Major depressive disorder (MDD) is a severe mental disorder characterized by persistent low mood. Patients experience prolonged emotional distress, which not only affects daily life but also negatively impacts self-esteem. Typical symptoms include loss of interest, self-blame, decreased appetite, lack of concentration, and suicidal tendencies. The disease has a high prevalence, is prone to recurrence, and can lead to high rates of disability and mortality. According to the latest clinical treatment statistics, approximately 280 million people worldwide suffer from major depressive disorder, and the number of affected individuals continues to rise annually. It is worth noting that in statistics ranking disabilities and deaths collectively, major depressive disorder ranks only after diseases such as heart disease, stroke, and HIV/AIDS. Moreover, this disease can accompany patients for many years. According to the World Health Organization’s forecast, by 2030, major depressive disorder may become one of the major diseases posing a significant threat to human health. Currently, the pathogenesis of major depressive disorder is not fully understood, and the scientific community has various hypotheses, such as the monoamine hypothesis, receptor hypothesis, and immune inflammation hypothesis. In recent years, an increasing number of studies have shown that immune regulation abnormalities are closely related to the pathogenesis of major depressive disorder. Although the immune inflammation hypothesis is not the mainstream direction of current research on anti-major depressive drugs, it remains important in the field of major depressive disorder mechanisms. This hypothesis suggests that overactivation of the immune system can produce inflammatory cytokines, which play an important role in the development of major depressive disorder.
Traditional Chinese medicine for the treatment of mood diseases has the advantages of multiple targets and minimal adverse reactions, with broad application prospects in the treatment of major depressive disorder and comorbidities. Arisaematis Rhizoma is a perennial medicinal plant in the family Araceae, containing various components such as flavonoids, lignans, terpenes, and alkaloids. Studies have found that a natural small molecule polyphenolic compound extracted from the dried rhizomes of Arisaematis Rhizoma has multiple pharmacological activities such as anti-inflammatory, antioxidant, and anti-tumor effects, with low toxicity and minimal side effects. This study aims to explore the anti-inflammatory pathways and potential targets of Arisaematis Rhizoma. Network pharmacology is an emerging discipline based on systems biology and network science theories, aiming to provide a more systematic and comprehensive theoretical basis for drug development by analyzing the comprehensive network structure of molecules, genes, proteins, signaling pathways, etc. in biological systems to reveal complex interactions between drugs and organisms. With the development of high-throughput sequencing technology, bioinformatics-based big data mining analysis has been widely used to explore various disease-related biomarkers. Bioinformatics is conducive to exploring potential therapeutic targets and related pathways of drugs. Supervised learning in the machine learning domain involves inputting features and outcome events into a classification model simultaneously, aiming to fit the relationship between them. This intelligent predictive and decision-making method plays an important role in screening potential therapeutic targets. Molecular docking is a key method in computer-aided drug research used to simulate and predict the interaction between molecules. This technique mainly predicts the binding mode and affinity between receptors and drug molecules by analyzing the structure of receptors and drug molecules through theoretical simulation. Compared to molecular docking, molecular dynamics simulation is a more complex and comprehensive computer simulation method used to study the motion and interactions of molecules at the time and spatial scales. Molecular dynamics simulations can simulate the molecular trajectory and provide detailed information on the molecular structure, dynamic behavior, and thermodynamic properties. In the field of pharmacy, molecular dynamics is commonly used in pharmacokinetic research for drug development, screening, and prediction.
This study uses bioinformatics methods combined with various machine learning models to identify potential anti-inflammatory targets for treating major depressive disorder and conducts molecular docking, drug similarity evaluation, and molecular dynamics simulations to screen key active components of Arisaematis Rhizoma. The aim is to provide new research ideas for the clinical treatment of major depressive disorder and scientific basis for the development of treatment plans and related therapeutic drugs for major depressive disorder.
1. Methodology
1.1 Data Source
Retrieve the major depressive disorder datasets (GSE76826, GSE98793) from the Gene Expression Omnibus (GEO) database.
1.2 Gene Set Variation Analysis (GSVA) and Immune Infiltration
Search for inflammation-related genes in the Gene Card disease database using “inflammation” and “inflammatory” as keywords, with a relevance score >8. Calculate the inflammation score in the GSE76826 dataset using GSVA.
1.3 Screening of Arisaematis Rhizoma Active Components and Prediction of Potential Target Networks
Search for active components of Arisaematis Rhizoma in the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database and screen based on oral bioavailability (OB) ≥30% and drug-likeness (DL) ≥0.18. Predict potential target networks for active components using the Swiss Target Prediction database.
1.4 Exploration and Prediction of Major Depressive Disorder-Related Targets
Use the Limma package in R software to select differentially expressed genes between the major depressive disorder group and the control group. Apply a threshold of |log2FC| ≥0.5 and adjusted P value (Padj) <0.05.
1.5 Association between Arisaematis Rhizoma and Major Depressive Disorder-Related Targets
Take the intersection of Arisaematis Rhizoma potential target networks and major depressive disorder-related targets, display them in a 2×2 contingency table, and analyze their association using Fisher’s exact test.
1.6 Weighted Gene Co-Expression Network Analysis (WGCNA)
Apply the WGCNA package in R to analyze the GSE76826 dataset, identify gene co-expression modules related to immune inflammation, and focus on the Turquoise module.
1.7 Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) Enrichment Analysis
Conduct GO and KEGG enrichment analysis for eight anti-inflammatory candidate targets to reveal their functional and pathway information.
1.8 Machine Learning for Selecting Potential Anti-Inflammatory Targets
Build eight machine learning models to screen potential anti-inflammatory targets and evaluate model performance using measures such as accuracy, precision, sensitivity, specificity, and area under the curve (AUC).
1.9 Molecular Docking and Drug Similarity Evaluation
Conduct molecular docking to predict the binding affinity between active components of Arisaematis Rhizoma and target proteins. Evaluate drug-likeness properties of active components.
1.10 Molecular Dynamics Simulation
Perform a 50 ns molecular dynamics simulation using GROMACS to validate the results of molecular docking and drug similarity evaluation for Arisaematis Rhizoma’s active components.
1.11 Statistical Analysis
Use R software for data analysis, apply the Wilcoxon rank-sum test for statistical analysis, and consider a significance level of P <0.05.
2. Results
2.1 Validation of the Inflammation Hypothesis and Exploration of Immune Biomarkers
