iblviewer.mouse_brain

Classes

AllenAtlasExt

This overwrites the constructor of AllenAtlas that is not designed to be used for the public, that is people outside of IBL.

IBLAtlasModel

IBL Atlas is a wrapper for the Allen Atlas with added features.

MouseBrainViewer

This is your entry point to International Brain Laboratory data visualization
class AllenAtlasExt(res_um=25, brainmap='Allen', scaling=array([1, 1, 1]), image_file_path=None, label_file_path=None)[source]

Bases: ibllib.atlas.atlas.AllenAtlas

This overwrites the constructor of AllenAtlas that is not designed to be used for the public, that is people outside of IBL. Typically, you’d want to display the Allen volume data in this viewer and perform additional tasks (such as loading your own extra data) with other libraries. Dev note: I’m forced to copy and modify the whole constructor in this case.

Instantiates an atlas.BrainAtlas corresponding to the Allen CCF at the given resolution using the IBL Bregma and coordinate system

remap_atlas(local_file_path, regions=None, ibregma=None)[source]

Remap the atlas label into a usable volume

lateralize(label, regions=None, ibregma=None)[source]

Breaks the symmetry in regions labels in the Allen Mouse Atlas where the id of a region in the left hemisphere is the same as the same region in the right hemisphere. But if we want to map recordings to the brain, we need separate ids.

Parameters

label – Segmented volume

Returns

Modified volume

class IBLAtlasModel(origin: numpy.ndarray = array([5739, 5400, 332]), atlas: Optional[ibllib.atlas.atlas.AllenAtlas] = None, atlas_lut: Optional[iblviewer.volume.LUTModel] = None, atlas_mapping_ids: Optional[list] = None, ibl_back_end: bool = True, atlas_volume: Optional[iblviewer.volume.VolumeModel] = None, dwi_volume: Optional[iblviewer.volume.VolumeModel] = None)[source]

Bases: object

IBL Atlas is a wrapper for the Allen Atlas with added features. The volume is also modified such that it fits functional needs.

origin: numpy.ndarray = array([5739, 5400,  332])
atlas: ibllib.atlas.atlas.AllenAtlas = None
atlas_lut: iblviewer.volume.LUTModel = None
atlas_mapping_ids: list = None
ibl_back_end: bool = True
IBL_TRANSPOSE = (1, 0, 2)
atlas_volume: iblviewer.volume.VolumeModel = None
dwi_volume: iblviewer.volume.VolumeModel = None
initialize(resolution=None, ibl_back_end=True)[source]

Get Allen Atlas metadata (.csv file that is copied into ibllib.atlas) and volume files

Parameters

resolution – Volume resolution in microns

get_atlas_model(atlas_mapping=None)[source]

Get a VolumeModel instance that represents the atlas (segmented) volume

Parameters

atlas_mapping – IBL Mapping that we want to use on this atlas. See ibllib.atlas.AllenAtlas.regions.mappings.keys()

Returns

VolumeModel

get_dwi_model()[source]

Get a VolumeModel instance of the DWI volume image :return: VolumeModel

get_num_regions()[source]

Get how many regions are labelled

get_value_from_scalar_map(scalar)[source]

Reverse look-up in array to find a corresponding value

Parameters

scalar – Scalar value

Returns

Raw volume value

get_mapped_data(value)[source]

Given a value from the segmented volume, we retrieve useful info

Parameters

value – Value from the volume

Returns

Dictionary of corresponding data

remap(ids, source='Allen', dest='Beryl')[source]

Remap ids/scalar values from source to destination Function by Olivier Winter

Parameters

  • ids – List of ids

  • source – Source mapping

  • dest – Destination mapping

get_allen_volume_file_name(resolution, raw_image=False)[source]

Get the Allen volume file name given its resolution

Parameters

  • resolution – Resolution of the volume

  • raw_image – Whether we want the raw volume file name or the segmented one

Returns

String

get_allen_volume_url(resolution, raw_image=False)[source]

Construct a URL with which we can download data sets

Parameters

  • resolution – Volume resolution, either 10, 25, 50 or 100 (um)

  • raw_image – Whether the volume is the segmented one (aka the atlas) or the DWI

Returns

String

load_volume(file_path, remap_scalars=False, mapping=None, make_current=True)[source]

Load a volume data file. Supports NRRD and many other formats thanks to vedo/VTK

Parameters

  • file_path – Volume file path. Could support other file types easily.

  • remap_scalars – Whether scalar values in the volume are replaced by their row id from a mapping that stores. This is necessary in the case of segmented volumes with regions that have a discontinuous id.

  • mapping – Pandas Series or a Dictionary

  • make_current – Set the volume data as the current one

Returns

3D array

get_name(*args)[source]

Get full name for a model, separated by underscores :return: String

get_mapped_volume(volume, atlas_mapping=None, ibl_back_end=True)[source]

Set the volume data according to a mapping

Parameters

  • volume – Given volume to display

  • atlas_mapping – Mapping, either a string for the name of the mapping or an integer.

  • ibl_back_end – If you are not using ibllib and want to load your own volume, set this to False so that there will be no transposition of the volume (needed for the ones from IBL)

Returns

Volume nd array

get_region_and_row_id(acronym)[source]

Get region and row id given an acronym

Parameters

acronym – Acronym of a brain region

Returns

Region id and row id

get_regions_mask(region_ids, alpha_map=None)[source]

Build an alpha map that reveals only the given region ids

Parameters

  • region_ids – List or numpy array of region ids

  • alpha_map – Optional alpha map that will be modified. If None provided, the method will attempt to use the current active alpha map

Returns

2D numpy array with region ids and corresponding alpha values

class MouseBrainViewer[source]

Bases: iblviewer.application.Viewer

This is your entry point to International Brain Laboratory data visualization

initialize(resolution=25, mapping='Beryl', add_atlas=True, add_dwi=False, dwi_color_map='viridis', dwi_alpha_map=None, local_allen_volumes_path=None, offscreen=False, jupyter=False, embed_ui=False, embed_font_size=15, plot=None, plot_window_id=0, num_windows=1, render=False, dark_mode=False)[source]

Initialize the controller, main entry point to the viewer

Parameters

  • resolution – Resolution of the atlas volume. Possible values are 10 (requires a lot of RAM), 25, 50, 100. Units are in microns

  • mapping – Optional mapping value. In the context of IBL, there is ‘Allen’ for the standard Allen map and ‘Beryl’ (random name) which aggregates cortical layers as one.

  • add_atlas – Whether the Atlas is included in the viewer

  • add_dwi – Whether the diffusion weighted imaging is loaded in the viewer (same boundaries as atlas)

  • context – Context of the visualization

  • embed_ui – Whether the UI is embed within the VTK window

  • embed_font_size – Embed font size. Defaults to 16 points. You might need larger values in case you have a small screen with high dpi (but VTK methods fail to detect that).

  • jupyter – Whether we’re running from a jupyter notebook or not

  • plot – A vedo Plotter instance. You can either create it by yourself before hand, in case you want to have multiple windows with other stats or let the controller create a new one

  • plot_window_id – Sub-window id where the 3D visualization will be displayed

  • num_windows – Number of subwindows, in case you want to display your own stuff later

  • dark_mode – Whether the viewer is in dark mode

add_atlas_segmentation()[source]

Add the Allen Atlas segmented volume (aka label)

add_atlas_dwi(color_map, alpha_map)[source]

Add the Allen Atlas diffusion weighted image

Parameters

  • color_map – Color map for the volume

  • alpha_map – Alpha map for the volume

load_bounding_mesh(add_to_scene=False, alpha_on_scene=0.3)[source]

Load the bounding mesh of the mouse brain that represents its approximate pial limit

find_region(term)[source]

Find a region with a substring

Parameters

term – Search term

Returns

List of matching entries and the corresponding mask

get_region_names()[source]

Get the region names :return: List

get_selection_info(line_length=40, precision=5)[source]

Get information about the current selection

Parameters

  • line_length – Region name line length after what it’s word-wrapped

  • precision – Scalar value floating precision displayed

Returns

Preformatted multiline text and a dictionary of extra data

add_origin()[source]

Add the origin on scene

add_many_points_test(positions, point_radius=2, auto_xy_rotate=True, add_to_scene=True)[source]

Test method that validates that VTK is fast enough for displaying 10 million points interactively (and it is :)

add_spheres(positions, radius=10, values=None, color_map='Accent', name='Spheres', use_origin=True, add_to_scene=True, noise_amount=0, trim_outliers=True, bounding_mesh=None, ibl_flip_yz=True, **kwargs)[source]

Add new spheres

Parameters

  • positions – 3D array of coordinates

  • radius – List same length as positions of radii. The default size is 5um, or 5 pixels in case as_spheres is False.

  • values – 1D array of values, one per neuron or a time series of such 1D arrays (numpy format)

  • color_map – A color map, it can be a color map built by IBLViewer or a color map name (see vedo documentation), or a list of values, etc.

  • name – All point neurons are grouped into one object, you can give it a custom name

  • use_origin – Whether the origin is added as offset to the given positions

  • add_to_scene – Whether the new lines are added to scene/plot and rendered

  • noise_amount – Amount of 3D random noise applied to each point. Defaults to 0

  • trim_outliers – If bounding_mesh param is given, then the spheres will be trimmed, only the ones inside the bounding mesh will be kept

  • bounding_mesh – A closed manifold surface mesh used for trimming segments. If None, the current self.bounding_mesh is used (if it exists)

  • ibl_flip_yz – If you have an IBL data set, its 3D coordinates will be multiplied by -1 on Y and Z axes in order to match Allen Brain Atlas volume and how it’s stored by IBL.

Returns

objects.Points

add_points(positions, radius=10, values=None, color_map='Accent', name='Points', screen_space=False, use_origin=True, add_to_scene=True, noise_amount=0, trim_outliers=True, bounding_mesh=None, ibl_flip_yz=True, **kwargs)[source]

Add new points

Parameters

  • positions – 3D array of coordinates

  • radius – List same length as positions of radii. The default size is 5um, or 5 pixels in case as_spheres is False.

  • values – 1D array of values, one per neuron or a time series of such 1D arrays (numpy format)

  • color_map – A color map, it can be a color map built by IBLViewer or a color map name (see vedo documentation), or a list of values, etc.

  • name – All point neurons are grouped into one object, you can give it a custom name

  • screen_space – Type of point, if True then the points are static screen-space points. If False, then the points are made to scale in 3D, ie you see them larger when you zoom closer to them, while this is not the case with screen-space points. Defaults to False.

  • use_origin – Whether the origin is added as offset to the given positions

  • add_to_scene – Whether the new lines are added to scene/plot and rendered

  • noise_amount – Amount of 3D random noise applied to each point. Defaults to 0

  • trim_outliers – If bounding_mesh param is given, then the spheres will be trimmed, only the ones inside the bounding mesh will be kept

  • bounding_mesh – A closed manifold surface mesh used for trimming segments. If None, the current self.bounding_mesh is used (if it exists)

  • ibl_flip_yz – If you have an IBL data set, its 3D coordinates will be multiplied by -1 on Y and Z axes in order to match Allen Brain Atlas volume and how it’s stored by IBL.

Returns

objects.Points

add_segments(points, end_points=None, line_width=2, values=None, color_map='Accent', name='Segments', use_origin=True, add_to_scene=True, relative_end_points=False, spherical_angles=None, radians=True, trim_outliers=True, bounding_mesh=None, ibl_flip_yz=True)[source]

Add a set of segments

Parameters

  • points – 3D numpy array of points of length n

  • end_points – 3D numpy array of points of length n

  • line_width – Line width, defaults to 2px

  • values – 1D list of length n, for one scalar value per line

  • color_map – A color map, it can be a color map built by IBLViewer or a color map name (see vedo documentation), or a list of values, etc.

  • name – Name to give to the object

  • use_origin – Whether the current origin (not necessarily absolute 0) is used as offset

  • add_to_scene – Whether the new lines are added to scene/plot and rendered

  • relative_end_points – Whether the given end point is relative to the start point. False by default, except is spherical coordinates are given

  • spherical_angles – 3D numpy array of spherical angle data of length n In case end_points is None, this replaces end_points by finding the relative coordinate to each start point with the given radius/depth, theta and phi

  • radians – Whether the given spherical angle data is in radians or in degrees

  • trim_outliers – Whether segments are cropped by the bounding mesh

  • bounding_mesh – A closed manifold surface mesh used for trimming segments. If None, the current self.bounding_mesh is used (if it exists)

  • ibl_flip_yz – If you have an IBL data set, its 3D coordinates will be multiplied by -1 on Y and Z axes in order to match Allen Brain Atlas volume and how it’s stored by IBL.

Returns

objects.Lines

add_lines(points, line_width=2, values=None, color_map='Accent', name='Lines', use_origin=True, add_to_scene=True, trim_outliers=True, bounding_mesh=None, ibl_flip_yz=True)[source]

Create a set of lines with given point sets

Parameters

  • points – List of lists of 3D coordinates

  • line_width – Line width, defaults to 2px

  • values – 1D list of length n, for one scalar value per line

  • color_map – A color map, it can be a color map built by IBLViewer or a color map name (see vedo documentation), or a list of values, etc.

  • name – Name to give to the object

  • use_origin – Whether the current origin (not necessarily absolute 0) is used as offset

  • add_to_scene – Whether the new lines are added to scene/plot and rendered

  • trim_outliers – Whether segments that are out of the brain envelope are trimmed or not. True by default

  • bounding_mesh – A closed manifold surface mesh used for trimming lines. If None, the current self.bounding_mesh is used (if it exists)

  • ibl_flip_yz – If you have an IBL data set, its 3D coordinates will be multiplied by -1 on Y and Z axes in order to match Allen Brain Atlas volume and how it’s stored by IBL.

Returns

objects.Lines

add_volume(data=None, resolution=None, file_path=None, color_map='viridis', alpha_map=None, select=False, add_to_scene=True, transpose=None)[source]

Add a volume to the viewer with box clipping and slicing enabled by default

Parameters

  • data – Volume image data or a file_path

  • resolution – Resoluton of the volume

  • file_path – File path of the volume. If you don’t provide an image volume data, then the file_path will be used to load the volume data

  • color_map – Color map for the volume

  • alpha_map – Alpha map for the volume. If None, it will assume that 0 values are transparent and maximum values are opaque

  • select – Whether the volume is selected

  • add_to_scene – Whether the volume is added to scene

  • transpose – Transposition parameter. If None. nothing happens. If True, then the default IBL transposition is applied. You can provide your own, that is, a list of 3 elements to reorder the volume as desired.

Returns

VolumeController

set_left_view()[source]

Set left sagittal view

set_right_view()[source]

Set right sagittal view

set_anterior_view()[source]

Set anterior coronal view

set_posterior_view()[source]

Set posterior coronal view

set_dorsal_view()[source]

Set dorsal axial view

set_ventral_view()[source]

Set ventral axial view