fafbseg.google.find_missed_branches

fafbseg.google.find_missed_branches(x, autoseg_instance, tag=False, tag_size_thresh=10, min_node_overlap=4, **kwargs)[source]

Use autoseg to find (and annotate) potential missed branches.

Parameters:

  • x (pymaid.CatmaidNeuron/List) – Neuron(s) to search for missed branches.

  • autoseg_instance (pymaid.CatmaidInstance) – CATMAID instance containing the autoseg skeletons.

  • tag (bool, optional) – If True, will tag nodes of x that might have missed branches with “missed branch?”.

  • tag_size_thresh (int, optional) – Size threshold in microns of cable for tagging potentially missed branches.

  • min_node_overlap (int, optional) – Minimum number of nodes that input neuron(s) x must overlap with given segmentation ID for it to be included.

  • **kwargs – Keyword arguments passed to fafbseg.neuron_from_segments.

Returns:

  • summary (pandas.DataFrame) – DataFrame containing a summary of potentially missed branches.

    If input is a single neuron:

  • fragments (pymaid.CatmaidNeuronList) – Fragments found to be potentially overlapping with the input neuron.

  • branches (pymaid.CatmaidNeuronList) – Potentially missed branches extracted from fragments.

Examples

Setup

>>> import fafbseg                                                  
>>> import pymaid                                                   

>>> # Set up connections to manual and autoseg CATMAID
>>> manual = pymaid.CatmaidInstance('URL', 'HTTP_USER', 'HTTP_PW', 'API_TOKEN') 
>>> auto = pymaid.CatmaidInstance('URL', 'HTTP_USER', 'HTTP_PW', 'API_TOKEN')   

>>> # Set a source for segmentation data
>>> fafbseg.google.use_google_storage("https://storage.googleapis.com/fafb-ffn1-20190805/segmentation")

Find missed branches and tag them

>>> # Fetch a neuron
>>> x = pymaid.get_neuron(16, remote_instance=manual)                           
>>> # Find and tag missed branches
>>> (summary,                                                                   
...  fragments,
...  branches) = fafbseg.google.find_missed_branches(x, autoseg_instance=auto)

>>> # Show summary of missed branches
>>> summary.head()                                                              
   n_nodes  cable_length   node_id
0      110     28.297424   3306395
1       90     23.976504  20676047
2       64     15.851333  23419997
3       29      7.494350   6298769
4       16      3.509739  15307841

>>> # Co-visualize your neuron and potentially overlapping autoseg fragments
>>> x.plot3d(color='w')                                                         
>>> fragments.plot3d()                                                          

>>> # Visualize the potentially missed branches
>>> pymaid.clear3d()                                                            
>>> x.plot3d(color='w')                                                         
>>> branches.plot3d(color='r')