build 2026-09-04 20:32
Or paste x, y, z into the x field — it splits automatically.
Great for learning what a cell type looks like (a new random example loads each click), or for quickly finding a cell that still needs a first identification.
On identify, the cell (root ID) and its nucleus are loaded automatically, so both segmentation layers switch on even if unchecked.
Grubb, “Ultrastructure of precapillary sphincters and the neurovascular unit,” Vascular Biology 5, VB-23-0011 (2023).
Grubb, “Ultrastructure of the brain waste-clearance pathway,” bioRxiv preprint (2023), not yet peer-reviewed.
Grubb, Chaddha, Lippincott-Schwartz, Ott & Mughal, “Pericyte and Endothelial Primary Cilia and Centrioles have Disparate Organization Across the Brain Microvasculature,” bioRxiv preprint (2025), not yet peer-reviewed.
Morris, Foster, Sutherland & Grubb, “Microglia contact cerebral vasculature through gaps between astrocyte endfeet,” Journal of Cerebral Blood Flow & Metabolism 44(12), 1472–1486 (2024).
Pericyte subtype naming (ensheathing / mesh / thin-strand) and the arteriole–capillary transition zone: Grant, Hartmann, Underly, Berthiaume, Bhat & Shih, “Organizational hierarchy and structural diversity of microvascular pericytes in adult mouse cortex,” J Cereb Blood Flow Metab 39(3), 411–425 (2019).
Pericyte subtype criteria used by this tool’s guided identification, tested against serial block-face SEM (circumferential coverage, cross-sectional area shared with the endothelium, peg-and-socket density, and ECM extensions unique to ensheathing pericytes): Abdelazim et al., “Pericyte heterogeneity identified by 3D ultrastructural analysis of the microvessel wall,” Frontiers in Physiology 13:1016382 (2022). Note that these subtypes grade into one another rather than forming discrete classes — see the Grubb 2023 review above, which argues the transitional mural cells are better described simply as contractile vs non-contractile, and which is why the guided tree always offers a plain “Pericyte” option.
Cell type predictions & dataset: The MICrONS Consortium, “Functional connectomics spanning multiple areas of mouse visual cortex,” Nature 640, 435–447 (2025).
Perisomatic cell-type classification (neuronal + non-neuronal — incl. contributions from Leila Elabbady): Elabbady et al., “Perisomatic ultrastructure efficiently classifies cells in mouse cortex,” Nature 640, 478–486 (2025).
Brain ultrastructure overview: “Brain Ultrastructure: Putting the Pieces Together,” PMC7930431.
Neurovascular volume EM: “Public Volume Electron Microscopy Data: An Essential Resource to Study the Brain Microvasculature,” Frontiers in Cell and Developmental Biology (2022).
Neurotransmitter prediction from synaptic EM: “Neurotransmitter classification from EM images at synaptic sites,” Cell (2024).
Synapse type I/II framework: Gray, E.G., “Axo-somatic and axo-dendritic synapses of the cerebral cortex: an electron microscope study,” J. Anat. 93:420–433 (1959) (foundational; no open-access link).
Cortical layer depth boundaries used for the layer estimate (rescaled proportionally to each location's locally measured pia-to-white-matter thickness, rather than used as fixed absolute microns): Ledderose et al., “Layer 1 of somatosensory cortex: an important site for input to a tiny cortical compartment,” Cerebral Cortex (2023). Proportional rescaling is used because both this source and MICrONS are chemically (aldehyde) fixed tissue, which is documented to shrink cortex linearly by ~15–18%: Korogod, Petersen & Knott, “Ultrastructural analysis of adult mouse neocortex comparing aldehyde perfusion with cryo fixation,” eLife 4:e05793 (2015).
Fine-grained m-type classification (dendritic-morphology / synaptic-target clusters): Schneider-Mizell et al., “Inhibitory specificity from a connectomic census of mouse visual cortex,” Nature 640, 448–458 (2025).
Excitatory dendritic-morphology map (subtypes as a continuum with L5/L6 exceptions) — basis for the guided excitatory-subtype step: Weis et al., “An unsupervised map of excitatory neuron dendritic morphology in the mouse visual cortex,” Nature Communications 16, 3361 (2025).
Layer 5 thick-tufted (ET) excitatory neurons — thick apical trunk, prominent L1 tuft, thick myelinated axon: Bodor et al., “The synaptic architecture of layer 5 thick tufted excitatory neurons in the visual cortex of mice,” Nature Neuroscience (2025).
Sst (Martinotti) inhibitory subtypes — distal-dendrite targeting, axon myelination & output: Gamlin et al., “Connectomics of predicted Sst transcriptomic types in mouse visual cortex,” Nature 640, 497–505 (2025).
Excitatory projection-neuron classes (IT/ET/NP/CT) and general layer circuitry: Harris & Shepherd, “The neocortical circuit: themes and variations,” Nature Neuroscience 18, 170–181 (2015).
Inhibitory interneuron subtype function (basket/Martinotti/bipolar/neurogliaform cells): Tremblay, Lee & Rudy, “GABAergic Interneurons in the Neocortex: From Cellular Properties to Circuits,” Neuron 91, 260–292 (2016).
Microglia function: Nimmerjahn, Kirchhoff & Helmchen, “Resting Microglial Cells Are Highly Dynamic Surveillants of Brain Parenchyma in Vivo,” Science 308, 1314–1318 (2005).
Oligodendrocyte function: Bergles & Richardson, “Oligodendrocyte Development and Plasticity,” Cold Spring Harbor Perspectives in Biology 8, a020453 (2016).
OPC & neurovascular unit function: Pfeiffer, “Reciprocal Interactions between Oligodendrocyte Precursor Cells and the Neurovascular Unit in Health and Disease,” Cells 11, 1954 (2022).
Astrocyte & neurovascular coupling function: Takahashi, “Metabolic Contribution and Cerebral Blood Flow Regulation by Astrocytes in the Neurovascular Unit,” Cells 11, 813 (2022).
Pericyte function: Girolamo et al., “Central Nervous System Pericytes Contribute to Health and Disease,” Cells 11, 1707 (2022).
Leptomeninges anatomy & function: Zedde & Pascarella, “Leptomeninges: Anatomy, Mechanisms of Disease and Neuroimaging,” Neurology International 17, 203 (2025).
Leptomeningeal cell subtypes (arachnoid barrier cell tight junctions; shared, location-defined fibroblast subtypes) — basis for the guided leptomeninges step: Pietilä et al., “Molecular anatomy of adult mouse leptomeninges,” Neuron 111, 3745–3764.e7 (2023).
White matter function: Fields, “White matter in learning, cognition and psychiatric disorders,” Trends in Neurosciences 31, 361–370 (2008).
Nearest-nucleus, IDs & predictions from nucleus_detection_v0 and aibs_metamodel_celltypes_V661 (minnie65 only).
Live statistics from the MICrONS cell classifier / πJump project — every chart is computed fresh from the current "Master cell list" (and community reports) each time you load or refresh this page, so a correction submitted through πJump shows up here too. The physical geometry behind a cell (its cortical layer, its nearest neighbours) can't change from a report and stays fixed, but which cell TYPE that geometry counts toward always reflects the latest identification (see each panel's note).
Every identified cell, best-available identity (own-verified > community majority vote > MICrONS prediction). Unclassified cells excluded — see the pie chart and per-layer breakdown for those. Cell types grouped and ordered as Excitatory neurons → Inhibitory neurons → Glia → Vascular cells → Immune & perivascular cells → Leptomeninges, same as πJump's own dropdown. LIVE — recomputed from "Master cell list" on every load/refresh.
Main-nucleus population only. LIVE.
Layer estimated by inverse-distance-weighted interpolation of hand-traced pia / white-matter surfaces (Ledderose et al. 2023 boundaries, rescaled to local cortical thickness). Unclassified cells excluded — see the chart below for those. Layer is fixed geometry; cell-type counts are LIVE.
Split out from the chart above so identified cell types aren't crowded by the (much larger) unclassified population. Same layer-estimation method. LIVE.
Only computable where a cell has been specifically checked for a cilium — MICrONS publishes no organelle-level data, so "checked" is that subset, not the whole dataset (this denominator can't grow from an ordinary identification report; only an explicit organelle check adds to it). Which TYPE each check counts toward is LIVE.
Base-to-tip Euclidean distance, µm. Gray dots = individual cells (jittered); coloured bar = median. LIVE — each measurement's cell-type bucket reflects the current identity, and any organelle report not yet folded into "Master cell list" is included too.
Gray dots = individual cells (jittered); coloured bar = median. LIVE — same as the cilium-length chart above.
All 4 bars LIVE from "Master cell list"'s current identity source, main nuclei only.
LIVE, from "Master cell list"’s precomputed nucleus volumes, identified cells only (unclassified cells excluded — see “Identified vs. unclassified” above for totals). Compare within one category (also used by the two volume charts below):
LIVE, grows with every signed-in “Compute volume” click, identified cells only (unclassified cells excluded). Uses the same category selector as the nucleus-volume chart above.
LIVE, grows with every signed-in “Compute volume” click, identified cells only (unclassified cells excluded). Uses the same category selector as the two volume charts above.
Which identified cell types most often appear among the 5 nearest physical neighbours of the selected type. Neighbour geometry is fixed; neighbour cell types are LIVE.
Real synaptic partner types (MICrONS connectomic synapse data), sampled per source cell type. This is a SEPARATE, occasional offline step (not something πJump computes live) — see the note below if it hasn't been generated yet.
Average per sampled cell — same offline data as the connectivity chart above.
One dot per cell-type pair — dot size scales with how many pairs of that type have been checked. LIVE, default-settings searches only.
Every report type (identifications, discrepancies, merged-nucleus splits, organelle reports, root-ID proposals/votes, computed volumes), cumulative by day. LIVE.
This tool was made by Søren Grubb to make identifying brain cortical cells in electron microscopy easy. It was built using Claude AI. Predictions from the MICrONS consortium are included, but you can help identify cells that lack a prediction, or revise ones already made — identification discrepancies and newly identified cells are shared back with the MICrONS consortium.