Documentation

Data on VFB

Types of data and datasets that are available on Virtual Fly Brain.

Virtual Fly Brain integrates Drosophila melanogaster nervous system data from many laboratories into one queryable resource. Two things make that integration work, and both apply to every datatype below: images are registered onto a small set of standard templates so that data from different studies can be compared in the same coordinate space, and everything is classified with terms from the Drosophila Anatomy Ontology (DAO) so that a query for a cell type returns data from every study that reported it, whatever that study called it.

608,512
registered 3D images
136
imaging datasets
34.7 M
synaptic connections
3,290
single-cell clusters
28,772
anatomy ontology terms

The figures above come from the content report, which is regenerated against the production knowledge base after every data release.

What is here

DatatypeWhat it coversPage
Electron microscopyDense and sparse EM reconstructions of the adult brain, VNC and larval CNS, with neuron morphology and synaptic connectivityEM data
ConnectivityNeuron-to-neuron and neuron-to-region synaptic connections derived from the EM datasetsConnectivity data
Light microscopyDriver line expression patterns, split-GAL4 combinations, single-neuron and clonal images from confocal microscopyLM data
Single-cell transcriptomicsCell clusters and their gene expression from scRNAseq and snRNAseq studiesscRNAseq data
Templates and regionsThe reference brains and VNCs everything is registered to, and the painted neuropil domains in eachTemplates

VFB also hosts the public CATMAID instances for several community EM datasets, in some cases as the only remaining public copy after the original host went offline.

Where the data comes from, and where it goes

Most data on VFB is published data that VFB has registered, classified and indexed — VFB is an integrator, not usually the originator. Every dataset therefore carries its original publication, and every image links back to the resource it came from. Cite the original study when you use the data, and cite Court et al. (2023) for VFB itself.

Data can be reached three ways: through the 3D browser, programmatically through the APIsVFB_connect for Python, the knowledge graph, and the MCP server — and by download of individual images from their Term Info pages, in NRRD, Woolz, OBJ and SWC as available.

Licences

Licence terms differ per dataset and are set by the original producer, not by VFB. Where a dataset carries a licence it is shown on its Term Info page. Some older datasets predate the practice and carry none recorded; absence of a stated licence on VFB is not a grant of permission, so check with the original resource before redistributing.