Mapping Dendritic Spines Using 2D Two-Photon Laser Scanning
Neurons transform complex spatiotemporal patterns of synaptic input into structured sequences of action potentials that relay meaningful information. Excitatory inputs converging onto dendrites engage or interact with local non-linear regenerative events, adding a computational layer that expands the variety and complexity of input-output transformations. In conjunction with non-linear conductance-mediated mechanisms, the spatial arrangement of active synapses along dendrites strategically shapes this local computation. Mapping synaptic input organization is thus critical to fully uncover neuronal input-output function. Spine calcium imaging offers a direct functional readout of the location of active contacts, but such mapping requires access to spines distributed on intricate three-dimensional dendritic trees. We present a modular software pipeline for targeted imaging and analysis of dendrites using sequential two-dimensional scanning on standard two-photon microscopes. Designed to work with conventional two-photon microscopy setups, the method is fully compatible with ScanImage. It includes a pre-acquisition tool (ROIpy) and a post-acquisition analysis suite (Spyne). ROIpy generates dendrite-aligned region-of-interests for scattered depth-specific acquisition of neuronal arborizations. Spyne includes deep-learning modules for spine identification (using DeepD3) and within-spine calcium events detection (via a custom-classifier). This method is compatible with a range of experimental designs, including simultaneous two-photon imaging and patch-clamp recordings, as well as fully optical setups. The acquisition pipeline supports plane-by-plane imaging of the whole-arbor, targeted to specific compartments or to user-defined branches of interest. Our work provides a versatile strategy for targeted dendritic imaging using two-dimensional scanning multiphoton microscopy. While ROIpy allows adaptation to diverse experimental goals beyond synaptic mapping, Spyne provides an analysis strategy for functional mapping of active synapses at single-cell resolution, offering a basis for modeling how the spatial organization of synaptic inputs shapes dendritic integration..