Hi all,
Making a new post following on from a previous post (as people’s roles may have changed in the meantime and I don’t want to cause them unnecessary noise)
I am working on maintaining the Brainrender tool. From what I understand in the post, Brainrender relied on a cache of some AIBS streamline data from experiment 174957972 hosted at https://neuroinformatics.nl/HBP/allen-connectivity-viewer/json/streamlines_174957972.json.gz .
However, this URL now returns 403 Forbidden (see this Brainrender issue for more details)
I am wondering
is there now a way to easily download the streamline data from Python (ideally again as a compressed JSON file)?
if not, would it be OK to manually download the data and host it on the BrainGlobe GIN server instead of the out-dated cache?
Any advice is appreciated!
1 Like
As a new user I couldn’t post more than two links in one post, so linking the previous post this follows on from in this comment.
Same question here, but I wonder whether one can use a post request to access this link: https://neuroinformatics.nl/HBP/allen-connectivity-viewer/streamline-downloader.html
Just bumping this, does anybody know the best way to retrieve the streamlines for a specific experiment?
I have repackaged this data from the server into a new format…a “precomputed” skeleton format that is compatible with neuroglancer and for which there is a python library that is available for reading.
here is a neuroglancer link: Neuroglancer
[click login at bottom of screen to unshorten link]
the precomputed source location from the link is: precomputed://gs://allen_neuroglancer_ccf/allen_mesoscale
I would use the python library cloud-volume to access this bucket and pull streamlines skeletons for each of the experiments. GitHub - seung-lab/cloud-volume: Read and write Neuroglancer datasets programmatically.
here is an example snippet…
import cloudvolume
Streamline has vertices in nanometers and edges as indices into the vertices, and if you perform connected components analysis of each graph you will find each streamline is a different component.
if you don’t want to use python you can get the steamline at this https address pattern.https://storage.googleapis.com/allen_neuroglancer_ccf/allen_mesoscale/skeleton/[EXPERIMENT_ID]
The resulting binary needs to be interpreted according to the precomputed format
# Skeleton representation of segmented objects
A skeleton representation of some or all segmented objects may be specified as a directory tree
consisting of the following files:
- `info` file in JSON format specifying the [metadata](#skeleton-info-json-file-format).
- For each segment ID for which there is a skeleton representation, a segment data file specifying
the [encoded skeleton](#encoded-skeleton-file-format) for a single segment.
The actual storage of the manifest and mesh fragment data depends on whether the unsharded or
[sharded](./sharded.md) format is used.
## Skeleton info JSON file format
The `info` file is a JSON-format text file. The root value must be a JSON object with the following
members:
- `"@type"`: Must be `"neuroglancer_skeletons"`.
- `"transform"`: JSON array of 12 numbers specifying a 4x3 homogeneous coordinate transform from the
"stored model" coordinate space to a "model" coordinate space. The "stored model" coordinate
show original
the reference info file for this dataset can be found herehttps://storage.googleapis.com/allen_neuroglancer_ccf/allen_mesoscale/skeleton/info
2 Likes
Thanks @fcollman , much appreciated.