New Imaging Technologies for Multiscale Connectomics
- Datum: 10.07.2025
- Uhrzeit: 14:00 - 15:00
- Vortragende(r): Aaron Kuan
- Yale School of Medicine
- Ort: MPI BI Martinsried
- Raum: MPI BI, T-Building, Large lecture hall
- Gastgeber: Lisa Fenk
- Kontakt: lisa.fenk@bi.mpg.de
- Rubrik: Vorträge

A major goal of neuroscience is understanding how neuronal circuits produce cognition. The growing field of connectomics promises to revolutionize this pursuit by providing comprehensive wiring diagrams of the brain. Connectomes of smaller insect nervous systems and isolated microcircuits in mammalian brains have been transformative for the field, but scaling connectomic imaging to brain-wide volumes in mammalian brains remains difficult with current volume electron microscopy (vEM) technologies.
Here we present multimodal imaging approach that leverages synchrotron-based X-ray Nano-Holography (XNH) and serial section electron tomography (ssET) to enable multiscale mapping of mammalian brains. Using intermediate voltage electron tomography, we show semi-thin brain sections (>500 nm) can be reconstructed at synapse resolution from a moderate number of projections (<30), and that successive sections can be stitched together to form large contiguous volumes. Imaging >10x thicker samples not only replaces the precarious process of ultra-thin sectioning with routine histological sectioning, but also vastly reduces the number of sections that need to be collected. We also show that non-destructive XNH can resolve individual myelinated axons in dense white matter within thick (~1 mm) tissue samples, enabling rapid mapping of long-range projections between brain regions. By stitching together XNH datasets, we map dense axonal projections across mesoscale regions of cortical white matter in the mouse brain. Moreover, we show that XNH and vEM are compatible on the very same samples, making it possible to efficiently obtain multiscale wiring diagrams characterizing the hierarchical structure of mammalian brains.