This commit is contained in:
2026-04-26 16:59:15 +02:00
parent c0d0e25ca2
commit 111ce77513
+141
View File
@@ -0,0 +1,141 @@
import math
import matplotlib.pyplot as plt
from matplotlib.collections import LineCollection
import matplotlib as mpl
import numpy as np
import squidpy as sq
import scipy
import spatialdata as sd
from spatialdata_io.experimental import to_legacy_anndata
from graph_tool.all import *
from src import centrality
from src import plot
from src import fitting
def merfish():
"""
Merfish dataset from `squidpy`.
"""
adata = sq.datasets.merfish()
adata = adata[adata.obs.Bregma == -9].copy()
return adata
def random_graph(n=5000, seed=None):
"""
Uniformly random point cloud generation.
`n` [int] Number of points to generate. Default 5000 seems like a good starting point in point density and corresponding runtime for the subsequent calculations.
@return [numpy.ndarray] Array of shape(n, 2) containing the coordinates for each point of the generated point cloud.
"""
if seed is None:
import secrets
seed = secrets.randbits(128)
rng = np.random.default_rng(seed=seed)
return rng.random((n, 2)), seed
def spatial_graph(adata):
"""
Generate the spatial graph using delaunay for the given `adata`.
`adata` will contain the calculated spatial graph contents in the keys
adata.obsm['spatial']` in case the `adata` is created from a dataset of *squidpy*.
@return [Graph] generated networkx graph from adata.obsp['spatial_distances']
"""
g, pos = graph_tool.generation.triangulation(adata, type="delaunay")
g.vp["pos"] = pos
weight = g.new_edge_property("double")
for e in g.edges():
weight[e] = math.sqrt(sum(map(abs, pos[e.source()].a - pos[e.target()].a)))**2
g.ep["weight"] = weight
return g, weight
def plot_graph_nodes(g, centralities, fig, ax, name):
pos = g.vp["pos"]
x = []
y = []
for v in g.vertices():
ver = pos[v]
x.append(ver[0])
y.append(ver[1])
sc = ax.scatter(x, y, s=1, c=centralities, cmap=plt.cm.plasma)
ax.set_title(name)
fig.colorbar(sc, ax=ax)
def plot_relationship_nodes(g, vp, convex_hull, fig, ax, name):
quantification = plot.quantification_data(g, vp, convex_hull)
# optimize model's piece-wise linear function
d = quantification[:, 0]
C = quantification[:, 1]
m_opt, c0_opt, b_opt, aic_opt = fitting.fit_piece_wise_linear(d, C, M=100)
# TODO
# should this be part of the plotting function itself, it should not be necessary for me to do this
d_curve = np.linspace(min(d), max(d), 500)
C_curve = np.piecewise(
d_curve,
[d_curve <= b_opt, d_curve > b_opt],
[lambda x: m_opt * x + c0_opt, lambda x: m_opt * b_opt + c0_opt]
)
ax.set_title(name)
ax.set_xlabel('Distance to Bounding-Box')
ax.set_ylabel('Centrality')
ax.scatter(quantification[:, 0], quantification[:, 1], c=quantification[:, 1], cmap=plt.cm.plasma, s=0.2)
ax.plot(d_curve, C_curve, color='g', linewidth=1, label=f"m = {m_opt} | b = {b_opt} | c = {c0_opt}")
ax.legend()
def plot_degree_distribution(g, fig, ax, name):
ax.set_title(name)
ax.set_ylabel('Degree')
ax.set_xlabel('# Occurances')
degrees = g.get_total_degrees(list(g.vertices()))
bins = degrees.max() - degrees.min()
counts, bins, patches = ax.hist(degrees, bins=bins, orientation='horizontal')
# Label the percentages below the x-axis...
# bin_centers = 0.5 * np.diff(bins) + bins[:-1]
# for count, x in zip(counts, bin_centers):
# # Label the percentages
# percent = '%0.000f%%' % (100 * float(count) / counts.sum())
# ax.annotate(percent, xy=(x, 0), xycoords=('data', 'axes fraction'),
# xytext=(0, -18), textcoords='offset points', va='top', ha='center')
points, seed = random_graph(n=2000, seed=231533135843957409942915332448253409428)
g, weight = spatial_graph(points)
# adata = merfish()
# g, weight = spatial_graph(adata.obsm['spatial'])
g = GraphView(g)
# relationship with betweenness scoring for both node and edges
vp = pagerank(g, weight=weight)
vp.a = np.nan_to_num(vp.a) # correct floating point values
# plot graph
fig = plt.figure(figsize=(16, 9), layout='constrained')
fig.suptitle(f"Artical (n = 2000 | seed {seed})", fontsize=16)
ax1, ax2 = fig.subplots(1, 2)
# relationship with betweenness scoring for both node and edges
vp = pagerank(g, weight=weight)
vp.a = np.nan_to_num(vp.a) # correct floating point values
# normalize centrality values
min_val, max_val = vp.a.min(), vp.a.max()
vp.a = (vp.a - min_val) / (max_val - min_val)
# compare relationships
convex_hull = centrality.convex_hull(g)
plot_degree_distribution(g, fig, ax2, "Degree Distribution")
plot_relationship_nodes(g, vp, convex_hull, fig, ax1, "Pagerank Centrality with fitted model")
fig.savefig(f"degree_distribution_vs_pagerank_artifical.pdf", format='pdf')