Wenjia WANG ; Xiaohua TU ; Weiyi TIAN ; Yanpeng CAI ; Chi CHEN ; Guangzhi HE
Acta Laboratorium Animalis Scientia Sinica 2024;32(11):1379-1389
Objective To investigate changes in the intestinal flora and function in rats with antibiotic-associated diarrhea(A AD)treated with GeGen QinLian Decoction(GQD).Methods Sixty male or female SPF-grade SD rats were fed for 7 days and then divided randomly into Control(Con)and modeling groups(1∶5 ratio).Rats in the modeling group received clindamycin 250 mg/kg by gavage once a day for 7 consecutive days.After successful modeling,the rats were divided randomly into model(Mod),high-dose GQD(GQD-H,10.08 g/kg),medium-dose GQD(GQD-M,5.04 g/kg),low-dose GQD(GQD-L,2.52 g/kg),and live Bifidobacterium power(LBP,0.15 g/kg)groups(n=10 rats per group).GQD and LBP were administered by gavage,and the Con and Mod groups were given an equal volume of saline by gavage once a day.Feces were collected after 7 consecutive days of administration for macrogenomics sequencing analysis.Results α diversity and β diversity suggested that intestinal microbial diversity differed between the Mod and GQD-treated groups.GQD increased the abundance of thick-walled bacteria and decreased the abundance of Aspergillus at the phylum level,and increased the relative abundances of the intestinal mucus bacteria Blautia,Bacteroides,Thomasclavelia,and Mediterraneibacter,and decreased the relative abundances of Adlercreutzia,Muribaculum,and Escherichia at the genus level.GQD also up-regulated the amino acid,carbohydrate,and immune disease pathways.Conclusions GQD improves the abundance ratio of beneficial and pathogenic intestinal bacteria in rats with antibiotic-associated diarrhea,which in turn reduces the intestinal inflammatory response and repairs the intestinal immune system.
8.Association between serum uric acid concentration and radiographic axial spondylarthritis: a cross-sectional study of 202 patients.
Yupeng LAI ; Yanpeng ZHANG ; Zhihao LEI ; Yihong HUANG ; Tongxin NI ; Pin HE ; Xiaoling LI ; Chiduo XU ; Jun XIA ; Meiying WANG
Chinese Medical Journal 2023;136(9):1114-1116
9.Comparison of 18F-PSMA-1007 and 18F-FDG PET/CT imaging in diagnosis of newly diagnosed hepatocellular carcinoma
Xiaoli MEI ; Yanpeng LI ; Shasha XU ; Yanxia YU ; Xiaobo NIU ; Xiaoting LIU ; Shuwei HE ; Xingmin HAN
Chinese Journal of Nuclear Medicine and Molecular Imaging 2023;43(11):660-664
Objective:To compare the clinical utility of 18F-prostate specific membrane antigen (PSMA)-1007 and 18F-FDG PET/CT imaging in newly diagnosed hepatocellular carcinoma (HCC). Methods:From April 2022 to July 2022, 17 patients (14 males, 3 females, age 36-73(54.4±10.1) years) with newly diagnosed HCC who underwent 18F-FDG and 18F-PSMA-1007 PET/CT imaging within 3 d in the First Affiliated Hospital of Zhengzhou University were prospectively enrolled. ROIs were drawn from normal liver tissue (L), abdominal aorta (A), right gluteus medius (M), and SUV max of these regions were compared with the SUV max of primary tumor (T). Wilcoxon rank sum test and Kruskal-Wallis rank sum test were used to analyze the data. Results:18F-FDG PT/CT, 18F-PSMA-1007 PET/CT and enhanced MRI detected 1(0, 2), 2(1, 5) and 2(1, 4) tumor lesions of the liver in each patient respectively ( H=7.10, P=0.029), and 18F-PSMA-1007 detected more lesions than 18F-FDG ( P=0.024). Although SUV max of 18F-PSMA-1007 in HCC was significantly higher than that of 18F-FDG (25.7(17.1, 45.1) vs 6.3(2.9, 12.4); z=3.39, P=0.001), there was no significant difference of T/L ratio between 18F-PSMA-1007 and 18F-FDG PET/CT imaging (2.7(2.1, 4.7) vs 1.6(1.0, 4.5); z=0.52, P=0.602). T/A and T/M ratios were significantly higher in 18F-PSMA-1007 PET/CT imaging than those in 18F-FDG PET/CT imaging ( z values: 3.15, 3.53, P values: 0.002, <0.001). 18F-PSMA-1007 PET/CT imaging found high uptake foci in the liver and ribs in 2 cases, which were pathologically confirmed as bone metastasis of HCC, while those lesions were not found by 18F-FDG imaging. Conclusion:Compared with 18F-FDG, 18F-PSMA-1007 PET/CT demonstrates higher tumor uptake, more intrahepatic tumors foci and distant bone metastases.
10.Bioinformatics analysis of differential gene expression in HEp-2 cells infected with human respiratory syncytial virus
Yanbin SU ; Yong LIU ; Ye WANG ; Shuo WANG ; Junyu XUE ; Tianxiao ZHANG ; Xianglei PENG ; Yanpeng ZHENG ; Lishu ZHANG ; Yuanhui FU ; Jinsheng HE
Chinese Journal of Experimental and Clinical Virology 2021;35(5):575-580
Objective:To find clues potentially valuable for fighting against infection with human respiratory syncytial virus (HRSV), the differentially expressed genes in HEp-2 cells infected with HRSV were analyzed.Methods:Gene expression profiles of HEp-2 cells infected with HRSV were collected from the public gene expression omnibus (GEO) database. The differentially expressed genes following HRSV infection at each time point of 4, 8, 12, and 15 hours were found using R language. The differentially expressed genes were analyzed by gene ontology (GO), KEGG pathway and protein-protein interaction network (PPI). Genes with relatively high protein interaction in PPI were randomly selected for quantitative reverse transcription-polymerase chain reaction (qRT-PCR) verification at the transcription level from HEp-2 cells after HRSV infection at 4 hours.Results:A total of 101 differentially expressed genes were determined, including 92 upregulated genes and 9 downregulated genes. Function enrichment analysis revealed that HRSV infection could cause significant changes in multiple signaling pathways such as immune response in HEp-2 cells. The results of qRT-PCR were consistent with the trend of transcriptome data.Conclusions:The differentially expressed genes and the change of signaling pathways in HRSV-infected HEp-2 cells is of great significance to the studies on pathogenic mechanism and prevention of HRSV infection.

Result Analysis
Print
Save
E-mail