Cardiotoxicity is a critical issue in drug development that poses serious health risks,including potentially fatal arrhythmias.The human ether-à-go-go related gene(hERG)potassium channel,as one of the pri-mary targets of cardiotoxicity,has garnered widespread attention.Traditional cardiotoxicity testing methods are expensive and time-consuming,making computational virtual screening a suitable alter-native.In this study,we employed machine learning techniques utilizing molecular fingerprints and descriptors to predict the cardiotoxicity of compounds,with the aim of improving prediction accuracy and efficiency.We used four types of molecular fingerprints and descriptors combined with machine learning and deep learning algorithms,including Gaussian naive Bayes(NB),random forest(RF),support vector machine(SVM),K-nearest neighbors(KNN),eXtreme gradient boosting(XGBoost),and Trans-former models,to build predictive models.Our models demonstrated advanced predictive performance.The best machine learning model,XGBoost Morgan,achieved an accuracy(ACC)value of 0.84,and the deep learning model,Transformer_Morgan,achieved the best ACC value of 0.85,showing a high ability to distinguish between toxic and non-toxic compounds.On an external independent validation set,it achieved the best area under the curve(AUC)value of 0.93,surpassing ADMETlab3.0,Cardpred,and CardioDPi.In addition,we explored the integration of molecular descriptors and fingerprints to enhance model performance and found that ensemble methods,such as voting and stacking,provided slight improvements in model stability.Furthermore,the SHapley Additive exPlanations(SHAP)explanations revealed the relationship between benzene rings,fluorine-containing groups,NH groups,oxygen in ether groups,and cardiotoxicity,highlighting the importance of these features.This study not only improved the predictive accuracy of cardiotoxicity models but also promoted a more reliable and scientifically interpretable method for drug safety assessment.Using computational methods,this study facilitates a more efficient drug development process,reduces costs,and improves the safety of new drug candidates,ultimately benefiting medical and public health.

Aim To investigate whether protein inhibitor of activated STAT3(PIAS3)deficiency exacerbates the occurrence and development of atherosclerosis(As)in female ApoE knockout mice.Methods PIAS3 gene knockout mice with ApoE-/-background(PIAS3-/-/ApoE-/-)and their littermate PIAS3+/+/ApoE-/-mice were bred and fed with a high-fat/high-cholesterol diet for 12 weeks to induce As.Body weight(every week)and plasma lipid levels including to-tal cholesterol,triglyceride,high density lipoprotein cholesterol and non-high density lipoprotein cholesterol(every 4 weeks)of the mice were measured.Oil red O staining,HE staining,immunohistochemistry staining and immunofluores-cence staining were performed on mouse aortic tree and frozen sections of aortic root to evaluate the area,cellular composi-tion and stability of As plaques.Moreover,the expression of estrogen receptor α(ERα)and its co-localization with vas-cular smooth muscle cells(VSMC)in plaques were determined by immunofluorescence staining.Results Compared with PIAS3+/+/ApoE-/-mice,PIAS3-/-/ApoE-/-mice showed no significant differences in body weight,major organ weight(heart,liver,spleen,kidney and epididymal fat)and plasma lipid levels;however,PIAS3 deficiency promoted the forma-tion of As in female PIAS3-/-/ApoE-/-mice.Compared with PIAS3+/+/ApoE-/-mice,PIAS3-/-/ApoE-/-mice showed an increased lipid accumulation and a decreased VSMC content in As plaques(P＜0.05),leading to a decrease in plaque stability.In addition,the expression of ERα in the As plaques of PIAS3-/-/ApoE-/-mice was significantly downregulated(P＜0.05),and there was a obvious co-localization between ERα and VSMC.The reduction of VSMC content in PIAS3-/-/ApoE-/-mouse plaques might lead to a decrease of ERα expression,thereby weakening the anti-As effect of es-trogen.Conclusion PIAS3 deficiency exacerbates the formation of As plaques in female PIAS3-/-/ApoE-/-mice,which might be due to the regulatory effect of PIAS3 on ERα expression in plaques.

Aim To investigate whether protein inhibitor of activated STAT3(PIAS3)deficiency exacerbates the occurrence and development of atherosclerosis(As)in female ApoE knockout mice.Methods PIAS3 gene knockout mice with ApoE-/-background(PIAS3-/-/ApoE-/-)and their littermate PIAS3+/+/ApoE-/-mice were bred and fed with a high-fat/high-cholesterol diet for 12 weeks to induce As.Body weight(every week)and plasma lipid levels including to-tal cholesterol,triglyceride,high density lipoprotein cholesterol and non-high density lipoprotein cholesterol(every 4 weeks)of the mice were measured.Oil red O staining,HE staining,immunohistochemistry staining and immunofluores-cence staining were performed on mouse aortic tree and frozen sections of aortic root to evaluate the area,cellular composi-tion and stability of As plaques.Moreover,the expression of estrogen receptor α(ERα)and its co-localization with vas-cular smooth muscle cells(VSMC)in plaques were determined by immunofluorescence staining.Results Compared with PIAS3+/+/ApoE-/-mice,PIAS3-/-/ApoE-/-mice showed no significant differences in body weight,major organ weight(heart,liver,spleen,kidney and epididymal fat)and plasma lipid levels;however,PIAS3 deficiency promoted the forma-tion of As in female PIAS3-/-/ApoE-/-mice.Compared with PIAS3+/+/ApoE-/-mice,PIAS3-/-/ApoE-/-mice showed an increased lipid accumulation and a decreased VSMC content in As plaques(P＜0.05),leading to a decrease in plaque stability.In addition,the expression of ERα in the As plaques of PIAS3-/-/ApoE-/-mice was significantly downregulated(P＜0.05),and there was a obvious co-localization between ERα and VSMC.The reduction of VSMC content in PIAS3-/-/ApoE-/-mouse plaques might lead to a decrease of ERα expression,thereby weakening the anti-As effect of es-trogen.Conclusion PIAS3 deficiency exacerbates the formation of As plaques in female PIAS3-/-/ApoE-/-mice,which might be due to the regulatory effect of PIAS3 on ERα expression in plaques.

Cardiotoxicity is a critical issue in drug development that poses serious health risks, including potentially fatal arrhythmias. The human ether-à-go-go related gene (hERG) potassium channel, as one of the primary targets of cardiotoxicity, has garnered widespread attention. Traditional cardiotoxicity testing methods are expensive and time-consuming, making computational virtual screening a suitable alternative. In this study, we employed machine learning techniques utilizing molecular fingerprints and descriptors to predict the cardiotoxicity of compounds, with the aim of improving prediction accuracy and efficiency. We used four types of molecular fingerprints and descriptors combined with machine learning and deep learning algorithms, including Gaussian naive Bayes (NB), random forest (RF), support vector machine (SVM), K-nearest neighbors (KNN), eXtreme gradient boosting (XGBoost), and Transformer models, to build predictive models. Our models demonstrated advanced predictive performance. The best machine learning model, XGBoost Morgan, achieved an accuracy (ACC) value of 0.84, and the deep learning model, Transformer_Morgan, achieved the best ACC value of 0.85, showing a high ability to distinguish between toxic and non-toxic compounds. On an external independent validation set, it achieved the best area under the curve (AUC) value of 0.93, surpassing ADMETlab3.0, Cardpred, and CardioDPi. In addition, we explored the integration of molecular descriptors and fingerprints to enhance model performance and found that ensemble methods, such as voting and stacking, provided slight improvements in model stability. Furthermore, the SHapley Additive exPlanations (SHAP) explanations revealed the relationship between benzene rings, fluorine-containing groups, NH groups, oxygen in ether groups, and cardiotoxicity, highlighting the importance of these features. This study not only improved the predictive accuracy of cardiotoxicity models but also promoted a more reliable and scientifically interpretable method for drug safety assessment. Using computational methods, this study facilitates a more efficient drug development process, reduces costs, and improves the safety of new drug candidates, ultimately benefiting medical and public health.

Objective:To observe and evaluate the clinical efficacy and safety of albumin-bound paclitaxel as first-line treatment for patients with advanced pancreatic cancer in China, and to explore the prognosis-related molecules in pancreatic cancer based on next-generation sequencing (NGS) of tumor tissues.Methods:From December 2018 to December 2020, patients with locally advanced or metastatic pancreatic cancer were recruited to accept albumin-bound paclitaxel as first-line treatment in the oncology departments of 24 hospitals in East China. The primary endpoints were overall survival (OS) and treatment related adverse events, and the secondary endpoint was progression-free survival (PFS). Adverse effects were graded using Common Terminology Criteria for Adverse Events 5.0 (CTCAE 5.0). NGS sequencing on the primary or metastatic tissue samples of pancreatic cancer obtained through surgical resection or biopsy was performed.Results:This study recruited 229 patients, including 70 patients with locally advanced pancreatic cancer (LAPC) and 159 patients with metastatic pancreatic cancer (mPC). The disease control rate was 79.9% and the objective response rate is 36.3%.The common adverse effects during treatment were anaemia (159 cases), leucopenia (170 cases), neutropenia (169 cases), increased aminotransferases (110 cases), and thrombocytopenia (95 cases), and the incidence of grade 3-4 neutropenia is 12.2% (28/229). The median follow-up time was 21.2 months (95% CI: 18.5-23.1 months). The median PFS (mPFS) was 5.3 months (95% CI: 4.37-4.07 months) and the median OS (mOS) was 11.2 months (95% CI: 9.5-12.9 months). The mPFS of patients with LAPC was 7.4 months (95% CI: 6.6-11.2 months), and their mOS was 15.5 months (95% CI: 12.6-NA months). The mPFS of patients with mPC was 3.9 months (95% CI: 3.4-5.1 months), and their mOS was 9.3 months (95% CI: 8.0-10.8 months). Multivariate Cox regression analysis showed that clinical stage ( HR=1.47, 95% CI: 1.06-2.04), primary tumor site ( HR=0.64, 95% CI: 0.48-0.86), Eastern Cooperative Oncology Group Performance Status (ECOG PS) score ( HR=2.66, 95% CI: 1.53-4.65), and whether to combine radiotherapy ( HR=0.65, 95% CI: 0.42-1.00) were independent influencing factors for the PFS of these patients. The primary tumor site ( HR=0.68, 95% CI: 0.48-0.95), ECOG score ( HR=5.82, 95% CI: 3.14-10.82), and whether to combine radiotherapy ( HR=0.58, 95% CI: 0.35-0.96) were independent influencing factors of the OS of these patients. The most frequent gene mutations in these advanced stage pancreatic patients were KRAS (89.66%), TP53 (77.01%), CDKN2A (32.18%), and SMAD4 (21.84%) by NGS of tumor tissues from 87 pancreatic cancer patients with sufficient specimens. Further analysis revealed that mutations in CDKN2B, PTEN, FGF6, and RBBP8 genes were significantly associated with an increased risk of death ( P＜0.05). Conclusion:Albumin-bound paclitaxel as first-line treatment demonstrated feasible anti-tumor efficacy and manageable safety for patients with advanced pancreatic cancer in China.

Objective:To observe and evaluate the clinical efficacy and safety of albumin-bound paclitaxel as first-line treatment for patients with advanced pancreatic cancer in China, and to explore the prognosis-related molecules in pancreatic cancer based on next-generation sequencing (NGS) of tumor tissues.Methods:From December 2018 to December 2020, patients with locally advanced or metastatic pancreatic cancer were recruited to accept albumin-bound paclitaxel as first-line treatment in the oncology departments of 24 hospitals in East China. The primary endpoints were overall survival (OS) and treatment related adverse events, and the secondary endpoint was progression-free survival (PFS). Adverse effects were graded using Common Terminology Criteria for Adverse Events 5.0 (CTCAE 5.0). NGS sequencing on the primary or metastatic tissue samples of pancreatic cancer obtained through surgical resection or biopsy was performed.Results:This study recruited 229 patients, including 70 patients with locally advanced pancreatic cancer (LAPC) and 159 patients with metastatic pancreatic cancer (mPC). The disease control rate was 79.9% and the objective response rate is 36.3%.The common adverse effects during treatment were anaemia (159 cases), leucopenia (170 cases), neutropenia (169 cases), increased aminotransferases (110 cases), and thrombocytopenia (95 cases), and the incidence of grade 3-4 neutropenia is 12.2% (28/229). The median follow-up time was 21.2 months (95% CI: 18.5-23.1 months). The median PFS (mPFS) was 5.3 months (95% CI: 4.37-4.07 months) and the median OS (mOS) was 11.2 months (95% CI: 9.5-12.9 months). The mPFS of patients with LAPC was 7.4 months (95% CI: 6.6-11.2 months), and their mOS was 15.5 months (95% CI: 12.6-NA months). The mPFS of patients with mPC was 3.9 months (95% CI: 3.4-5.1 months), and their mOS was 9.3 months (95% CI: 8.0-10.8 months). Multivariate Cox regression analysis showed that clinical stage ( HR=1.47, 95% CI: 1.06-2.04), primary tumor site ( HR=0.64, 95% CI: 0.48-0.86), Eastern Cooperative Oncology Group Performance Status (ECOG PS) score ( HR=2.66, 95% CI: 1.53-4.65), and whether to combine radiotherapy ( HR=0.65, 95% CI: 0.42-1.00) were independent influencing factors for the PFS of these patients. The primary tumor site ( HR=0.68, 95% CI: 0.48-0.95), ECOG score ( HR=5.82, 95% CI: 3.14-10.82), and whether to combine radiotherapy ( HR=0.58, 95% CI: 0.35-0.96) were independent influencing factors of the OS of these patients. The most frequent gene mutations in these advanced stage pancreatic patients were KRAS (89.66%), TP53 (77.01%), CDKN2A (32.18%), and SMAD4 (21.84%) by NGS of tumor tissues from 87 pancreatic cancer patients with sufficient specimens. Further analysis revealed that mutations in CDKN2B, PTEN, FGF6, and RBBP8 genes were significantly associated with an increased risk of death ( P＜0.05). Conclusion:Albumin-bound paclitaxel as first-line treatment demonstrated feasible anti-tumor efficacy and manageable safety for patients with advanced pancreatic cancer in China.

Objective:To analyze the patterns of dynamic functional connectivity(dFC) at rest in different obsessive-compulsive disorder(OCD) subtypes based on gray matter volume, and explore the relationship between abnormal gray matter volume and dFC patterns and clinical symptoms in different OCD subtypes.Methods:Clinical data were prospectively collected from a total of 48 patients with OCD from inpatient and outpatient departments of the Fourth Affiliated Hospital of Qiqihar Medical College(20 cases) and Qiqihar Mental Health Center(28 cases) from September 2013 to March 2020. There were 33 males and 15 females, aged 18 to 42(27.1±6.7) years. During the same period, 39 age-matched healthy controls were also recruited, including 31 males and 8 females, aged 18 to 43(28.5±8.6) years. Structural and resting-state functional magnetic resonance imaging scans were performed on 48 OCD patients(OCD group) who met the ICD-10 diagnostic criteria and 39 healthy controls(control group). Using voxel-based morphometry(VBM) and semi-supervised machine learning methods, the patient group was characterized by gray matter volume to perform subtype analysis, and brain regions with abnormal gray matter volume were selected as regions of interest for whole-brain voxel-wise dFC analysis. Correlations between abnormal gray matter volume and dFC and clinical symptoms in OCD patients were analyzed.Results:Based on gray matter volume, OCD patients were computed into two subtypes, of which subtype 1 increased gray matter volume in the right central posterior gyrus(809 voxels; t=4.31; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed), and subtype 2 showed reduced gray matter volume in the left superior temporal gyrus(1 118 voxels; t=-4.37; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed). In subtype 1, the dFC between the right central posterior gyrus and left medial superior frontal gyrus was significantly reduced(187 voxels; t=-4.42; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed); in subtype 2, dFC between the left superior temporal gyrus and left hippocampus, and left central opercular cortex was significantly reduced(272 voxels、99 voxels; t=-4.69, -5.01; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed). In subtype 1, dFC between the right central posterior gyrus and left medial superior frontal gyrus was negatively correlated with illness duration( r=-0.579; P=0.002, uncorrected). In subtype 2, dFC between the left superior temporal gyrus and left hippocampus was positively correlated with illness duration( r=0.578; P=0.003, uncorrected). Conclusions:In the resting state, patients with different subtypes of OCD may have different patterns of dFC abnormalities based on gray matter volume.

Objective:To analyze the patterns of dynamic functional connectivity(dFC) at rest in different obsessive-compulsive disorder(OCD) subtypes based on gray matter volume, and explore the relationship between abnormal gray matter volume and dFC patterns and clinical symptoms in different OCD subtypes.Methods:Clinical data were prospectively collected from a total of 48 patients with OCD from inpatient and outpatient departments of the Fourth Affiliated Hospital of Qiqihar Medical College(20 cases) and Qiqihar Mental Health Center(28 cases) from September 2013 to March 2020. There were 33 males and 15 females, aged 18 to 42(27.1±6.7) years. During the same period, 39 age-matched healthy controls were also recruited, including 31 males and 8 females, aged 18 to 43(28.5±8.6) years. Structural and resting-state functional magnetic resonance imaging scans were performed on 48 OCD patients(OCD group) who met the ICD-10 diagnostic criteria and 39 healthy controls(control group). Using voxel-based morphometry(VBM) and semi-supervised machine learning methods, the patient group was characterized by gray matter volume to perform subtype analysis, and brain regions with abnormal gray matter volume were selected as regions of interest for whole-brain voxel-wise dFC analysis. Correlations between abnormal gray matter volume and dFC and clinical symptoms in OCD patients were analyzed.Results:Based on gray matter volume, OCD patients were computed into two subtypes, of which subtype 1 increased gray matter volume in the right central posterior gyrus(809 voxels; t=4.31; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed), and subtype 2 showed reduced gray matter volume in the left superior temporal gyrus(1 118 voxels; t=-4.37; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed). In subtype 1, the dFC between the right central posterior gyrus and left medial superior frontal gyrus was significantly reduced(187 voxels; t=-4.42; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed); in subtype 2, dFC between the left superior temporal gyrus and left hippocampus, and left central opercular cortex was significantly reduced(272 voxels、99 voxels; t=-4.69, -5.01; GRF corrected voxel P<0.001, cluster P<0.05 two-tailed). In subtype 1, dFC between the right central posterior gyrus and left medial superior frontal gyrus was negatively correlated with illness duration( r=-0.579; P=0.002, uncorrected). In subtype 2, dFC between the left superior temporal gyrus and left hippocampus was positively correlated with illness duration( r=0.578; P=0.003, uncorrected). Conclusions:In the resting state, patients with different subtypes of OCD may have different patterns of dFC abnormalities based on gray matter volume.

OBJECTIVE: To isolate the phenolic amides from the dried bulbs of Allium chinense and investigate their myocardium protective activities. METHODS: The chemical constituents were isolated and purified by combining with silica gel column, Sephadex LH-20 column, HPLC and other chromatography techniques. Their structures were elucidated by NMR techniques and mass spectrometry. The isolated compounds were evaluated to determine their protective effect for myocardium cells in vitro. RESULTS: Two new phenolic amides, namely, alichinemide I ( 1) and alichinemide II ( 2), and six konwn amides were isolated from the dried bulbs of A. chinense. The structures of compounds 3- 8 were identified as 3-indolcarbaldehyde ( 3), 1-(2-aminophenyl)urea ( 4), 2,3,4,9-tetrahydro-1H-pyrido[3,4-b]indole-3-carboxylic acid ( 5), N-trans-feruloyltyramine ( 6), N-trans-p-coumaroyltyramine ( 7), and N-(3,4-dimethoxyphenethyl) acetamide ( 8). Compound 3 (50 μmol/L) showed significant inhibitory effect on the damage of H9c2 myocardial cells induced by H₂O₂in vitro. CONCLUSION Compounds 1 and 2 were new phenolic amides. Compound 3 could be one of the potential myocardium protective constituents of A. chinense.

Objective The chemical composition and blood components of Tianmaxingnao capsules were discovered and examined using UPLC-Q/TOF-MS,and the possible pharmacological substance basis was preliminarily elucidated.Methods An UPLC-Q/TOF-MS method was developed in this study to determine the chemical composition of Tianma Xingnao capsules.After administration of Tianmaxingnao capsules,gather and examine rat plasma samples to investigate the exposed components of Tianmaxingnao capsules in rats.Results A total of 195 chemical components were identified in Tianmaxingnao capsules,including phenols,triterpenoid saponins,phenylethanol glycosides,cyclic ether terpenes,lipids,and phenylpropanoids.These components include those that are typical of Gastrodia elata Bl,Pheretima,Cistanche deserticola Ma,Rehmannia glutinosa,Polygala hybrida DC,and Acorus tatarinowii.Rat plasma samples were used to identify 37 prototype components and 3 metabolites of Tianmaxingnao capsules after they were administered.Conclusion This approach is easy to use,effective,sensitive,and precise.It may be used to investigate Tianmaxingnao capsules and the components that reach the bloodstream in detail,as well as to provide a first understanding of the pharmacological basis of the capsules.This study serves as a foundation for clarifying the pharmacological basis of Tianmaxingnao capsules and holds some reference value in exposing the pharmacological mechanism of the product.