The biomedical knowledge graph of symptom phenotype in coronary artery plaque: machine learning-based analysis of real-world clinical data

A knowledge graph can effectively showcase the essential characteristics of data and is increasingly emerging as a significant means of integrating information in the field of artificial intelligence. Coronary artery plaque represents a significant etiology of cardiovascular events, posing a diagnostic challenge for clinicians who are confronted with a multitude of nonspecific symptoms. To visualize the hierarchical relationship network graph of the molecular mechanisms underlying plaque properties and symptom phenotypes, patient symptomatology was extracted from electronic health record data from real-world clinical settings. Phenotypic networks were constructed utilizing clinical data and protein‒protein interaction networks. Machine learning techniques, including convolutional neural networks, Dijkstra's algorithm, and gene ontology semantic similarity, were employed to quantify clinical and biological features within the network. The resulting features were then utilized to train a K-nearest neighbor model, yielding 23 symptoms, 41 association rules, and 61 hub genes across the three types of plaques studied, achieving an area under the curve of 92.5%. Weighted correlation network analysis and pathway enrichment were subsequently utilized to identify lipid status-related genes and inflammation-associated pathways that could help explain the differences in plaque properties. To confirm the validity of the network graph model, we conducted coexpression analysis of the hub genes to evaluate their potential diagnostic value. Additionally, we investigated immune cell infiltration, examined the correlations between hub genes and immune cells, and validated the reliability of the identified biological pathways. By integrating clinical data and molecular network information, this biomedical knowledge graph model effectively elucidated the potential molecular mechanisms that collude symptoms, diseases, and molecules.

Effect of botanical drugs in improving symptoms of hypertensive nephropathy: Analysis of real-world data, retrospective cohort, network, and experimental assessment

Background/aim: Hypertensive nephropathy (HN) is a common complication of hypertension. Traditional Chinese medicine has long been used in the clinical treatment of Hypertensive nephropathy. However, botanical drug prescriptions have not been summarized. The purpose of this study is to develop a prescription for improving hypertensive nephropathy, explore the evidence related to clinical application of the prescription, and verify its molecular mechanism of action.

Methods: In this study, based on the electronic medical record data on Hypertensive nephropathy, the core botanical drugs and patients’ symptoms were mined using the hierarchical network extraction and fast unfolding algorithm, and the protein interaction network between botanical drugs and Hypertensive nephropathy was established. The K-nearest neighbors (KNN) model was used to analyze the clinical and biological characteristics of botanical drug compounds to determine the effective compounds. Hierarchical clustering was used to screen for effective botanical drugs. The clinical efficacy of botanical drugs was verified by a retrospective cohort. Animal experiments were performed at the target and pathway levels to analyze the mechanism.

Results: A total of 14 botanical drugs and five symptom communities were obtained from real-world clinical data. In total, 76 effective compounds were obtained using the K-nearest neighbors model, and seven botanical drugs were identified as Gao Shen Formula by hierarchical clustering. Compared with the classical model, the Area under the curve (AUC) value of the K-nearest neighbors model was the best; retrospective cohort verification showed that Gao Shen Formula reduced serum creatinine levels and Chronic kidney disease (CKD) stage [OR = 2.561, 95% CI (1.025–6.406), p < 0.05]. With respect to target and pathway enrichment, Gao Shen Formula acts on inflammatory factors such as TNF-α, IL-1β, and IL-6 and regulates the NF-κB signaling pathway and downstream glucose and lipid metabolic pathways.

Conclusion: In the retrospective cohort, we observed that the clinical application of Gao Shen Formula alleviates the decrease in renal function in patients with hypertensive nephropathy. It is speculated that Gao Shen Formula acts by reducing inflammatory reactions, inhibiting renal damage caused by excessive activation of the renin-angiotensin-aldosterone system, and regulating energy metabolism.

Predicting Coupled Herbs for the Treatment of Hypertension Complicated with Coronary Heart Disease in Real-World Data Based on a Complex Network and Machine Learning

Hypertension and coronary heart disease are the most common cardiovascular diseases, and traditional Chinese medicine is applied as an auxiliary treatment for common cardiovascular diseases. This study is based on 3 years of electronic medical record data from the Affiliated Hospital of Shandong University of Traditional Chinese Medicine. A complex network and machine learning algorithm were used to establish a screening model of coupled herbs for the treatment of hypertension complicated with coronary heart disease. A total of 5688 electronic medical records were collected to establish the prescription network and symptom database. The hierarchical network extraction algorithm was used to obtain core herbs. Biological features of herbs were collected from public databases. At the same time, five supervised machine learning models were established based on the biological features of the coupled herbs. Finally, the K-nearest neighbor model was established as a screening model with an AUROC of 91.0%. Seventy coupled herbs for adjuvant treatment of hypertension complicated with coronary heart disease were obtained. It was found that the coupled herbs achieved the purpose of adjuvant therapy mainly by interfering with cytokines and regulating inflammatory and metabolic pathways. These results show that this model can integrate the molecular biological characteristics of herbs, preliminarily screen combinations of herbs, and provide ideas for explaining the value in clinical applications.