Set Up Walkthrough Data Path

All data read and written in this tutorial will be in the Precomputed data format and using cloud-volume library.

To run the walkthrough, you need to specify the storage location where the results and intermediary data will be written. Provided walkthrough commands expect the location to be specified by the CORGIE_WALKTHROUGH_PATH environment variable. CORGIE_WALKTHROUGH_PATH can point either to a local directory or to cloud storage, as described bellow.

Using Local Directory

To use a local directory, set the CORGIE_WALKTHROUGH_PATH environment variable to the chosen directory path with file:// protocol prefix. To visualize cloud-volume datasets stored in your local directory, use cloud-volume built-in viewing tools. Make sure that the specified directory exists. Example setup can be done through:

export CORGIE_DIR=${HOME}/corgie_data/walkthrough
mkdir -p $CORGIE_DIR
export CORGIE_WALKTHROUGH_PATH=file://${CORGIE_DIR}

Using Cloud Storage

To use cloud storage, set the CORGIE_WALKTHROUGH_PATH to the cloud path. Refer to cloud-volume Documentation for path formatting options. cloud-volume datasets stored in cloud storage can be visualized in Neuroglancer. The cloud storage bucket may need special configuration in order for the data to be accessible through Neuroglancer ([GCP instructions], [AWS instructions]).