lobemap

lobemap is a small napari viewer for Drosophila antennal lobe atlases.

https://github.com/user-attachments/assets/2b9711c9-3246-49c4-946b-a29fad0fce52

https://github.com/user-attachments/assets/d83978bd-f2de-48fe-9f0d-6aff19e3b663

It lets you:

• switch between atlases
• slice through 3D volumes
• show and hide glomeruli
• see glomerulus names, receptors, and sensilla
• compare reference maps and tables from different sources

Reference Table

The reference table is a cross-dataset index for glomeruli, receptors, sensilla, ligands, valence, driver lines, VFB IDs, and atlas coverage.

This preview shows selected columns only; it is not the full table schema.

Available columns: canonical_glomerulus, receptor_consensus, receptor_benton_2025, receptor_potter_task_2022, receptor_odour_scenes, receptor_door, sensillum_consensus, sensory_organ_consensus, sensillum_benton_2025, sensory_organ_benton_2025, sensillum_potter_task_2022, sensillum_odour_scenes, sensillum_door, sensory_organ_door, neuron_name_benton_2025, essential_coreceptor_benton_2025, co_receptor_door, orco_gal4_grabe_2015, orco_t2a_qf2, ir8a_t2a_qf2, ir76b_t2a_qf2, ir25a_t2a_qf2, key_agonists_benton_2025, sensory_scene_benton_2025, key_ligand, odour_scene, valence, fbbt_id, vfb_name, vfb_synonyms, projection_neuron_lines, projection_neuron_line_source, gh146_gal4, gh146_pn_female, gh146_pn_male, chat_gal4, chat_soma_count, chat_adpn, chat_lpn, chat_vpn, sensory_neuron_lines, present_grabe_2015, present_hemibrain, present_flywire, present_door_map, present_door_mappings, present_potter_task_2022, present_benton_2025, present_bates_schlegel_2020.

Open the full table or the name reconciliation table.

Requirements

• uv
• A desktop environment that can open napari / Qt windows

Python and package dependencies are managed by uv from pyproject.toml.

Start

From PyPI:

pip install lobemap
lobemap

From a source checkout:

uv sync
./lobemap

Choose an atlas from the dropdown in napari.

You can also open one atlas directly:

./lobemap --atlas flywire

Generated Data

Some viewers use derived cache files under datasets/*/data/derived/ for fast startup and slicing. These generated files are tracked so a fresh checkout can open the viewer without rebuilding caches. The PyPI package includes runtime source tables/volumes plus tracked data/derived/ and data/validation/ files needed by the installed viewer. Rebuild all generated visual data after source-data changes with:

uv run python scripts/regenerate_visual_data.py

The rebuild script uses the tracked source files and needs pdftoppm to render the Potter Task 2022 PDF preview.

What Is Included

• 3D atlas viewers for Grabe 2015, Bates Schlegel 2020, hemibrain, FlyWire, JRC2018Unisex, and Benton 2025
• 2D reference viewers for DoOR and Potter Task 2022
• Virtual Fly Brain and BANC resource lookup
• reference tables for glomerulus names, receptors, sensilla, and direct line labels

Demo

Data And Credit

This repo combines source data from several papers and public resources. Source files are kept in datasets/*/data/source/ folders when they can be shared. Paper PDFs are not tracked in git.

See docs/data-sources.md for the source list and links.

More

• How to use lobemap
• Data sources
• Reference tables

License

Code in this repo is released under the MIT License. Data files keep the rights and citation requirements of their original sources.