Atlas Tools

Some important static methods used throughout the Django project.

neuroglancer.atlas.align_atlas(animal, annotator_id, source, reverse=False, reference_scales=None)

This prepares the data for the align_point_sets method. Make sure we have at least 3 points

Parameters:

  • animal – the animal we are aligning to

  • input_type_id – the int defining what type of input. Taken from the input_type table with column=id

  • owner_id – the int defining the person. Taken from the auth_user table column=id

Returns:

a 3x3 matrix and a 1x3 matrix

neuroglancer.atlas.brain_to_atlas_transform(brain_coord, r, t)

Taken from abakit Takes an x,y,z brain coordinates, and a rotation matrix and translation vector.

Parameters:

  • atlas_coord – tuple of x,y,z coordinates of the atlas in micrometers

  • r – float of the rotation matrix

  • t – vector of the translation matrix

Returns:

the point in atlas coordinates in micrometers.

neuroglancer.atlas.get_scales(prep_id)

A generic method to safely query and return resolutions. Note, since prep_id is a varchar, you can’t use objects.get, you must use objects.filter

Parameters:

prep_id – varchar of the primary key of the animal

neuroglancer.atlas.make_ontology_graph_CCFv3()

Load the allen CCFv3 ontology into a graph object

neuroglancer.atlas.make_ontology_graph_pma()

Load the Princeton Mouse Atlas ontology into a graph object

neuroglancer.atlas.umeyama(src, dst, with_scaling=True)

The Umeyama algorithm to register landmarks with rigid transform.

See the paper ‘Least-squares estimation of transformation parameters between two point patterns’.

Parameters:

  • src – List of data points.

  • dst – List of data points.

  • with_scaling – A boolean determining if we should scale or not.