Computational Drug Repurposing for Alzheimer’s Disease via Sheaf Theoretic Population-Scale Analysis of snRNA-Seq Data

Single-cell and single-nucleus RNA sequencing are used to reveal heterogeneity in cells, showing a growing potential for precision and personalized medicine. Nevertheless, sustainable drug discovery must be based on a population-level understanding of molecular mechanisms, which calls for a population-scale analysis of this data. This work introduces a sequential target-drug selection model for drug repurposing against Alzheimer's Disease (AD) targets inferred from snRNA-seq data of AD progression- involving hundreds of thousands of nuclei from multipatient and multiregional studies. We utilize Persistent Sheaf Laplacians (PSL) to facilitate a Protein-Protein Interaction (PPI) analysis inferred from disease related differential gene expression (DEG). We then use an ensemble of machine learning models to predict repurpose-able compounds. We screen the efficacy of different small compounds and further examine their central nervous system relevant ADMET properties, resulting in a list of potential molecular targets as well as pharmaceutical lead candidates for AD treatment.