A transcriptomic axis aligns with in vivo functional dynamics in hippocampal inhibitory circuits

Linking molecular identity to function in vivo at single-cell resolution remains an outstanding challenge in neuroscience. Here, we bridge this gap in the mouse hippocampus with an end-to-end pipeline of cell-resolved two-photon imaging and spatial transcriptomics. CA1 interneurons exhibiting heterogeneous physiological responses during a virtual-reality navigation task were post hoc clustered by gene expression into 5 GABAergic subclasses and 14 types. Physiological responses of individual cells aligned with a transcriptomic axis, and a classifier trained on physiological features alone recovered the same ordered organization. Our approach establishes a direct, scalable framework for linking in vivo circuit dynamics to constituent cell identity, revealing a transcriptomic axis that encompasses the structural and functional diversity of hippocampal inhibitory neurons.