Skip to content

Latest commit

 

History

History
133 lines (108 loc) · 5.78 KB

README.md

File metadata and controls

133 lines (108 loc) · 5.78 KB
Raw

Atlas Download Tools Logo

Search, download, and prepare atlas data.

Latest release DOI License
Black Isort Pydocstyle Pydocstyle Checked with mypy
Build status Documentation Status

Among different sources of data, Allen Brain Institute hosts a rich database of gene expression images, Nissl volumes, and annotation atlases. The Atlas-Download-Tools library can help you to download single section images and entire datasets, as well as the corresponding metadata. It can further pre-process the image data to place it in the standard reference space.

Installation

Installation from PyPI

The atldld package can be easily installed from PyPI.

pip install atldld

Installation from source

If you want to try the latest version, you can install from source.

pip install git+https://github.com/BlueBrain/Atlas-Download-Tools

Installation for development

If you want a dev install, you should install the latest version from source with all the extra requirements for running test.

git clone https://github.com/BlueBrain/Atlas-Download-Tools
cd Atlas-Download-Tools
pip install -e '.[dev]'

How to use the package

Atlas-Download-Tools can be used through a command line interface (CLI), as well as programmatically through a python API.

At present the CLI is rather limited, but we are working on adding the most useful functionality as soon as possible.

Using the CLI

All functionality can be accessed through the atldld command and its sub-commands. For example:

$ atldld

# Information about the version of atldld
$ atldld info version
$ atldld info version --help
# Information about the cache folder 
$ atldld info cache

One can also get dataset information through the CLI.

$ atldld dataset info $DATASET_ID
$ atldld dataset preview $DATASET_ID

To download a dataset use the atldld download command.

$ atldld download dataset $DATASET_ID

For further information please refer to the help part of the corresponding command.

Using the python API

The package atldld has several functionalities to download data from Allen Brain Institute:

  • One can find dataset IDs from a gene expression name:
from atldld.utils import get_experiment_list_from_gene
dataset_ids = get_experiment_list_from_gene("Pvalb", axis='sagittal')
  • One can download any dataset from a dataset ID:
from atldld.sync import DatasetDownloader
downloader = DatasetDownloader(dataset_id=DATASET_ID, **kwargs)
# One needs to fetch metadata before downloading a dataset.
downloader.fetch_metadata()
dataset = downloader.run()
image_id, section_number, img, img_exp, df = next(dataset)

Note that this functionality makes a simplifying assumption that the slices are perfectly parallel to one of the 3 axes.

  • One can obtain metadata of a dataset:
from atldld.utils import CommonQueries, get_3d
# The axis {'sagittal', 'coronal'}
axis = CommonQueries.get_axis(dataset_id=DATASET_ID)
# The reference space
ref_space = CommonQueries.get_reference_space(dataset_id=DATASET_ID)
# The 3d transformation of the dataset
matrix_3d = get_3d(dataset_id=DATASET_ID)

Funding & Acknowledgment

The development of this software was supported by funding to the Blue Brain Project, a research center of the École polytechnique fédérale de Lausanne (EPFL), from the Swiss government’s ETH Board of the Swiss Federal Institutes of Technology.

Copyright (c) 2021-2022 Blue Brain Project/EPFL