brainrender 0.3.4.1

Python scripts to use Allen Brain Map data for analysis and rendering

BrainRender

brainrender is a python package for the visualization of three dimensional anatomical data from mice brains registered to the Common Coordinate Framework (CCF) from the Allen Institute. Please check the user guide and the examples notebooks for more information on how to use BrainRender.

For more information about brainrender and how to use it, checkout the brainrender preprint.

Installation

To install brainrender, use an existing python environment or create one [with python < 3.8] and install with:

pip install brainrender

Description

brainrender can be used to download and visualize anatomical data from publicly available datasets as well as user generated content. Data from the Allen Mouse Brain Atlas and Allen Mouse Connectome projects can be used to visualise the three dimensional morphology of brain region and obtain a qualitative understanding of the I/O connective of a region of interest. Brainrender can also be used to visualise reconstructed neuronal morphologies, either by downloading them directly from MouseLight project from Janelia Research campus, or by rendering reconstructions obtained from elsewhere (e.g. from neuromorpho.org).

In addition, brainrender can be used to visualise data generated within your lab. Brainrender can be used to visualise the results of tracing experiments (e.g. location of the injection site and labelled cells) and to visualise the location of chronically implanted device (such as electrodes arrays or optic fibres).

Publicly available datasets

Allen Mouse Brain Atlas

Anatomical and projection data is downloaded from the Allen Brain Atlas reference atlas and connectivity atlas using the Allen API (© 2015 Allen Institute for Brain Science. Allen Brain Atlas API. Available from: brain-map.org/api/index.html) ([1], [2]).

Brain structures

Brain structures are the most fundamental element of any brainrender scene. You can use brainrender to select which brain regions you'd like to visualise, and how they should look (color, transparency...). Brainrender then takes care of downloading and rendering the 3D mesh for the selected regions.

Efferent projections (Streamlines)

Efferent anatomical projections from a region of interest as determined by local injections of an anterogradely transported virus (see Allen's connectivity atlas) can be rendered as 'streamlines'.

Streamlines reconstructions are made by https://neuroinformatics.nl using the mouse connectome data from Allen (see here for more details). Brainrender automatically downloads and renders streamlines data for a brain region of interest.

Brain structures	Afferent projections

Efferent projections	MouseLight neurons

Afferent projections (tractography)

The same data used to extract information about efferent projections, can be visualised in an alternative way to look at afferent projections to a region of interest. In brainrender this type of data visualisation is called tractography. Brainrender can be used to select the subset of experiment (injection of an anterograde virus expressing a fluorescent marker) which resulted in fluorescence in the region of interest from the >1000 experiments in the Allen Mouse Connectome project. The location of virus injection and the projection from that location to the region of itnerest is then displayed.

Volumetric projections

Brainrender can now use the model from [4] to visualise projection spatialised projections from one (or more) source brain region to one (or more) target brain region. By that we mean that you can see where in the target brainregion you have the strongest projections from the source region, as shown in this example:

MOs to STR	SSs to STR	SSp to STR

Mouselight and neurons morphology

To look at the morphology of single neurons in the mouse brain, brainrender can be used to download and render single neurons morphological data from Janelia's mouse light database (see the neurons browser). [3] These reconstructions are aligned to the Allen brain atlas, so the neurons morphology can be visualised along other types of data (e.g. efferent/afferent projections) in brainrender.

Brainrender can be used to visualise neuronal morphologies from other sources by loading .swc files directly. Additionally, brainrender can also be used to download morphologies from the Allen Cell Types project. A large number of neuronal morphologies can be found at neuromorpho.org.

The same approach can also be used to see which part of the source region projects to to the targetl

User generated content

Brainrender can be used to visualise data generated within individual labs, such as the results of a tracing experiments.

Injection site	Labelled neurons

This often requires that the raw data be registered to the Allen atlas, before being visualised in brainrender. This functionality is not supported in brainrender, however brainrender can be used to visualise data registered with other packages such as amap (github repo) and cellfinder (github repo).

You can also use brainrender to visualise the position of devices implanted in the brain (e.g. optic cannula). At the moment this can only be done by rendering colored cylinders at the location where the implant is, but we're happy to add more functionality, get in touch if you need anything specific!

Optic fibre	Electrodes array

Making figures

Brainrender's high quality renderings can be exported as .png images directly within brainrender, facilitating the creation of figure for scientific talks and publications.

MouseLight neuron	Labelled cells

Brainrender can also be used to create videos and animations. At the moment this functionality is not well supported, if you need to do something that brainrender can't yet do, please get in touch!

Behind the scenes

Brainrender was deisgned to be a powerful and flexible software for downloading and rendering neuroanatomical data while still being relatively easy to use (using brainrender requires minimal coding experience). This was achieved by:

1. handling the interaction with the API, SDK and databases services used to find and download the data behind the scenes, requiring minimal user input.
2. Using vtkplotter to handle the rendering. Vtkplotter (github repo) is a powerful rendering engine that produces high quality three dimensional rendering. Vtkplotter is also flexible, meaning that it can handle data provided in various file formats, allowing brainrender to render data from various sources.

Getting in touch

If you're unsure how to use brainrender, please start by having a look at the user guide and the examples notebooks. If you still have unanswered questions, please do not hesitate to get in touch (the easiest way is to open an issue on github).

For any bug report or feature request, please open an issue with a brief description of the matter. Although brainrender can already to much, we are always happy to add more functionality that could be useful for users. If you spot some feature that is missing, we'd love to hear about is so please get in touch!

Referencing Brain Render

If you found BrainRender useful and decided to include a rendering in your talks, posters or article, please acknowledge BrainRender's contribution by citing the brainrender preprint as:

Brainrender. A python based software for visualisation of neuroanatomical and morphological data.
Federico Claudi, Adam L. Tyson, Tiago Branco
bioRxiv 2020.02.23.961748; doi: https://doi.org/10.1101/2020.02.23.961748 

Similar tools

In R

cocoframer is an R library for interacting with the Allen's Mouse CCF github repository.

mouselightr package generates 3D CCF mouse brain plots, along with MouseLight neuron reconstructions github repository

References

• [1] Lein, E.S. et al. (2007) Genome-wide atlas of gene expression in the adult mouse brain, Nature 445: 168-176. doi:10.1038/nature05453
• [2] Oh, S.W. et al. (2014) A mesoscale connectome of the mouse brain, Nature 508: 207-214. doi:10.1038/nature13186
• [3] Winnubst, J. et al. (2019) Reconstruction of 1,000 Projection Neurons Reveals New Cell Types and Organization of Long-Range Connectivity in the Mouse Brain, Cell 179: 268-281
• [4] Knowx et al (2018). High-resolution data-driven model of the mouse connectome.