Optional Neuroglancer Layers

Each module’s Neuroglancer scene includes several optional layers in addition to the electron-microscopy imagery and the neuron segmentation. They appear as tabs across the top of the viewer; Left-Click a tab to toggle the layer on or off, and a tab to open its settings.

There are two kinds of optional layer:

Hosted layers load from a public data source (a precomputed:// path), so they can be added to any scene by pasting that path.
Local annotation layers (points or lines) are drawn directly into a specific scene and have no source path — the only way to reuse one is to drag it from a scene that already contains it.

The layer catalogue below lists every optional layer, what it shows, its source path (for hosted layers), and an example scene you can open to see it or drag it from. First, the two ways to add a layer to your own scene.

Adding a layer by dragging between windows

You can copy any layer — hosted or local — from one Neuroglancer window into another by dragging its layer tab. This is the only way to reuse a local annotation layer (the column boundaries, organelle markers, and KnownInputs), since those have no source path.

Open two Neuroglancer windows: the scene that has the layer, and the scene you want to add it to.
Left-Click and hold the layer’s tab in the top bar of the source window, then drag it onto the top bar of the destination window and release.

Dragging a layer tab from one Neuroglancer window into another.

Adding a hosted layer

To add a hosted (precomputed://) layer using its source path:

Click the + at the top of the viewer to create a new layer.
Paste the layer’s source path (from the catalogue) into the box that says “Enter a layer source…” and press Enter.
Neuroglancer picks the layer type automatically (imagery, segmentation, or mesh). Meshes appear once you toggle them on.

Adding a hosted layer by pasting its source path.

Layer catalogue

Reference meshes

3D surfaces that provide anatomical context around the neurons.

Layer	Shows	Source (paste to add)
Cortical layer boundaries (`layermeshes`)	The boundaries between cortical layers — L1/L2‑3, L2‑3/L4, L4/L5, L5/L6, and L6/white matter (six surfaces).	`precomputed://gs://allen-minnie-phase3/layermeshes_smooth_v2#type=MESH`
Cortical area boundaries (`areameshes`)	Rough boundaries between primary visual cortex (V1) and the higher-order visual areas (HVA).	`precomputed://gs://allen-minnie-phase3/areameshes#type=MESH`

How they were generated: meshed from the corresponding layer/area boundary segmentation for the MICrONS volume (Allen minnie-phase3).

Example scenes: Cortical Layers (layermeshes) · V1 and HVA Columns (layermeshes + areameshes).

Column boundary annotations

Layer	Shows	Source
`V1_cylinder`	The cylindrical boundary of the sampled V1 cortical column (blue points).	Local annotation — drag from the example scene
`HVA_cylinder`	The cylindrical boundary of the sampled higher-order visual area column (green points).	Local annotation — drag from the example scene

Example scene: V1 and HVA Columns.

Synapse layers

Layer	Shows	Source
Synapse predictions (`synapse_target_predictions_ssa`)	Synapses onto the selected cell, colored by predicted postsynaptic compartment: spine (coral), shaft (cyan), soma (gray). Filters to the selected cell; off by default in Module 4.	`gs://keith-dev/ng_precomputed_annotations/synapses/mat-v1300_w_types_idx-v1/`
`KnownInputs`	Input synapses onto the displayed neuron whose presynaptic partner is a proofread cell, colored by target compartment (spine / shaft / soma) via tags.	Local annotation — drag from the example scene

How they were generated: the synapse-prediction layer is a static, precomputed export of the v1300 synapse table with target-type predictions. KnownInputs was built from a CAVE query for input synapses whose presynaptic partner is proofread (proofreading_status_and_strategy), restricted to partners valid at the v1300 materialization version.

Example scenes: Spine Walkthrough (synapse_target_predictions_ssa) · Known Inputs (KnownInputs).

Organelle annotations (Module 2)

Small local point layers marking example locations of a subcellular structure, for identification practice: Nucleus, Mitochondria, Smooth ER, Golgi, Microtubules, Cilia, Myelinated axons, and Synapses. Each is a local annotation — drag it from the example scene.

Example scene: Organelles.

CCF context layers (Module 1 overview)

The Module 1 overview scene places the MICrONS volume inside the whole mouse brain using the Allen Common Coordinate Framework (CCF).

Layer	Shows	Source (paste to add)
`average_template_10_8bit`	The Allen CCF average-template mouse brain (grayscale imagery).	`precomputed://gs://allen_neuroglancer_ccf/average_template_10_8bit`
`AllenCCF` / `ccf_outline`	CCF brain-region meshes and the whole-brain outline.	`precomputed://gs://allen_neuroglancer_ccf/ccf_test1`
`seg35`	The MICrONS `minnie35` segmentation (a second, smaller EM volume).	`precomputed://gs://iarpa_microns/minnie/minnie35/seg`

Note

The CCF overview scene is being revised, so it is not yet published to the layer bucket. The sources above are listed for reference; their exact configuration may change in a future update.