22 Novembre 2026

Psicologia Punto One | Psicologia e Psicologi nel Web

Menu

Menu

SCHEDA SCIENTIFICA2026Altra ricerca

A curvilinear coordinate flatmap for visualizing hippocampal structure and development.

Ashwin Anirudha Bhandiwad, Fae Kronman, Josephine Liwang, Pan Gao, Cyrille Rossant, Daniel Birman, Luis Puelles, Song-Lin Ding, Xiangmin Xu, Lydia Ng, Yongsoo Kim, Tyler Mollenkopf

Imaging neuroscience (Cambridge, Mass.) · 2026

Abstract originale

The hippocampal formation is a highly curved and topographically complex forebrain structure. This complex geometry presents persistent challenges for analyzing subregional, laminar, and connectivity patterns. Here, we present a computational workflow that generates curvilinear-coordinate flatmaps from mouse Common Coordinate Framework (CCF) registered hippocampal and retrohippocampal regions by solving the Laplacian equation to derive geodesic streamlines. This transformation unfolds the hippocampus into a planar slab, bounded by the meningeal and ventricular surfaces, with the depth defined along the radial axis. We apply this transform to mouse CCF-registered image volumes, single neuron reconstructions, and point cloud data, including spatial transcriptomic and rabies tracing datasets, revealing topographic variations in the dorsoventral and radial axes that are obscured in the CCF coordinate space. As proof of principle, we use flatmaps to show connectivity loss in a mouse model of Alzheimer's disease and track postnatal development of microglial distribution in the hippocampus. This work provides an efficient and accessible resource for visualizing hippocampal organization across development and disease, offering new opportunities to interrogate the structure and function of this important brain region.

Parole chiave

AnimalsHippocampusMiceNeuronsAlzheimer DiseaseImage Processing, Computer-Assistedatlascoordinate systemdata resourceflatmapmousenonlinear coordinates
Leggi il testo integrale originale

Testo estratto dalla pubblicazione: sono esclusi immagini, media, supplementi e parti soggette a diritti distinti.

A curvilinear coordinate flatmap for visualizing hippocampal structure and development

Abstract

The hippocampal formation is a highly curved and topographically complex forebrain structure. This complex geometry presents persistent challenges for analyzing subregional, laminar, and connectivity patterns. Here, we present a computational workflow that generates curvilinear-coordinate flatmaps from mouse Common Coordinate Framework (CCF) registered hippocampal and retrohippocampal regions by solving the Laplacian equation to derive geodesic streamlines. This transformation unfolds the hippocampus into a planar slab, bounded by the meningeal and ventricular surfaces, with the depth defined along the radial axis. We apply this transform to mouse CCF-registered image volumes, single neuron reconstructions, and point cloud data, including spatial transcriptomic and rabies tracing datasets, revealing topographic variations in the dorsoventral and radial axes that are obscured in the CCF coordinate space. As proof of principle, we use flatmaps to show connectivity loss in a mouse model of Alzheimer’s disease and track postnatal development of microglial distribution in the hippocampus. This work provides an efficient and accessible resource for visualizing hippocampal organization across development and disease, offering new opportunities to interrogate the structure and function of this important brain region.

1Introduction

Topographic organization is a defining feature of neural circuits, linking spatial position to computation. In sensory and motor systems, topographic maps define how information is represented, integrated, and organized (Kaas, 1997). Understanding a region’s function, therefore, requires not only identifying cell types and their connections but also understanding their spatial and geometric context.

Advances in imaging, spatial transcriptomics, and single neuron morphological reconstruction have recently enabled the creation of large datasets that uncover cell type organization at a brain-wide scale (Kim et al., 2017; Liu et al., 2025; Wu et al., 2022; Yao et al., 2023). When integrated within a standardized atlas that share a common coordinate system, regional annotations, and consistent terminology, these datasets provide a powerful reference for understanding regional organization (Hawrylycz et al., 2023; Kleven, Gillespie, et al., 2023; Leergaard & Bjaalie, 2022).

Although whole-brain atlases exist for various species and developmental stages (Evans et al., 2012; Frey et al., 2011; Kleven, Bjerke, et al., 2023; Kronman et al., 2024; Liwang et al., 2025; Wang et al., 2020), the topographic organization of cell types and their connectivity remains difficult to assess. A key limitation to understanding topography is the use of a coordinate system with linear and orthogonal axes in the anterior-posterior, dorsal-ventral, and medial-lateral planes. While intuitive, this linear coordinate system fails to capture intrinsic curvature and nonlinear biological variation of brain structures, particularly across development (Bernhardt et al., 2022; Qiu et al., 2024; Thompson et al., 2014; Zhao et al., 2024). Therefore, resolving topographic organization requires a nonlinear coordinate system aligned with the primary axis of variation (Fig. 1A).

Geodesic flatmaps address this limitation by unfolding structures onto simpler manifolds while preserving intrinsic geometry (Fischl et al., 1999; Hahn & Duckworth, 2023; Hahn et al., 2021). Early implementations were qualitative and communicated three-dimensional data on paper, a two-dimensional medium (Sherk, 1992; Van Essen & Maunsell, 1980). More recently, quantitative flatmaps, created using streamlines, meshes, and physical models (Bolaños-Puchet et al., 2024; DeKraker et al., 2018; Diers et al., 2023; Jones et al., 2000; Ng et al., 2010; Swanson & Hahn, 2020; Wu et al., 2022), provide a data-driven approach to understanding a structure’s nonlinear geometry. Streamline-based flatmaps have been applied to visualize connectivity, density, and vascular organization in mouse cortex (Bennett et al., 2024; Harris et al., 2019; Peng et al., 2021; Wu et al., 2022).

The hippocampal formation (HPF), a key structure for learning, memory, and emotional control (Knierim, 2015), poses a unique challenge for geodesic unfolding because of its high curvature. Existing mouse hippocampal flatmaps remain largely qualitative without computational means to apply them to data assets (Swanson & Hahn, 2020). A promising approach for hippocampal unfolding uses Laplace’s equation to human hippocampi into a rectangular flatmap along the anterior-posterior, proximal-distal, and laminar axes (DeKraker et al., 2022, 2023, 2025). Unfolding the hippocampal ribbon into a standard rectangular shape permits comparison of signals across individuals and across species (Eichert et al., 2024). A similar approach in mouse would enable a common space for integrating multimodal data from mesoscale connectivity, single-neuron reconstructions, and spatial transcriptomic datasets.

Here, we introduce a resource for transforming mouse hippocampal volumes into a curvilinear coordinate system that remains topologically consistent across development. Using image volumes from the Allen CCFv3 mouse atlas and early postnatal development atlases (Liwang et al., 2025; Wang et al., 2020), we define geodesic hippocampal coordinate spaces for adult (P56) HPF and two developmental stages, postnatal day (P)4 and P14. This framework not only preserves laminar and longitudinal organization as in previous Laplacian-based maps but also preserves the native shape of the hippocampal formation and the dentate gyrus over development. The resulting flatmaps reveal hippocampal organization in an intuitive coordinate system, offering an accessible tool for exploring hippocampal topography and its developmental origins.

2Methods

2.1Ethics statement

All animal experiments were conducted according to the National Institutes of Health (NIH) guidelines for animal care and use and were authorized by the University of California, Irvine Institutional Animal Care and Use Committee and Institutional Biosafety Committee (Protocol number: AUP-25-150).

2.2Data sources

The average template and annotations for the adult P56 brain were published previously (Wang et al., 2020) and accessed through the Allen software development kit (Allen SDK). Mesoscale anterograde tracing data were accessed through the Allen Mouse Brain Connectivity portal (https://connectivity.brain-map.org), and the MERFISH spatial transcriptomic data were accessed through the Allen Brain Cell (ABC) atlas webpage (https://portal.brain-map.org/atlases-and-data/bkp/abc-atlas). Single neuron reconstructions were accessed from the Janelia MouseLight browser (https://ml-neuronbrowser.janelia.org) and the Mouse Projectome Atlas (https://mouse.digital-brain.cn/projectome/hipp). All neurons with soma in the HPF were selected for transformation.

Developmental templates and annotations for postnatal stages P4-P14 and registration transformations between adult and developmental stages were published previously (Liwang et al., 2025). CCFv3 annotations were mapped onto the P4 template by applying these transforms using Advanced Normalization Tools (ANTs) (Avants et al., 2011) using the AntsPy python library (https://github.com/ANTsX/ANTsPy/).

2.3Meningeal and ventricular surface annotation

Meningeal and ventricular surfaces were annotated manually using ImageJ. Each annotation volume was masked to generate a volume for the hippocampal formation (HPF). The hippocampal formation was split into the dentate gyrus and a combined structure of the Ammon’s horn, subiculum, presubiculum, postsubiculum, parasubiculum, and entorhinal cortex. The dentate gyrus is discontinuous at these developmental stages and was processed as a separate volume.

The annotation volume was filtered to include all descendants of the HPF (CA, PRE, SUB, ProS, PARA) and DG in the CCFv3 ontology (Wang et al., 2020). The filtered volume was binarized to generate a region mask, and the edges of the region mask were used as the bounding surface of the region of interest. Because the CCFv3 volume is symmetric, the following steps were performed on a single hemisphere to minimize computational costs.

Meningeal and ventricular surfaces were manually defined as the outermost voxels of the image volume. Surfaces were defined using developmental references and inherent template contrast features. For each section, edges were compared to the surfaces in the filtered annotation and annotated as either meningeal, ventricular, or unassigned, by manually assigning an integer value (1–3) to each surface voxel. The meningeal surface was defined as the surface derived from the meningeal layer (Fig. 2B). Where possible, the ventricular layer (e.g., L6 of entorhinal cortices, CA stratum oriens) was used as the ventricular surface. For the dentate gyrus, meningeal and ventricular boundaries were extended from the contiguous regions in CA3. Unassigned (“edge”) surfaces were defined as any surface that was neither meningeal nor ventricular. In hippocampal regions with no laminar subdivisions, the known laminar meningeal/ventricular boundaries were extended from layered structures by visual inspection to fill the continuous plane.

2.4Streamline computation

To find the shortest-distance path through the curved volume, the protocol used streamlines derived from a Laplacian gradient (Fig. 2A). Theory and details of streamline generation have been published previously (Jones et al., 2000; Ng et al., 2010; Wang et al., 2020). Streamline computation was performed using a forked implementation of the IBL atlas project (https://github.com/int-brain-lab/atlas). Streamlines were generated from the solution to Laplace’s equation. Briefly, the annotated edges and region volume combined and length-normalized orthogonal vectors to the meningeal surface were computed (Fig. 1E). Laplace’s equation was defined with Dirichlet boundary conditions at the meningeal surface and Neumann boundary conditions at the ventricular and edge surfaces. The partial differential equation was solved using numerical approximation using an iterative procedure in a GPU implementation. Following empirical data from cortical streamline generation, n = 10,000 iterations were used to arrive at convergence. The solution to Laplace’s equation was used to estimate a gradient from the meningeal to ventricular surfaces in a stepwise process with forward and backward differences depending on whether the voxel was within the volume or on the bounding surfaces. Curvilinear streamlines were computed by integrating the Laplacian gradient using an ordinary differential equation. Gradients were integrated from meningeal to ventricular surfaces using the Euler method. Finally, streamlines were rescaled so that each streamline consisted of n = 100 points sampled evenly along the meningeal-ventricular axis.

2.5Mapping CCF to curvilinear (flatmap) slab

To generate the flattened slab space from streamlines, a 2D manifold of the 3D meningeal surface was generated using Isomap (Tenenbaum et al., 2000). Isomap is a dimension reduction technique that preserves local nearest-neighbor geodesic (i.e. shortest path) distances along a surface. Isomap was performed using python’s SciPy library. A local neighborhood of n = 25 nearest neighbors was empirically determined as the neighborhood that preserved distances and balanced against high spatial clustering.

To reconstruct streamlines, the meningeal surface was subsampled from 3.88 x 105 to 1.2 x 105 voxels (3.2x reduction), preserving its shape while reducing memory load and computational costs. In cases where the meningeal surface was significantly smaller than the ventricular surface, such as in the dentate gyrus, the ventricular surface was treated as the streamline origin. This adjustment improved the representation of the volume because the streamline source had a larger surface area than the sink and reduced compression artifacts. After integrating these adjustments, the flatmap space was represented as the Isomap flattened source surface as the (x, y) coordinate and streamline depth (n = 100 steps) as the depth coordinate (Fig. 2D). All streamlines were oriented with the meningeal surface as the source—in the case of the dentate gyrus, the streamlines were inverted to follow the meningeal-to-ventricular convention.

Bilateral planar slabs were generated by orienting the HPF and DG surfaces along primarily the dorsoventral axis (Fig. 2G). Because HPF and DG were independently processed, surfaces were manually aligned so that the distal CA3 boundary and proximal DG boundary were roughly aligned. Moreover, independent Isomap transformation and streamline generation led to an overestimation of the DG surface, which was corrected by scaling the DG surface by the ratio of HPF to DG meningeal surface areas in CCF space. Finally, the contralateral hemisphere was assumed to be mirror-symmetric and flipped to generate a bilateral flatmap slab.

After slab generation, each voxel in CCF space was mapped to the closest voxel in flatmap space. For each voxel in the HPF and DG volumes, a 3D KDTree search was performed to find the closest voxel that contained a streamline point. In some cases, multiple CCF voxels mapped to the same point in flatmap space, indicating that the flatmap transformation was at a lower resolution compared to the original volume. Mappings between CCF and flatmap space were saved as a lookup table and used for mapping volumetric data. In parallel, a KDTree search was conducted in the flatmap space for each voxel to find the nearest nonzero neighbor—this table was subsequently used for nearest-neighbor interpolation for volumetric data.

2.6Rabies tracing and viral injections

To map neural circuit connectivity, mice were anesthetized with 1–2% isoflurane and positioned in a stereotaxic frame (Leica Angle Two™). Following scalp disinfection and craniotomy, viral injections were made into the dorsal CA1 (AP—1.9 mm, ML—1.4 mm, and DV—1.4 mm from bregma) using a glass pipette (20–30 μm tip) and delivered by a Picospritzer at a rate of 20–30 nl/min with a 10-ms pulse duration. The helper AAVs, AAV8-DIO-TC66T-2A-GFP-2A-oG (3.8 × 10¹³ GC/ml) and AAV1-CamKII 0.4.Cre.SV40 (1.9 × 10¹³ GC/ml) were co-injected (0.1 μl). After 17 days, EnvA-RV-SADΔG-tdTomato rabies virus (CNCM, UCI) was injected at the same site (5 × 108 infectious units/ml, 0.2 μl). Mice received 5 mg/kg carprofen subcutaneously to mitigate pain and inflammation and survived for 9 days to allow retrograde rabies labeling of presynaptic neurons. The mice were transcardially perfused with 20 ml of PBS and then with 40 ml of PBS containing 4% paraformaldehyde (PFA) before brain sample collection.

After perfusion, brains were processed and imaged using serial two photon tomography on a TissueCyte system with a previously published workflow (Ding et al., 2024). Cell segmentation was performed using Napari to identify putative starter cells and monosynaptic input neurons. Image volumes and cell coordinates were registered to the Allen CCFv3 template space and flatmap transformed using the CCF-to-flatmap mapping table.

3Results

To generate developmentally consistent flatmaps of the hippocampal formation (HPF), we began with its embryonic origin. The HPF begins as a sheet derived from the cortical hem and curves into its characteristic crescent shape by embryonic day 18 (E18) (Angevine, 1975; Bayer, 1980; Swanson & Hahn, 2020). By E21, the dentate gyrus emerges as a distinct cap at the end of the cornu ammonis (CA) (Fig. 1B). During this process, the original sheet differentiates into outer (meningeal) and inner (ventricular) surfaces that remain topologically consistent across development (Fig. 1B, C). Based on the prosomeric model (Puelles et al., 2013), we aligned slab depth along the radial dimension with meningeal and ventricular surfaces as boundaries (Fig. 1D, E). This framework provides a common geodesic space for comparing hippocampal organization across development.

Based on this scaffold, we implemented a curvilinear coordinate framework based on the streamline-based coordinate framework for the isocortex (Wang et al., 2020). We defined the meningeal and ventricular boundaries of the adult mouse hippocampal formation (HPF) using anatomical annotations from the Allen Mouse Brain CCFv3 (Wang et al., 2020). HPF subregions (CA fields, subiculum, presubiculum, postsubiculum, parasubiculum, and entorhinal cortex) and the dentate gyrus (DG) were isolated, with the DG processed separately because of its developmental discontinuity. Surface voxels were manually classified as meningeal, ventricular, or unassigned (Fig. 1E, Fig. 2A, B) to serve as boundary conditions for constructing a curvilinear coordinate system.

We solved Laplace’s equation between the meningeal and ventricular surfaces to obtain a smooth harmonic potential and integrated streamlines through this field to generate unique, non-intersecting trajectories linking the two surfaces (Fig. 2C, D). Each voxel was assigned to a single streamline, defining a continuous mapping between CCF and curvilinear coordinates.

To obtain a two-dimensional embedding, we projected the meningeal surface into a geodesic slab using Isomap, which preserves local structure by computing shortest paths along a neighborhood graph (Tenenbaum et al., 2000) (Fig. 2E). Nearest-neighbor correlations between CCF and streamline coordinates confirmed preservation of local spatial relationships (HPF: r = 0.70; DG: r = 0.17; Fig. 2F). Although the embedding introduced some compression due to streamline downsampling, local spatial relationships remained well preserved (Fig. 2G).

Flattening a convex structure to a planar slab introduces either distance or volume distortion (Milnor, 1969). We quantified local volumetric distortion by calculating the local Jacobian determinant of the transformation. Compression was most pronounced in ventral CA, particularly near the ventricular surface, and more evenly distributed near the meningeal surface (Supplementary Fig. S1), indicating that volumetric inferences across subregions should be interpreted with caution.

To assess biological interpretability, we examined how CCF (Cartesian x, y, z) axes were transformed into flatmap space. Gradients along the dorsal–ventral, anterior–posterior, and medial–lateral axes were mapped through the combined HPF and DG mask. The dorsal–ventral axis was largely preserved, whereas the other two axes underwent nonlinear transformations across the slab (Fig. 2H, I). Given the well-established functional and molecular gradients along the dorsal-ventral axis (also known as the septal-temporal or longitudinal axis) (Fanselow & Dong, 2010), we oriented flatmaps with the vertical axis aligned approximately to the CCF dorsal-ventral axis. This orientation helps highlight features along a well-understood and biologically relevant dimension.

3.1Application to mesoscale anterograde and single neuron reconstruction data

We next tested whether the curvilinear coordinate space captures known spatial connectivity patterns. Average template and CCFv3 structure labels (Wang et al., 2020) were first transformed into flatmap space (Fig. 3A).The hippocampal formation unfolded into a geodesic planar slab with well-preserved laminar organization.

We then applied the transformation to mesoscale anterograde connectivity data from the Allen Mouse Connectivity Atlas (Oh et al., 2014). Hippocampal connectivity exhibits characteristic topographic gradients along the dorsoventral and transverse axes (Fanselow & Dong, 2010; Strange et al., 2014; Witter et al., 1989), providing a benchmark for validating the flatmap coordinate space. Aggregated CA1 injection experiments (n = 8) revealed dense projections to the subiculum, prosubiculum, and in layers 5–6 of the entorhinal cortex, consistent with previous work (Amaral & Witter, 1989; Cenquizca & Swanson, 2007) (Fig. 3B). CA3 injections (n = 7 experiments) showed ipsilateral and contralateral projections to CA1, contralateral CA3, and DG polymorph layer, corroborating previous work (Ishizuka et al., 1990; Li et al., 1994; Swanson et al., 1978) (Fig. 3C). Notably, coordinate transformation preserved the relative absence of signal in the CA1 pyramidal layer, underscoring the accuracy of the transformation in maintaining anatomical detail.

We then evaluated whether flatmaps captured laminar specificity of entorhinal inputs to DG. Projections from the lateral entorhinal cortex (ENTl) terminate in the outer DG stratum moleculare, while medial entorhinal cortex (ENTm) projections terminate in the inner DG (van Groen et al., 2003). Averaged ENTl and ENTm projections localized to the expected layers after transformation (Fig. 3D). Laminar segregation was also present in the CA1 stratum lacunosum moleculare (slm), demonstrating that the transformation preserves laminar specificity alongside broader topographic structure.

Because mesoscale data aggregates fibers from multiple layers, cell types, and includes fibers of passage, we extended the curvilinear transformation to single neuron reconstructions. We transformed reconstructions from the Janelia MouseLight browser (Chandrashekar, 2017) and the mouse projectome atlas (Qiu et al., 2024), and selected 10,158 neurons with soma within the HPF. In flatmap coordinates, axons were aligned vertically along the meningeal-ventricular axis, reflecting radial geometric organization (Fig. 3E, F). Two example neurons from the lateral entorhinal cortex—one in layer 1 and one in layers 2/3—showed distinct projection zones in dorsal DG (Fig. 3G). At the population level, CA1 outputs also showed dorsoventral gradients. Dorsal CA1 neurons targeted subiculum and dorsal entorhinal cortex, whereas ventral CA1 neurons projected to ventral targets (Fig. 3H).

Together, these analyses show that curvilinear coordinates preserve laminar fidelity while revealing large-scale topographies in both mesoscale and single-cell datasets. Flatmaps, therefore, provide a unified framework for visualizing hippocampal connectivity across scales.

3.2Topographic organization of cell types in flatmap coordinate space

We next tested whether flatmaps could resolve the spatial organization of molecularly defined cell types. We transformed a registered spatial transcriptomic dataset comprising 812,084 HPF cells from multiple sources (Yao et al., 2023; Zhang et al., 2023), each mapped to a cell type ontology. As a first step, we transformed cells labeled by putative neurotransmitter identity, which showed uniform distribution across the HPF (Fig. 4A).

To examine topographic organization of glutamatergic and GABAergic cell subtypes, we focused on the dentate gyrus granule cell layer (DGsg), which contains the primary excitatory cell type within the dentate gyrus. Most glutamatergic cells were immature DG (DG-IMN Glut) granule cells, but two additional classes showed striking positional biases: Olfactory bulb-Cajal Retzius (OB-CR Glut) cells were enriched ventrally, while long-range projecting glutamatergic cells localized dorsally (Fig. 4B). Among GABAergic cells, medial ganglionic eminence-derived classes dominated overall, with Vip GABA cells enriched dorsally and caudal ganglionic eminence derived Lamp5 GABA cells confined ventrally (Fig. 4C). This pattern suggests that inhibitory interneuron distribution reflects developmental lineage along the dorsoventral axis.

Next, we focused on cell type laminar distribution in the entorhinal cortex along the meningeal-ventricular axis. Corticothalamic layer 6b (L6b CT ENT Glut) cells mapped as expected to deep layers 5–6, whereas intratelencephalic layer 2/3 (L2/3 IT ENT Glut) cells were present in superficial layers but unexpectedly were also present in deeper ventral ENTm (Fig. 4D). Flatmaps thus enable visualization of subtle laminar deviations that are difficult to discern in CCF coordinate space.

Finally, we assessed supertype-level organization of CA1 and CA3 pyramidal neurons. CA1-ProS Glut cells (six supertypes) showed a bimodal distribution along the dorsoventral axis, with the rare CA1-ProS Glut 6 subtype enriched ventrally (Fig. 4E). In contrast, CA3 Glut neurons were more evenly distributed, with a modest bias toward ventral CA3 (Fig. 4F). Together, these analyses demonstrate that curvilinear transformation captures both laminar and topographic organization of cell types across multiple levels of the ontology, linking transcriptomic identity to spatial patterning within the HPF.

3.3Application of flatmaps to retrograde labeling data in a model of Alzheimer’s disease

To test whether flatmaps can reveal spatial features of pathology, we applied the coordinate transformation to retrograde tracing data from 5xFAD mice, a transgenic model of Alzheimer’s disease. 5xFAD and wild-type (WT; C57BL/6J) littermates received injects in the dorsal CA1 with AAV helper viruses (AAV1-CamKII0.4.Cre.SV40 and AAV8-DIO-TC66T-2A-GFP). Seventeen days later, EnvA-SADΔG-tdTomato was injected in the same coordinates, and mice were perfused 9 days post-rabies injection, at approximately 4.3 months of age.

In WT animals, monosynaptic rabies virus mapped input neurons were detected in CA1, CA3, subicular complex, and entorhinal cortex (Fig. 5A). Flatmap transformation revealed a laminar arrangement of these inputs along the meningeal–ventricular axis, aligned with starter cell distribution. This organization was not apparent in CCF space. Comparing WT and 5xFAD mice with matched starter cell numbers (n = 17 WT, n = 14 5xFAD) (Fig. 5B, Supplementary Fig. S3) showed that both groups preserved laminar organization, but 5xFAD mice displayed fewer presynaptic neurons overall, especially in the entorhinal cortex and subiculum. Presynaptic CA1 neurons were also reduced and occupied a shorter span along the hippocampal longitudinal axis.

These findings demonstrate that coordinate transformation can capture subtle spatial changes in connectivity and propose flatmaps as a practical tool for visualizing differences in cell distribution and connectivity in models of neurodegeneration.

3.4Curvilinear coordinate space maps for spatiotemporal analyses across development

Finally, we asked whether flatmaps could track hippocampal topography across development. Early postnatal development involves inhibitory neuron migration, synaptogenesis, and changes in neuronal activity that shape hippocampal function (Cossart & Khazipov, 2022; Garaschuk et al., 1998; Olsen et al., 2023; Wong et al., 2005). Because our model is defined based on meningeal and ventricular boundaries, we expected curvilinear coordinates to yield a developmentally invariant latent space with preserved topology.

Annotated surfaces were registered sequentially from P56 to P14 and P4 and manually refined to correct minor registration errors. Curvilinear coordinates were computed independently for HPF and DG at each stage, and corresponding CCFv3 annotations were transformed for comparison and labeling (Fig. 5C). Despite differences in the shape and size of the HPF between P4 and P14, CCFv3 labels maintained a consistent structure, suggesting feasibility for cross-development analyses (Fig. 5D). However, subregion volumes showed nonlinear distortions between CCF and flatmap space. CA pyramidal layers preserved relative proportions, but regions with high curvature (e.g. ENTm, DG, prosubiculum) diverged across development (Supplementary Fig. S2). These distortions emphasize that distance-based rather than volumetric measures provide more reliable developmental comparisons.

To test whether flatmaps capture developmental dynamics, we transformed published Cx3cr1-eGFP volumes, which label microglia, at P4, P14, and P56 (Liwang et al., 2025). Microglia, which regulate synapse formation and cell death, proliferate and colonize brain regions with axis-specific density differences during early postnatal development (Barry-Carroll & Gomez-Nicola, 2024; Paolicelli et al., 2011). At P4, expression was enriched in ventral DG (ventricular side) and dorsal DG (meningeal side) (Fig. 5E, F). By P14, Cx3cr1 expression was more uniform dorsoventrally but shifted toward the meningeal surface in the ventral DG, consistent with radial migration. In the entorhinal cortex, Cx3cr1 expression increased from the ventricular to meningeal surfaces between P4 and P14. By P56, expression was uniformly distributed across the HPF, consistent with prior observations (Barry-Carroll et al., 2023; Jinno et al., 2007). These axis-biased shifts recapitulate known microglial migration and colonization patterns, demonstrating the value of flatmaps for developmental analysis.

4Discussion

In this work, we present a curvilinear coordinate system that “unfolds” the hippocampal and retrohippocampal formation into a planar slab with preserved nearest-neighbor distances. Using multiple data modalities, we show that this approach captures topographic variation that is obscured in CCF coordinates. While hippocampal organization across the longitudinal axis is well defined, this coordinate space extends analysis to the radial axis in a geodesic space. The resulting “2.5D” representation captures both surface and depth-related variation.

The data assets and transformation scripts are available at https://github.com/AllenInstitute/flatmap. Using this resource, a 10 μm CCFv3 registered adult mouse hippocampal volume (~4.1 x 107 voxels) encoded using 32-bit floating point format can be transformed into the flatmap space in ~4 min on a single CPU workstation with 5 GB RAM. Point coordinates (e.g., from spatial transcriptomic cells) can be mapped from Cartesian CCF coordinates to flatmap space at a rate of ~40 sec per 100 points. These speeds allow flatmapping to be added as a step in data processing pipelines, making it accessible for a wide range of spatial datasets.

Our model is based on a radial organization inspired by the concentric ring model of development (Puelles et al., 2019). Anchoring to meningeal and ventricular surfaces yields a developmentally grounded axis that supports comparison across postnatal development. This approach is similar to previous Laplacian-based methods that use the longitudinal, proximal-distal, and laminar axes to represent the human hippocampal formation and dentate gyrus as a rectangular slab (DeKraker et al., 2018, 2025). In contrast, by using Isomap, our approach preserves the intrinsic shape of the hippocampal formation and produces a topology consistent with previous physical models (Swanson & Hahn, 2020). This model also separates the dentate gyrus and transforms it independently of the hippocampal formation to reflect the distinct developmental and topological organization of the two interlinked regions.

Because these approaches share a common organization along the radial axis, the mouse flatmap space is compatible with previous human and primate flatmaps. Flatmaps could be used to align hippocampal structures across species using streamline-derived coordinates (DeKraker et al., 2018; Eichert et al., 2024) or for within-species comparisons, for example, between HPF and isocortex. Although primate and mouse hippocampi differ with respect to overall shape, orientation, and surface features (Garcia & Buffalo, 2020; Strange et al., 2014), their conserved relative subfield organization in the flatmap space suggests that transformation could facilitate cross-species comparisons and inferences from mouse disease models.

A key advantage of transforming data into a geodesic space is its ability to reveal spatial details that are not apparent in CCF space. For example, mesoscale connectivity revealed laminar segregation of DG inputs at the ENTm-ENTl boundary, a pattern consistent with previous observations but now quantifiable from volumetric data (Fig. 3D). Likewise, developmental flatmaps captured microglial migration along the meningeal–ventricular axis, consistent with histological reports of colonization gradients. These correspondences indicate that the transformation preserves biological signals and highlight its utility for hypothesis generation, for example, testing whether dendritic targeting by ENT inputs is systematically coupled to dorsoventral cell type variation.

One limitation of this approach is area and volume distortion, particularly in developmental flatmaps (Supplementary Fig. S1, Fig. 5D). Distortions arise from redundancy and downsampling when mapping to streamlines, particularly in smaller structures or those with high curvature, such as the ventral CA and meningeal DG. While distance-based analyses are robust, volumetric inferences require caution. Because distortions accumulate at larger volumes, applications where volumetric precision is required may require alternative approaches, such as mesh-based transforms (Bolaños-Puchet et al., 2023).

Boundary artifacts are another challenge, particularly at meningeal and ventricular edges and the CA3–DG interface (Fig. 4G). Mis-mapping can obscure critical pathways such as the perforant path, an important connectivity pathway between the entorhinal cortex and the DG (Amaral & Witter, 1989), and fragment single neuron morphologies when axons exit the HPF boundary. These artifacts highlight the need for refinement of surface definitions and expansion of this approach to more brain regions to preserve continuity. Addressing these boundary-related artifacts will be crucial for improving the accuracy of structural mapping, particularly in developmental and connectivity studies.

An important question moving forward is how flatmaps generalize across data modalities. This study highlights flatmap transformation for point data (e.g., MERFISH soma) and continuous fields (e.g., mesoscale connectivity). These data are differentially affected by distortions. Systematic evaluation of these differences, and integration with ground-truth histology, will be essential for defining reliable use cases of the transform.

Despite these caveats, the approach is generalizable to other labeled volumes. Although it requires additional manual segmentation of two boundary surfaces, the subsequent workflow can be applied across species and to other structures with complex topography in the brain and body. The method also integrates with emerging spatial omics approaches such as MERFISH (Hu et al., 2021; Townes & Engelhardt, 2023). Application of curvilinear coordinate transformation to spatial single-cell data also offers the opportunity to link spatial context to cell identity, connectivity, and dynamic processes such as development and disease. Flatmaps thus provide a computational framework for analyzing topologically complex structures, enabling comparative and developmental analyses that are otherwise obscured in CCF coordinate space.

Supplementary Material

References

Amaral , D. G. , & Witter , M. P. (1989 ). The three-dimensional organization of the hippocampal formation: A review of anatomical data . Neuroscience , 31 (3 ), 571 –591 . 10.1016/0306-4522(89)90424-7 2687721

Angevine , J. B. (1975 ). Development of the hippocampal region . In R. L. Isaacson & K. H. Pribram (Eds.), The hippocampus: Volume 1: Structure and development (pp. 61 –94 ). Springer US. 10.1007/978-1-4684-2976-3_4

Avants , B. B. , Tustison , N. J. , Song , G. , Cook , P. A. , Klein , A. , & Gee , J. C. (2011 ). A reproducible evaluation of ANTs similarity metric performance in brain image registration . NeuroImage , 54 (3 ), 2033 –2044 . 10.1016/j.neuroimage.2010.09.025 PMC3065962 20851191

Barry-Carroll , L. , & Gomez-Nicola , D. (2024 ). The molecular determinants of microglial developmental dynamics . Nature Reviews. Neuroscience , 25 (6 ), 414 –427 . 10.1038/s41583-024-00813-1 38658739

Barry-Carroll , L. , Greulich , P. , Marshall , A. R. , Riecken , K. , Fehse , B. , Askew , K. E. , Li , K. , Garaschuk , O. , Menassa , D. A. , & Gomez-Nicola , D. (2023 ). Microglia colonize the developing brain by clonal expansion of highly proliferative progenitors, following allometric scaling . Cell Reports , 42 (5 ), 112425 . 10.1016/j.celrep.2023.112425 37099424

Bayer , S. A. (1980 ). Development of the hippocampal region in the rat. II. Morphogenesis during embryonic and early postnatal life . The Journal of Comparative Neurology , 190 (1 ), 115 –134 . 10.1002/cne.901900108 7381049

Bennett , H. C. , Zhang , Q. , Wu , Y.-T. , Manjila , S. B. , Chon , U. , Shin , D. , Vanselow , D. J. , Pi , H.-J. , Drew , P. J. , & Kim , Y. (2024 ). Aging drives cerebrovascular network remodeling and functional changes in the mouse brain . Nature Communications , 15 (1 ), 6398 . 10.1038/s41467-024-50559-8 PMC11289283 39080289

Bernhardt , B. C. , Smallwood , J. , Keilholz , S. , & Margulies , D. S. (2022 ). Gradients in brain organization . NeuroImage , 251 (118987 ), 118987 . 10.1016/j.neuroimage.2022.118987 35151850

Bolaños-Puchet , S. , Teska , A. , Hernando , J. B. , Lu , H. , Romani , A. , Schürmann , F. , & Reimann , M. W. (2024 ). Enhancement of brain atlases with laminar coordinate systems: Flatmaps and barrel column annotations . Imaging Neuroscience (Cambridge, Mass.) , 2 (imag–2–00209 ), 1 –20 . 10.1162/imag_a_00209 PMC12272257 40800415

Bolaños-Puchet , S. , Teska , A. , & Reimann , M. W. (2023 ). Enhancement of brain atlases with region-specific coordinate systems: Flatmaps and barrel column annotations . bioRxiv (p. 2023.08.24.554204). 10.1101/2023.08.24.554204 PMC12272257 40800415

Cenquizca , L. A. , & Swanson , L. W. (2007 ). Spatial organization of direct hippocampal field CA1 axonal projections to the rest of the cerebral cortex . Brain Research Reviews , 56 (1 ), 1 –26 . 10.1016/j.brainresrev.2007.05.002 PMC2171036 17559940

Chandrashekar , J. (2017 ). MouseLight neuron browser - collection [Data set] . 10.25378/janelia.c.3924088.v1

Cossart , R. , & Khazipov , R. (2022 ). How development sculpts hippocampal circuits and function . Physiological Reviews , 102 (1 ), 343 –378 . 10.1152/physrev.00044.2020 34280053

DeKraker , J. , Cabalo , D. G. , Royer , J. , Ngo , A. , Khan , A. R. , Karat , B. G. , Benkarim , O. , Rodriguez-Cruces , R. , Frauscher , B. , Pana , R. , Hansen , J. Y. , Misic , B. , Valk , S. L. , Lau , J. C. , Kirschner , M. , Bernasconi , A. , Bernasconi , N. , Muenzing , S. E. A. , Axer , M. , … Bernhardt , B. C . (2025 ). HippoMaps: Multiscale cartography of human hippocampal organization . Nature Methods , 22 (10 ), 2211 –2222 . 10.1038/s41592-025-02783-3 PMC12510884 41034610

DeKraker , J. , Ferko , K. M. , Lau , J. C. , Köhler , S. , & Khan , A. R. (2018 ). Unfolding the hippocampus: An intrinsic coordinate system for subfield segmentations and quantitative mapping . NeuroImage , 167 , 408 –418 . 10.1016/j.neuroimage.2017.11.054 29175494

DeKraker , J. , Haast , R. A. M. , Yousif , M. D. , Karat , B. , Lau , J. C. , Köhler , S. , & Khan , A. R. (2022 ). Automated hippocampal unfolding for morphometry and subfield segmentation with HippUnfold . eLife , 11 , e77945 . 10.7554/elife.77945 PMC9831605 36519725

DeKraker , J. , Palomero-Gallagher , N. , Kedo , O. , Ladbon-Bernasconi , N. , Muenzing , S. E. A. , Axer , M. , Amunts , K. , Khan , A. R. , Bernhardt , B. , & Evans , A. C. (2023 ). Evaluation of surface-based hippocampal registration using ground-truth subfield definitions . eLife , 12, RP88404. 10.7554/elife.88404.2 PMC10642966 37956092

Diers , K. , Baumeister , H. , Jessen , F. , Düzel , E. , Berron , D. , & Reuter , M. (2023 ). An automated, geometry-based method for hippocampal shape and thickness analysis . NeuroImage , 276 , 120182 . 10.1016/j.neuroimage.2023.120182 PMC12811610 37230208

Ding , Y. , Huang , Y. , Gao , P. , Thai , A. , Chilaparasetti , A. N. , Gopi , M. , Xu , X. , & Li , C. (2024 ). Brain image data processing using collaborative data workflows on Texera . Frontiers in Neural Circuits , 18 , 1398884 . 10.3389/fncir.2024.1398884 PMC11266044 39050044

Eichert , N. , DeKraker , J. , Howard , A. F. D. , Huszar , I. N. , Zhu , S. , Sallet , J. , Miller , K. L. , Mars , R. B. , Jbabdi , S. , & Bernhardt , B. C. (2024 ). Hippocampal connectivity patterns echo macroscale cortical evolution in the primate brain . Nature Communications , 15 (1 ), 5963 . 10.1038/s41467-024-49823-8 PMC11252401 39013855

Evans , A. C. , Janke , A. L. , Collins , D. L. , & Baillet , S. (2012 ). Brain templates and atlases . NeuroImage , 62 (2 ), 911 –922 . 10.1016/j.neuroimage.2012.01.024 22248580

Fanselow , M. S. , & Dong , H.-W . (2010 ). Are the dorsal and ventral hippocampus functionally distinct structures? Neuron , 65 (1 ), 7 –19 . 10.1016/j.neuron.2009.11.031 PMC2822727 20152109

Fischl , B. , Sereno , M. I. , & Dale , A. M. (1999 ). Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system . NeuroImage , 9 (2 ), 195 –207 . 10.1006/nimg.1998.0396 9931269

Frey , S. , Pandya , D. N. , Chakravarty , M. M. , Bailey , L. , Petrides , M. , & Collins , D. L. (2011 ). An MRI based average macaque monkey stereotaxic atlas and space (MNI monkey space) . NeuroImage , 55 (4 ), 1435 –1442 . 10.1016/j.neuroimage.2011.01.040 21256229

Garaschuk , O. , Hanse , E. , & Konnerth , A. (1998 ). Developmental profile and synaptic origin of early network oscillations in the CA1 region of rat neonatal hippocampus . The Journal of Physiology , 507 (Pt 1 ), 219 –236 . 10.1111/j.1469-7793.1998.219bu.x PMC2230780 9490842

Garcia , A. D. , & Buffalo , E. A. (2020 ). Anatomy and function of the primate entorhinal cortex . Annual Review of Vision Science , 6 (1 ), 411 –432 . 10.1146/annurev-vision-030320-041115 PMC12889097 32580662

Hahn , J. D. , & Duckworth , C. (2023 ). A brain flatmap data visualization tool for mouse, rat, and human . The Journal of Comparative Neurology , 531 (10 ), 1008 –1016 . 10.1002/cne.25478 37002855

Hahn , J. D. , Swanson , L. W. , Bowman , I. , Foster , N. N. , Zingg , B. , Bienkowski , M. S. , Hintiryan , H. , & Dong , H.-W . (2021 ). An open access mouse brain flatmap and upgraded rat and human brain flatmaps based on current reference atlases . The Journal of Comparative Neurology , 529 (3 ), 576 –594 . 10.1002/cne.24966 PMC7721992 32511750

Harris , J. A. , Mihalas , S. , Hirokawa , K. E. , Whitesell , J. D. , Choi , H. , Bernard , A. , Bohn , P. , Caldejon , S. , Casal , L. , Cho , A. , Feiner , A. , Feng , D. , Gaudreault , N. , Gerfen , C. R. , Graddis , N. , Groblewski , P. A. , Henry , A. M. , Ho , A. , Howard , R. , … Zeng , H . (2019 ). Hierarchical organization of cortical and thalamic connectivity . Nature , 575 (7781 ), 195 –202 . 10.1038/s41586-019-1716-z PMC8433044 31666704

Hawrylycz , M. , Martone , M. E. , Ascoli , G. A. , Bjaalie , J. G. , Dong , H.-W. , Ghosh , S. S. , Gillis , J. , Hertzano , R. , Haynor , D. R. , Hof , P. R. , Kim , Y. , Lein , E. , Liu , Y. , Miller , J. A. , Mitra , P. P. , Mukamel , E. , Ng , L. , Osumi-Sutherland , D. , Peng , H. , … Zingg , B . (2023 ). A guide to the BRAIN Initiative Cell Census Network data ecosystem . PLoS Biology , 21 (6 ), e3002133 . 10.1371/journal.pbio.3002133 PMC10313015 37390046

Hu , J. , Li , X. , Coleman , K. , Schroeder , A. , Ma , N. , Irwin , D. J. , Lee , E. B. , Shinohara , R. T. , & Li , M. (2021 ). SpaGCN: Integrating gene expression, spatial location and histology to identify spatial domains and spatially variable genes by graph convolutional network . Nature Methods , 18 (11 ), 1342 –1351 . 10.1038/s41592-021-01255-8 34711970

Ishizuka , N. , Weber , J. , & Amaral , D. G. (1990 ). Organization of intrahippocampal projections originating from CA3 pyramidal cells in the rat . The Journal of Comparative Neurology , 295 (4 ), 580 –623 . 10.1002/cne.902950407 2358523

Jinno , S. , Fleischer , F. , Eckel , S. , Schmidt , V. , & Kosaka , T. (2007 ). Spatial arrangement of microglia in the mouse hippocampus: A stereological study in comparison with astrocytes . Glia , 55 (13 ), 1334 –1347 . 10.1002/glia.20552 17647290

Jones , S. E. , Buchbinder , B. R. , & Aharon , I. (2000 ). Three-dimensional mapping of cortical thickness using Laplace’s equation . Human Brain Mapping , 11 (1 ), 12 –32 . 10.1002/1097-0193(200009)11:1<12::aid-hbm20>3.0.co;2-k PMC6872107 10997850

Kaas , J. H. (1997 ). Topographic maps are fundamental to sensory processing . Brain Research Bulletin , 44 (2 ), 107 –112 . 10.1016/s0361-9230(97)00094-4 9292198

Kim , Y. , Yang , G. R. , Pradhan , K. , Venkataraju , K. U. , Bota , M. , García Del Molino , L. C. , Fitzgerald , G. , Ram , K. , He , M. , Levine , J. M. , Mitra , P. , Huang , Z. J. , Wang , X.-J. , & Osten , P. (2017 ). Brain-wide maps reveal stereotyped cell-type-based cortical architecture and subcortical sexual dimorphism . Cell , 171 (2 ), 456 -469.e22 . 10.1016/j.cell.2017.09.020 PMC5870827 28985566

Kleven , H. , Bjerke , I. E. , Clascá , F. , Groenewegen , H. J. , Bjaalie , J. G. , & Leergaard , T. B. (2023 ). Waxholm Space atlas of the rat brain: A 3D atlas supporting data analysis and integration . Nature Methods , 20 (11 ), 1822 –1829 . 10.1038/s41592-023-02034-3 PMC10630136 37783883

Kleven , H. , Gillespie , T. H. , Zehl , L. , Dickscheid , T. , Bjaalie , J. G. , Martone , M. E. , & Leergaard , T. B. (2023 ). AtOM, an ontology model to standardize use of brain atlases in tools, workflows, and data infrastructures . Scientific Data , 10 (1 ), 486 . 10.1038/s41597-023-02389-4 PMC10372146 37495585

Knierim , J. J. (2015 ). The hippocampus . Current Biology , 25 (23 ), R1116 –R1121 . 10.1016/j.cub.2015.10.049 26654366

Kronman , F. N. , Liwang , J. K. , Betty , R. , Vanselow , D. J. , Wu , Y.-T. , Tustison , N. J. , Bhandiwad , A. , Manjila , S. B. , Minteer , J. A. , Shin , D. , Lee , C. H. , Patil , R. , Duda , J. T. , Xue , J. , Lin , Y. , Cheng , K. C. , Puelles , L. , Gee , J. C. , Zhang , J. , … Kim , Y . (2024 ). Developmental mouse brain common coordinate framework . Nature Communications , 15 (1 ), 9072 . 10.1038/s41467-024-53254-w PMC11494176 39433760

Leergaard , T. B. , & Bjaalie , J. G. (2022 ). Atlas-based data integration for mapping the connections and architecture of the brain . Science (New York, N.Y.) , 378 (6619 ), 488 –492 . 10.1126/science.abq2594 36378966

Li , X. G. , Somogyi , P. , Ylinen , A. , & Buzsáki , G. (1994 ). The hippocampal CA3 network: An in vivo intracellular labeling study . The Journal of Comparative Neurology , 339 (2 ), 181 –208 . 10.1002/cne.903390204 8300905

Liu , L. , Yun , Z. , Manubens-Gil , L. , Chen , H. , Xiong , F. , Dong , H. , Zeng , H. , Hawrylycz , M. , Ascoli , G. A. , & Peng , H. (2025 ). Connectivity of single neurons classifies cell subtypes in mouse brains . Nature Methods , 22 (4 ), 861 –873 . 10.1038/s41592-025-02621-6 PMC11978518 40119176

Liwang , J. K. , Kronman , F. N. , Pi , H.-J. , Wu , Y.-T. , Vanselow , D. J. , Manjila , S. B. , Parmaksiz , D. , Shin , D. , Ben-Simon , Y. , Taormina , M. , Way , S. W. , Zeng , H. , Tasic , B. , Ng , L. , & Kim , Y. (2025 ). epDevAtlas: Mapping GABAergic cells and microglia in the early postnatal mouse brain . Nature Communications , 16 (1 ), 9538 . 10.1038/s41467-025-64549-x PMC12572312 41162351

Milnor , J. (1969 ). A problem in cartography . The American Mathematical Monthly: The Official Journal of the Mathematical Association of America , 76 (10 ), 1101 –1112 . 10.1080/00029890.1969.12000424

Ng , L. , Lau , C. , Sunkin , S. M. , Bernard , A. , Chakravarty , M. M. , Lein , E. S. , Jones , A. R. , & Hawrylycz , M. (2010 ). Surface-based mapping of gene expression and probabilistic expression maps in the mouse cortex . Methods , 50 (2 ), 55 –62 . 10.1016/j.ymeth.2009.10.001 19818854

Oh , S. W. , Harris , J. A. , Ng , L. , Winslow , B. , Cain , N. , Mihalas , S. , Wang , Q. , Lau , C. , Kuan , L. , Henry , A. M. , Mortrud , M. T. , Ouellette , B. , Nguyen , T. N. , Sorensen , S. A. , Slaughterbeck , C. R. , Wakeman , W. , Li , Y. , Feng , D. , Ho , A. , … Zeng , H . (2014 ). A mesoscale connectome of the mouse brain . Nature , 508 (7495 ), 207 –214 . 10.1038/nature13186 PMC5102064 24695228

Olsen , L. C. , Galler , M. , Witter , M. P. , Saetrom , P. , & O’Reilly , K. C. (2023 ). Transcriptional development of the hippocampus and the dorsal-intermediate-ventral axis in rats . Hippocampus , 33 (9 ), 1028 –1047 . 10.1002/hipo.23549 37280038

Paolicelli , R. C. , Bolasco , G. , Pagani , F. , Maggi , L. , Scianni , M. , Panzanelli , P. , Giustetto , M. , Ferreira , T. A. , Guiducci , E. , Dumas , L. , Ragozzino , D. , & Gross , C. T. (2011 ). Synaptic pruning by microglia is necessary for normal brain development . Science (New York, N.Y.) , 333 (6048 ), 1456 –1458 . 10.1126/science.1202529 21778362

Peng , H. , Xie , P. , Liu , L. , Kuang , X. , Wang , Y. , Qu , L. , Gong , H. , Jiang , S. , Li , A. , Ruan , Z. , Ding , L. , Yao , Z. , Chen , C. , Chen , M. , Daigle , T. L. , Dalley , R. , Ding , Z. , Duan , Y. , Feiner , A. , … Zeng , H . (2021 ). Morphological diversity of single neurons in molecularly defined cell types . Nature , 598 (7879 ), 174 –181 . 10.1038/s41586-021-03941-1 PMC8494643 34616072

Puelles , L. , Alonso , A. , García-Calero , E. , & Martínez-de-la-Torre , M. (2019 ). Concentric ring topology of mammalian cortical sectors and relevance for patterning studies . The Journal of Comparative Neurology , 527 (10 ), 1731 –1752 . 10.1002/cne.24650 30737959

Puelles , L. , Harrison , M. , Paxinos , G. , & Watson , C. (2013 ). A developmental ontology for the mammalian brain based on the prosomeric model . Trends in Neurosciences , 36 (10 ), 570 –578 . 10.1016/j.tins.2013.06.004 23871546

Qiu , S. , Hu , Y. , Huang , Y. , Gao , T. , Wang , X. , Wang , D. , Ren , B. , Shi , X. , Chen , Y. , Wang , X. , Wang , D. , Han , L. , Liang , Y. , Liu , D. , Liu , Q. , Deng , L. , Chen , Z. , Zhan , L. , Chen , T. , … Xu , C . (2024 ). Whole-brain spatial organization of hippocampal single-neuron projectomes . Science (New York, N.Y.) , 383 (6682 ), eadj9198 . 10.1126/science.adj9198 38300992

Sherk , H. (1992 ). Flattening the cerebral cortex by computer . Journal of Neuroscience Methods , 41 (3 ), 255 –267 . 10.1016/0165-0270(92)90090-z 1513183

Strange , B. A. , Witter , M. P. , Lein , E. S. , & Moser , E. I. (2014 ). Functional organization of the hippocampal longitudinal axis . Nature Reviews. Neuroscience , 15 (10 ), 655 –669 . 10.1038/nrn3785 25234264

Swanson , L. W. , & Hahn , J. D. (2020 ). A qualitative solution with quantitative potential for the mouse hippocampal cortex flatmap problem . Proceedings of the National Academy of Sciences of the United States of America , 117 (6 ), 3220 –3231 . 10.1073/pnas.1918907117 PMC7022173 31988117

Swanson , L. W. , Wyss , J. M. , & Cowan , W. M. (1978 ). An autoradiographic study of the organization of intrahippocampal association pathways in the rat . The Journal of Comparative Neurology , 181 (4 ), 681 –715 . 10.1002/cne.901810402 690280

Tenenbaum , J. B. , de Silva , V. , & Langford , J. C. (2000 ). A global geometric framework for nonlinear dimensionality reduction . Science , 290 (5500 ), 2319 –2323 . 10.1126/science.290.5500.2319 11125149

Thompson , C. L. , Ng , L. , Menon , V. , Martinez , S. , Lee , C.-K. , Glattfelder , K. , Sunkin , S. M. , Henry , A. , Lau , C. , Dang , C. , Garcia-Lopez , R. , Martinez-Ferre , A. , Pombero , A. , Rubenstein , J. L. R. , Wakeman , W. B. , Hohmann , J. , Dee , N. , Sodt , A. J. , Young , R. , … Jones , A. R . (2014 ). A high-resolution spatiotemporal atlas of gene expression of the developing mouse brain . Neuron , 83 (2 ), 309 –323 . 10.1016/j.neuron.2014.05.033 PMC4319559 24952961

Townes , F. W. , & Engelhardt , B. E. (2023 ). Nonnegative spatial factorization applied to spatial genomics . Nature Methods , 20 (2 ), 229 –238 . 10.1038/s41592-022-01687-w PMC9911348 36587187

Van Essen , D. C. , & Maunsell , J. H. (1980 ). Two-dimensional maps of the cerebral cortex . The Journal of Comparative Neurology , 191 (2 ), 255 –281 . 10.1002/cne.901910208 7410593

van Groen , T. , Miettinen , P. , & Kadish , I. (2003 ). The entorhinal cortex of the mouse: Organization of the projection to the hippocampal formation . Hippocampus , 13 (1 ), 133 –149 . 10.1002/hipo.10037 12625464

Wang , Q. , Ding , S.-L. , Li , Y. , Royall , J. , Feng , D. , Lesnar , P. , Graddis , N. , Naeemi , M. , Facer , B. , Ho , A. , Dolbeare , T. , Blanchard , B. , Dee , N. , Wakeman , W. , Hirokawa , K. E. , Szafer , A. , Sunkin , S. M. , Oh , S. W. , Bernard , A. , … Ng , L . (2020 ). The Allen mouse brain common coordinate framework: A 3D reference atlas . Cell , 181 (4 ), 936 -953.e20 . 10.1016/j.cell.2020.04.007 PMC8152789 32386544

Witter , M. P. , Groenewegen , H. J. , Lopes da Silva , F. H. , & Lohman , A. H. (1989 ). Functional organization of the extrinsic and intrinsic circuitry of the parahippocampal region . Progress in Neurobiology , 33 (3 ), 161 –253 . 10.1016/0301-0082(89)90009-9 2682783

Wong , T. , Zhang , X. L. , Asl , M. N. , Wu , C. P. , Carlen , P. L. , & Zhang , L. (2005 ). Postnatal development of intrinsic GABAergic rhythms in mouse hippocampus . Neuroscience , 134 (1 ), 107 –120 . 10.1016/j.neuroscience.2005.04.019 15961234

Wu , Y.-T. , Bennett , H. C. , Chon , U. , Vanselow , D. J. , Zhang , Q. , Muñoz-Castañeda , R. , Cheng , K. C. , Osten , P. , Drew , P. J. , & Kim , Y. (2022 ). Quantitative relationship between cerebrovascular network and neuronal cell types in mice . Cell Reports , 39 (12 ), 110978 . 10.1016/j.celrep.2022.110978 PMC9271215 35732133

Yao , Z. , van Velthoven , C. T. J. , Kunst , M. , Zhang , M. , McMillen , D. , Lee , C. , Jung , W. , Goldy , J. , Abdelhak , A. , Aitken , M. , Baker , K. , Baker , P. , Barkan , E. , Bertagnolli , D. , Bhandiwad , A. , Bielstein , C. , Bishwakarma , P. , Campos , J. , Carey , D. , … Zeng , H . (2023 ). A high-resolution transcriptomic and spatial atlas of cell types in the whole mouse brain . Nature , 624 (7991 ), 317 –332 . 10.1038/s41586-023-06812-z PMC10719114 38092916

Zhang , M. , Pan , X. , Jung , W. , Halpern , A. R. , Eichhorn , S. W. , Lei , Z. , Cohen , L. , Smith , K. A. , Tasic , B. , Yao , Z. , Zeng , H. , & Zhuang , X. (2023 ). Molecularly defined and spatially resolved cell atlas of the whole mouse brain . Nature , 624 (7991 ), 343 –354 . 10.1038/s41586-023-06808-9 PMC10719103 38092912

Zhao , K. , Wang , D. , Wang , D. , Chen , P. , Wei , Y. , Tu , L. , Chen , Y. , Tang , Y. , Yao , H. , Zhou , B. , Lu , J. , Wang , P. , Liao , Z. , Chen , Y. , Han , Y. , Zhang , X. , & Liu , Y. (2024 ). Macroscale connectome topographical structure reveals the biomechanisms of brain dysfunction in Alzheimer’s disease . Science Advances , 10 (41 ), eado8837 . 10.1126/sciadv.ado8837 PMC11809497 39392880

Come citare questo studio

Ashwin Anirudha Bhandiwad, Fae Kronman, Josephine Liwang, Pan Gao, Cyrille Rossant, Daniel Birman, Luis Puelles, Song-Lin Ding, Xiangmin Xu, Lydia Ng, Yongsoo Kim, Tyler Mollenkopf. (2026). A curvilinear coordinate flatmap for visualizing hippocampal structure and development.. Imaging neuroscience (Cambridge, Mass.). DOI: 10.1162/imag.a.1429

Consulta la pubblicazione originale ↗