brain-indexer

brain-indexer is a library for efficient spatial queries on large datasets (TBs and more). It is currently based on boost::rtree.

It provides a high level Python API for indexes of simple geometric shapes (boxes, spheres and cylinders) and the required functionalities to create indexes of synapses and morphologies.

Installation

From PyPI

A version of brain-indexer without processing on multiple nodes is available on PyPI and can easily be installed with:

pip install brain-indexer

From source

brain-indexer requires Boost with a minimum version of 1.79.0, but preferably 1.80.0 or newer. If your system does not provide such a version, one can install a more recent one into /opt/boost/ as follows:

wget -qO- https://boostorg.jfrog.io/artifactory/main/release/1.85.0/source/boost_1_85_0.tar.bz2 | tar xjf -
cd boost_1_85_0
./bootstrap.sh
./b2 --prefix=/opt/boost --with-serialization --with-filesystem --with-test install

When using a custom Boost installation like this, it is imperative to set the following environment variables to make sure that all libraries are found:

export SKBUILD_CMAKE_DEFINE="CMAKE_INSTALL_RPATH_USE_LINK_PATH=ON"
export CMAKE_PREFIX_PATH=/opt/boost

If multi-node processing via MPI is required, the following system dependencies have to be installed:

sudo apt-get install -y libopenmpi-dev

Then the installation of brain-indexer can proceed as usual, either with MPI support:

gh repo clone BlueBrain/brain-indexer
pip install "./brain-indexer[mpi]"

or without support:

gh repo clone BlueBrain/brain-indexer
pip install ./brain-indexer/

Where to start

We provide two Jupyter notebooks with a hands-on introduction to brain-indexer:

A basic introduction in basic_tutorial.ipynb
More complex use-cases in advanced_tutorial.ipynb

Examples

More examples on how to use brain-indexer are available in the examples folder:

segment_index.py: simple indexing and querying of a segment index
synapses_index.py: simple indexing and querying of a synapse index
segment_index_sonata.py: indexing and querying of a segment index using SONATA files
segment_multi_index_sonata.py: indexing and querying of a segment multi-index using SONATA files
synapse_multi_index_sonata.py: indexing and querying of a synapse multi-index using SONATA files

Also, the tests folder contains some tests that double also as examples on how to use brain-indexer.

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.