Wired for Survival: Modeling Gene Networks to Identify Resilience and Neuron Subtypes in ALS via Graphical Optimization

Overview

Can neuron resilience be predicted as topological failure of cellular identity? Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease marked by TDP-43 aggregation and motor neuron (MN) death. While current research emphasizes degeneration, mechanisms of neuronal resilience and their potential as therapeutic targets remain unclear. This study engineered a topologically-aware graph neural network via protein-protein interactions to model ALS resilience through network stability. The model distinguished ALS with 85% accuracy and identified a 20-gene resilience hub enriched in chromatin remodeling (p=7.61e-04). This hub serves as a biomarker of early disease state determination and a scaffold for epigenetic therapy. scRNA-seq data (GSE226482) revealed MN subtypes exist on a state-specific spectrum as immature neurons show resilience, while mature, vulnerable MNs display end-stage markers. Type 7, expressing MNX1, is unique in being highly resilient despite a mature state. Spatial expression of these subtypes was confirmed via spinal cord tissue imaging. These neuron subtypes, identified as prognostic biomarkers, were validated in patient data: Type 7 was enriched in long-surviving samples (p=1.93e-03), and high resilience correlated with increased survival (p=0.044). Mathematical modeling using path-finding algorithms defined a “differentiation tax,” the cost to neuron identity under stress, and identified low-cost efficient resilience networks. By redefining ALS as topological failure, this research uncovers an epigenetic gene hub for resilient neuronal subtypes. This scalable, clinically-relevant framework identifies therapeutic targets to preserve neuron identity, providing hope for ALS patients via active neuroprotection.