MGDDI: A multi-scale graph neural networks for drug-drug interaction prediction.

Drug-drug interaction (DDI) prediction is crucial for identifying interactions within drug combinations, especially adverse effects due to physicochemical incompatibility. While current methods have made strides in predicting adverse drug interactions, limitations persist. Most methods rely on handcrafted features, restricting their applicability. They predominantly extract information from individual drugs, neglecting the importance of interaction details between drug pairs. To address these issues, we propose MGDDI, a graph neural network-based model for predicting potential adverse drug interactions. Notably, we use a multiscale graph neural network (MGNN) to learn drug molecule representations, addressing substructure size variations and preventing gradient issues. For capturing interaction details between drug pairs, we integrate a substructure interaction learning module based on attention mechanisms. Our experimental results demonstrate MGDDI's superiority in predicting adverse drug interactions, offering a solution to current methodological limitations.

Postoperative delirium prediction after cardiac surgery using machine learning models.

OBJECTIVE: Postoperative delirium (POD) is a common postoperative complication in elderly patients, especially those undergoing cardiac surgery, which seriously affects the short- and long-term prognosis of patients. Early identification of risk factors for the development of POD can help improve the perioperative management of surgical patients. In the present study, five machine learning models were developed to predict patients at high risk of delirium after cardiac surgery and their performance was compared. METHODS: A total of 367 patients who underwent cardiac surgery were retrospectively included in this study. Using single-factor analysis, 21 risk factors for POD were selected for inclusion in machine learning. The dataset was divided using 10-fold cross-validation for model training and testing. Five machine learning models (random forest (RF), support vector machine (SVM), radial based kernel neural network (RBFNN), K-nearest neighbour (KNN), and Kernel ridge regression (KRR)) were compared using area under the receiver operating characteristic curve (AUC-ROC), accuracy (ACC), sensitivity (SN), specificity (SPE), and Matthews coefficient (MCC). RESULTS: Among 367 patients, 105 patients developed POD, the incidence of delirium was 28.6 %. Among the five ML models, RF had the best performance in ACC (87.99 %), SN (69.27 %), SPE (95.38 %), MCC (70.00 %) and AUC (0.9202), which was far superior to the other four models. CONCLUSION: Delirium is common in patients after cardiac surgery. This analysis confirms the importance of the computational ML models in predicting the occurrence of delirium after cardiac surgery, especially the outstanding performance of the RF model, which has practical clinical applications for early identification of patients at risk of developing POD.

[Analgesic effect and mechanism of electroacupuncture on SNI rats based on microglia-BDNF-neuron signal].

OBJECTIVE: To observe the effect of electroacupuncture (EA) at "Huantiao" (GB 30) and "Weizhong" (BL 40) on the activation of glial cells, the expression of brain-derived neurotrophic factor (BDNF), excitability and the number of dendritic spines of neurons in the spinal dorsal horn in rats with spared nerve injury (SNI) of sciatic nerve, and to explore the analgesic mechanism of EA on SNI. METHODS: PartⅠ: Sixty SD rats were randomly divided into a sham operation group, a model group, an EA group and a sham EA group, 15 rats in each group. Except the sham operation group, the SNI rat model was established in the remaining groups. The rats in the sham operation group were only treated with incision without damaging the nerve. The rats in the EA group were treated with EA at "Huantiao" (GB 30) and "Weizhong" (BL 40) on the affected side, continuous wave, frequency of 2 Hz, current intensity of 1 mA, 30 minutes each time, once a day, for 14 days. The rats in the sham EA group were treated with EA at points 0.5 cm next to "Huantiao" (GB 30) and "Weizhong" (BL 40) on the affected side; the manipulation, EA parameters and treatment course were the same as the EA group. The latency of thermal foot contraction reflex and the threshold of mechanical foot contraction reflex were detected 1 day before modeling and 3, 7 and 14 days after modeling. Fourteen days after modeling, Western blot was used to detect the protein expressions of ionized binding adapter junction protein 1 (Iba-1), glial fibrillary acidic protein (GFAP), BDNF and c-Fos in the spinal dorsal horn; the expressions of Iba-1 and c-Fos proteins in the spinal dorsal horn were detected by immunofluorescence staining; immunohistochemical method was used to detect the expression of GFAP protein in the spinal dorsal horn; Golgi staining was used to detect the number of dendritic spines in spinal dorsal horn neurons. PartⅡ: Thirty SD rats were randomly divided into a control group, a BDNF group and a BDNF+anti-TrkB group, 10 rats in each group. The control group was treated with intrathecal injection of 10 µL mixture with 1︰1 of 0.9% sodium chloride solution and dimethyl sulfoxide (DMSO); the BDNF group was treated with intrathecal injection of 10 µg rat recombinant BDNF dissolved in 10 µL mixture with 1︰1 of 0.9% sodium chloride solution and DMSO; the BDNF+anti-TrkB group was treated with intrathecal injection of 10 µg rat recombinant BDNF and 30 µg tyrosine kinase receptor B (TrkB) antibody dissolved in 10 µL mixture with 1︰1 of 0.9% sodium chloride solution and DMSO. The threshold of mechanical foot retraction reflex was detected 1 day before intrathecal injection and 1, 3 and 7 days after injection. Seven days after injection, the expression of c-Fos protein in the spinal dorsal horn was detected by Western blot and immunofluorescence staining. RESULTS: PartⅠ: Compared with the sham operation group, 3, 7 and 14 days after modeling, the latency of thermal foot contraction reflex and the threshold of mechanical foot contraction reflex in the model group were decreased (P<0.05); 7 and 14 days after modeling, compared with the model group, the latency of thermal foot contraction reflex and the threshold of mechanical foot contraction reflex in the EA group were increased (P<0.05). The expressions of Iba-1, GFAP, BDNF, c-Fos proteins and the number of neuronal dendritic spines in the spinal dorsal horn in the model group were higher than those in the sham operation group (P<0.05); the expressions of Iba-1, BDNF, c-Fos proteins and the number of neuronal dendritic spines in the EA group were lower than those in the model group (P<0.05). PartⅡ: 3 and 7 days after intrathecal injection, the threshold of mechanical foot retraction reflex in the BDNF group was lower than that in the control group (P<0.05); the threshold of mechanical foot retraction reflex in the BDNF+anti-TrkB group was higher than that in the BDNF group (P<0.05). The expression of c-Fos protein in spinal dorsal horn in the BDNF group was higher than that in the control group (P<0.05); the expression of c-Fos protein in spinal dorsal horn in the BDNF+anti-TrkB group was lower than that in the BDNF group (P<0.05). CONCLUSION: The analgesic effect of EA at "Huantiao" (GB 30) and "Weizhong" (BL 40) on SNI rats may be related to inhibiting the activation of microglia in the dorsal horn of the spinal cord, thereby blocking the signal of microglia-BDNF-neuron, and finally reducing the excitability of neurons.

Using a machine learning-based risk prediction model to analyze the coronary artery calcification score and predict coronary heart disease and risk assessment.

OBJECTIVES: To calculate the coronary artery calcification score (CACS) obtained from coronary artery computed tomography angiography (CCTA) examination and combine it with the influencing factors of coronary artery calcification (CAC), which is then analyzed by machine learning (ML) to predict the probability of coronary heart disease(CHD). METHODS: All patients who were admitted to the Affiliated Hospital of Traditional Chinese Medicine of Southwest Medical University from January 2019 to March 2022, suspected of CHD, and underwent CCTA inspection were retrospectively selected. The degree of CAC was quantified based on the Agatston score. To compare the correlation between the CACS and clinical-related factors, we collected 31 variables, including hypertension, diabetes, smoking, hyperlipidemia, among others. ML models containing the random forest (RF), radial basis function neural network (RBFNN),support vector machine (SVM),K-Nearest Neighbor algorithm (KNN) and kernel ridge regression (KRR) were used to assess the risk of CHD based on CACS and clinical-related factors. RESULTS: Among the five ML models, RF achieves the best performance about accuracy (ACC) (78.96%), sensitivity (SN) (93.86%), specificity(Spe) (51.13%), and Matthew's correlation coefficient (MCC) (0.5192).It also has the best area under the receiver operator characteristic curve (ROC) (0.8375), which is far superior to the other four ML models. CONCLUSION: Computer ML model analysis confirmed the importance of CACS in predicting the occurrence of CHD, especially the outstanding RF model, making it another advancement of the ML model in the field of medical analysis.

The Effect of Dexmedetomidine on Cognitive Function and Protein Expression of Aß, p-Tau, and PSD95 after Extracorporeal Circulation Operation in Aged Rats.

Postoperative cognitive dysfunction (POCD) is a kind of serious neurologic complications and dexmedetomidine has a certain effect on POCD. However, functional mechanism of dexmedetomidine on POCD still remains unclear, so the research mainly studied the effect of dexmedetomidine on cognitive function and protein expression in hippocampus and prefrontal cortex cerebrospinal fluid after extracorporeal circulation operation in aged rats. We Found that, compared with POCD group, the cognitive function was improved in POCD + Dex group. We speculate that dexmedetomidine could improve the cognitive function after extracorporeal circulation operation in aged rats and Aß, p-Tau, and PSD95 protein might have contributed to this favorable outcome.

Urine and serum microRNA-1 as novel biomarkers for myocardial injury in open-heart surgeries with cardiopulmonary bypass.

MicroRNA-1 (miR-1) is a cardio-specific/enriched microRNA. Our recent studies have revealed that serum and urine miR-1 could be a novel sensitive biomarker for acute myocardial infarction. Open-heart surgeries with cardiopulmonary bypass (CPB) are often accompanied with surgery injury and CPB-associated injury on the hearts. However, the association of miR-1 and these intra-operative and post-operative cardiac injures is unknown. The objective of this study was to test the hypothesis that urine and serum miR-1 might be a novel biomarker for myocardial injuries in open-heart surgeries with CPB. Serum and urine miR-1 levels in 20 patients with elective mitral valve surgery were measured at pre-surgery, pre-CPB, 60 min post-CBP, and 24h post-CBP. Serum cardiac troponin-I (cTnI) was used as a positive control biomarker for cardiac injury. Compared with these in pre-operative and pre-CPB groups, the levels of miR-1 in serum and urine from patients after open-heart surgeries and CPB were significant increased at all observed time points. A similar pattern of serum cTnI levels and their strong positive correlation with miR-1 levels were identified in these patients. The results suggest that serum and urine miR-1 may be a novel sensitive biomarker for myocardial injury in open-heart surgeries with CPB.