plot_fibers_in_striatum_coronal_view.py

generates a coronal view of the striatum with the fiber tips

1. Import required packages

import pandas as pd
import numpy as np
from pathlib import Path
from PIL import Image
import urllib.request
from os.path import exists
from brain_locations_visualizer import config_parser
from brain_locations_visualizer.plotting_functions import generate_coronal_figure

2. Get the configuration file and define variables

# use the default config file for the documentation
config_file = Path("config_for_documentation.json")

# variables are assigned in the config_parser function
config_parser.config_parser(config_file)

plot with a different marker the flat fibers and the tapered fibers (this is specified in the mouse name)

ff_marker = "_"
tf_marker = "|"

define the output directory as the parent of the config file

parent = Path(config_file).parent

read the file of points

coords = pd.read_csv(config_parser.file_path, header=0)
X = coords.x
Y = coords.y
Z = coords.z
Animal_Name = coords.Mouse_name

select only the fibers used in the analysis CAREFUL HERE WITH WHERE IS LEFT AND WHERE IS RIGHT!! animals that are not included have a # in front of their name

animal_mask = [not an.startswith("#") for an in Animal_Name]
X = np.array(list(X[animal_mask])).astype(float)
Y = np.array(list(Y[animal_mask])).astype(float)
Z = np.array(list(Z[animal_mask])).astype(float)
Animal_Name = np.array(list(Animal_Name[animal_mask]))

2.1 This part decides which slices to show

# download the atlas from the server into the data folder
url = "https://zenodo.org/record/7501966/files/" + config_parser.atlas_path.name
# download if data is not there
if not exists(config_parser.atlas_path):
    print('Downloading data...')
    urllib.request.urlretrieve(url, config_parser.atlas_path)
else:
    print('Data already in directory')

Data already in directory

read atlas and get its dimensions

atlas = Image.open(config_parser.atlas_path)
try:
    h, w, _ = np.shape(atlas)
except Exception:
    h, w = np.shape(atlas)

show images evenly if the precise slices are not specified in the config file

if config_parser.sl_list == []:
    step = int(
        np.floor(
            (config_parser.z_limits[1] - config_parser.z_limits[0])
            / config_parser.n_images
        )
    )
    sl_list = list(
        range(config_parser.z_limits[0], config_parser.z_limits[1], step)
    )
    sl_list = sl_list[-config_parser.n_images :]

else:
    sl_list = config_parser.sl_list

Mirror all to the right hemisphere (optional)

atlas_mid_point = w / 2
for i in range(len(Z)):
    if Z[i] < atlas_mid_point:
        dist_to_center = atlas_mid_point - Z[i]
        Z[i] = atlas_mid_point + dist_to_center

define masks for the different animals and fiber types

mask_1 = [x.startswith(config_parser.id_1) for x in Animal_Name]
mask_2 = [x.startswith(config_parser.id_2) for x in Animal_Name]
mask_other = np.logical_and(
    [not e for e in mask_1], [not e for e in mask_2]
)

ff_mask = [x.endswith("_flat") for x in Animal_Name]
tf_mask = [not x for x in ff_mask]

2. Generate the figure

generate_coronal_figure(
    config_parser,
    sl_list,
    w,
    h,
    atlas,
    X,
    Y,
    Z,
    mask_1,
    mask_2,
    mask_other,
    ff_mask,
    tf_mask,
    ff_marker,
    tf_marker,
    parent,
)