Network and Transcriptome-Guided Explainable Machine Learning for Identifying Disease-Specific Drug–Target Interactions in Alzheimer’s and Parkinson’s Disease
Vir Rastogi and Nirupma Singh
21/07/2026
Alzheimer’s disease and Parkinson’s disease are neurodegenerative disorders that pose significant global health challenges. Traditional drug discovery approaches are often time consuming and costly, emphasizing the need for more efficient computational strategies to identify potential drug targets. In this study, gene expression datasets associated with Alzheimer’s and Parkinson’s diseases were obtained from the Gene Expression Omnibus (GEO) database and analyzed using GEO2R to identify differentially expressed genes (DEGs). The identified DEGs were further investigated through gene interaction network analysis to examine their connectivity and pathway enrichment analysis to determine their significance. Features from their respective gene expression profiles and network characteristics were integrated to construct a dataset for predictive modeling between the two diseases. Two machine learning algorithms, Random Forest and XGBoost, were trained and evaluated using these features. Both models demonstrated strong predictive performance in identifying significant genes, with XGBoost showing slightly better results across most evaluation metrics. Feature importance analysis revealed that DEG significance and network connectivity were among the most influential predictors. The findings demonstrate that integrating gene expression data with features related to the network enhances the identification of relevant genes associated with neurodegenerative diseases. This computational model provides an efficient approach for prioritizing potential drug targets and may contribute to the development of improved strategies for the further research and the potential treatment of Alzheimer’s and Parkinson’s diseases.