Statistical and Machine Learning Approaches to Single-Nucleus Transcriptomics of the Mouse Spinal Cord in Health and Disease

The spinal cord undergoes transcriptional changes during aging and in motor neuron diseases such as amyotrophic lateral sclerosis (ALS) and spinal and bulbar muscular atrophy (SBMA). To investigate these processes, we analyzed single-nucleus RNA sequencing (snRNA-seq) data from mouse spinal cord across healthy, aged, and diseased states. Using established preprocessing pipelines and clustering methods, we characterized major cell types and explored their transcriptional alterations, applying statistical and explainable machine learning (ML) methods. These approaches combine predictive modeling with biological interpretability, enabling the identification of gene signatures that distinguish conditions. Our results highlight how explainable ML can bridge the gap between computational prediction and biological insight, establishing a framework for integrating single-nucleus transcriptomics with interpretable ML, offering new perspectives on the mechanisms of spinal cord aging and disease.