[Preprint] sc4D: spatio-temporal single-cell transcriptomics analysis through embedded optimal transport identifies joint glial response to Alzheimer’s disease pathology

In this work, we introduce sc4D, a spatio-temporal single-cell transcriptomics framework that jointly models cellular state, organization, and disease progression. By integrating autoencoder embeddings with optimal transport, sc4D reconstructs interpretable disease trajectories and predicts in silico perturbation response.

Applied to Alzheimer’s disease, sc4D recovers known pathological dynamics and improves our mechanistic understanding of glial behavior surrounding amyloid-beta plaques with spatial and temporal specificity. From the cellular programs active within these niches, we nominate therapeutic interventions predicted to restore protective, anti-inflammatory microglia that can improve plaque clearance.