Example notebook neuralib.plot
[1]:
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
from neuralib.io.dataset import load_example_rois
from neuralib.plot import (
dotplot,
violin_boxplot,
scatter_histplot,
diag_histplot,
diag_heatmap,
grid_subplots,
VennDiagram
)
[2]:
%load_ext autoreload
%autoreload
Plot the data with violin and box with dots
[3]:
df = load_example_rois()
print(df)
shape: (45_163, 17)
┌────────────┬─────────┬───────────┬───────────┬───┬───────────┬───────────┬───────────┬───────────┐
│ name ┆ acronym ┆ AP_locati ┆ DV_locati ┆ … ┆ merge_ac_ ┆ merge_ac_ ┆ merge_ac_ ┆ family │
│ --- ┆ --- ┆ on ┆ on ┆ ┆ 2 ┆ 3 ┆ 4 ┆ --- │
│ str ┆ str ┆ --- ┆ --- ┆ ┆ --- ┆ --- ┆ --- ┆ str │
│ ┆ ┆ f64 ┆ f64 ┆ ┆ str ┆ str ┆ str ┆ │
╞════════════╪═════════╪═══════════╪═══════════╪═══╪═══════════╪═══════════╪═══════════╪═══════════╡
│ Ectorhinal ┆ ECT5 ┆ -3.03 ┆ 4.34 ┆ … ┆ ECT ┆ ECT ┆ ECT ┆ ISOCORTEX │
│ area/Layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 5 ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Perirhinal ┆ PERI6a ┆ -3.03 ┆ 4.42 ┆ … ┆ PERI ┆ PERI ┆ PERI ┆ ISOCORTEX │
│ area layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Perirhinal ┆ PERI6a ┆ -3.03 ┆ 4.55 ┆ … ┆ PERI ┆ PERI ┆ PERI ┆ ISOCORTEX │
│ area layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Perirhinal ┆ PERI6a ┆ -3.03 ┆ 4.5 ┆ … ┆ PERI ┆ PERI ┆ PERI ┆ ISOCORTEX │
│ area layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Entorhinal ┆ ENTl5 ┆ -3.03 ┆ 4.92 ┆ … ┆ ENT ┆ ENT ┆ ENTl ┆ HPF │
│ area ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ lateral ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ part l… ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … ┆ … │
│ Temporal ┆ TEa6a ┆ -2.91 ┆ 3.79 ┆ … ┆ TEa ┆ TEa ┆ TEa ┆ ISOCORTEX │
│ associatio ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ n areas ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ lay… ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Ectorhinal ┆ ECT6a ┆ -2.91 ┆ 4.31 ┆ … ┆ ECT ┆ ECT ┆ ECT ┆ ISOCORTEX │
│ area/Layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Ventral ┆ AUDv6a ┆ -2.91 ┆ 3.52 ┆ … ┆ AUD ┆ AUD ┆ AUDv ┆ ISOCORTEX │
│ auditory ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ area layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Ectorhinal ┆ ECT6a ┆ -2.91 ┆ 4.14 ┆ … ┆ ECT ┆ ECT ┆ ECT ┆ ISOCORTEX │
│ area/Layer ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ 6a ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ Temporal ┆ TEa5 ┆ -2.91 ┆ 4.02 ┆ … ┆ TEa ┆ TEa ┆ TEa ┆ ISOCORTEX │
│ associatio ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ n areas ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
│ lay… ┆ ┆ ┆ ┆ ┆ ┆ ┆ ┆ │
└────────────┴─────────┴───────────┴───────────┴───┴───────────┴───────────┴───────────┴───────────┘
[4]:
sns.set_theme(style='white', font_scale=1.2)
fig, ax = plt.subplots()
violin_boxplot(data=df, x='source', y='AP_location', hue='source', ax=ax)
plt.show()
See correlation and histogram of two variables
[5]:
iris = sns.load_dataset('iris')
x = iris['petal_width']
y = iris['sepal_width']
scatter_histplot(x, y, linear_reg=True, bins=20)
Diagonal plot value comparison
same measurement(metric) in different condition (xy)
[12]:
N = 1000
# create higher density with bias distribution
x = np.random.beta(a=2, b=5, size=N)
y = np.random.beta(a=5, b=2, size=N)
_, ax = plt.subplots(1, 2)
# with histogram
diag_histplot(x, y, ax=ax[0], scatter_kws={'s': 5, 'c': 'gray', 'marker': '.', 'edgecolors': 'none'})
# with heatmap
diag_heatmap(x, y, ax=ax[1], extent=(0, 1, 0, 1), scatter_kws={'s': 5, 'c': 'w', 'marker': '.', 'edgecolors': 'none'})
ax[0].set(xlim=(0, 1), ylim=(0, 1))
[12]:
[(0.0, 1.0), (0.0, 1.0)]
Dotsplot with colormap
[7]:
# (X,): different animals
xlabel = ['animal_A', 'animal_B', 'animal_C', 'animal_D', 'animal_E']
# (Y,): subregion in Visual Cortex as an example
ylabel = ['VISam', 'VISp', 'VISpm', 'VISpor', 'VISl', 'VISal', 'VISli', 'VISpl']
# values: Array[float, [X, Y]]
nx = len(xlabel)
ny = len(ylabel)
values = np.random.sample((nx, ny)) * 100
dotplot(xlabel, ylabel, values,
scale='area',
max_marker_size=700,
with_color=True,
figure_title='example of dotplot')
Venn Diagram
[8]:
# 2 sets
subsets = {'setA': 10, 'setB': 20}
vd = VennDiagram(subsets, colors=('pink', 'palegreen'))
vd.add_intersection('setA & setB', 5)
vd.add_total(100)
vd.plot()
vd.show()
[9]:
# 3 sets
subsets = {'setA': 20, 'setB': 100, 'setC': 50}
vd = VennDiagram(subsets)
vd.add_intersection('setA & setB', 10)
vd.add_intersection('setB & setC', 10)
vd.add_intersection('setA & setC', 10)
vd.add_intersection('setA & setB & setC', 2)
vd.add_total(200)
vd.plot()
vd.show()
Grid subplots
[10]:
data = np.random.sample((30, 10, 2))
grid_subplots(data, 5, 'plot', dtype='xy', ls='dotted')
[11]:
def generate_example_2d_tuning(blobs: int = 6) -> np.ndarray:
"""example mimic place cell place cell firing"""
x, y = np.meshgrid(np.linspace(0, 6, 50), np.linspace(0, 6, 50))
z = np.zeros_like(x)
for _ in range(blobs):
x0 = np.random.uniform(0, 6)
y0 = np.random.uniform(0, 6)
sigma = 0.5
z += np.exp(-((x - x0) ** 2 + (y - y0) ** 2) / (2 * sigma ** 2))
return z
data = [generate_example_2d_tuning() for i in range(30)]
grid_subplots(data, images_per_row=6, plot_func='imshow', dtype='img')
