<p>Multimodal imaging-based on single-cell genomics and spatial transcriptomics has shed new light on the taxonomy of genetically defined cell clusters in the mammalian brain. While transcriptomic approaches have revolutionized our ability to classify brain cells, their true value emerges when they are interpreted in conjunction with anatomical, physiological, and translational frameworks. Accordingly, significant progress has been made to elucidate relationships between gene expression, electrical and morphological properties of some of these clusters. This rapidly growing body of work shows not only that the cell cluster composition varies across brain regions but also evolves over time and changes during the progression of disease states like Alzheimer’s disease. Given this complexity, integrating transcriptomic, structural, and functional data is now becoming essential for drawing meaningful comparisons across studies. In this review, we summarize these findings and discuss how this knowledge base is shifting towards more integrative approaches, quickly challenging current ideas regarding the genetic, molecular, and cellular underpinnings of Alzheimer’s disease.</p> Graphical Abstract <p></p> <p>Multimodal cell type classification in Alzheimer’s disease: Now more than ever, classifying brain cells in Alzheimer’s disease needs to integrate information about gene expression, morphology, spatial context, connectivity, and electrophysiology. This multimodal approach ensures reproducibility across studies, revealing functional diversity, and tracking changes in cell cluster composition during disease progression, ultimately challenging our understanding of the molecular and cellular basis of Alzheimer’s disease.</p>