Spaces:
Sleeping
Sleeping
File size: 18,667 Bytes
31d5c57 a233044 31d5c57 a079145 d9ac8bf 31d5c57 a233044 31d5c57 a233044 d9ac8bf a233044 31d5c57 9ed0f34 31d5c57 d9ac8bf 31d5c57 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 | import streamlit as st
import plotly.graph_objects as go
import plotly.express as px
import numpy as np
import pandas as pd
from PIL import Image
import io
from datetime import datetime
import matplotlib.pyplot as plt
import scanpy as sc
from itertools import combinations
from typing import Optional
from scipy.sparse import issparse
from scipy.stats import mannwhitneyu
from src.backend.flux_distribution import adata_to_long_df, p_to_star
# Standard color map for metabolic interaction types
INTERACTION_COLORS = {
"Competition": "#d32f2f", # Red
"Release": "#1976d2", # Blue
"Cooperation": "#388e3c", # Green
"Amensalism": "#fbc02d", # Amber
"Neutralism": "#7b1fa2", # Purple
"Interaction": "#607d8b" # Grey (fallback)
}
try:
from statsmodels.stats.multitest import multipletests
_HAS_STATSMODELS = True
except ImportError:
_HAS_STATSMODELS = False
def display_help_button(help_text, plot_name):
"""
Shows a help popover with insights for the plot.
"""
if help_text:
with st.popover("", icon=":material/help:", help="Click for insights", use_container_width=True):
st.markdown(f"#### <i class='fas fa-lightbulb'></i> Plot Insights", unsafe_allow_html=True)
st.markdown(help_text)
def display_plot_with_download(fig, plot_name: str = "plot", help_text: str = None):
"""
Display a matplotlib figure with aligned help and download buttons on top right.
Reuses same figure object to prevent flickering.
"""
# Use consistent column ratios: Spacer, Help, Download.
cols = st.columns([0.7, 0.2, 0.1], gap="small")
with cols[1]:
display_help_button(help_text, plot_name)
with cols[2]:
# Generate PDF file
pdf_buffer = io.BytesIO()
fig.savefig(pdf_buffer, format='pdf', dpi=300, bbox_inches='tight')
file_data = pdf_buffer.getvalue()
st.download_button(
label="",
data=file_data,
file_name=f"{plot_name}_{datetime.now().strftime('%Y%m%d_%H%M%S')}.pdf",
mime="application/pdf",
key=f"download_{plot_name}_{id(fig)}",
help="Download as PDF",
icon=":material/download:",
use_container_width=True
)
# Display the plot
st.pyplot(fig, use_container_width=False)
def display_plotly_with_download(fig, plot_name: str = "plot", help_text: str = None):
"""
Display a Plotly figure with aligned help button on top right.
Optimized to prevent flickering on reruns.
"""
cols = st.columns([0.7, 0.2, 0.1], gap="small")
with cols[1]:
display_help_button(help_text, plot_name)
with cols[2]:
st.empty()
# Use a unique key to prevent redraws and add config to disable animations for faster rendering
st.plotly_chart(
fig,
use_container_width=False,
key=f"plotly_{plot_name}_{id(fig)}",
config={"displayModeBar": False, "responsive": True, "staticPlot": False}
)
def display_interactive_spatial_plot(adata, color_key="domain", spot_size = 6, plot_name="spatial_plot", title: Optional[str] = None, help_text: Optional[str] = None):
# spot_size = spot_size
try:
# Create columns for help/download above the plot if help_text is provided
if help_text:
col_space, col_help, col_download = st.columns([5.0, 0.5, 0.5], gap="small")
with col_help:
display_help_button(help_text, plot_name)
library_id = list(adata.uns["spatial"].keys())[0]
img_key = "hires" if "hires" in adata.uns["spatial"][library_id]["images"] else "downscaled_fullres"
img = adata.uns["spatial"][library_id]["images"][img_key]
sf_key = f"tissue_{img_key}_scalef"
sf = adata.uns["spatial"][library_id]["scalefactors"][sf_key]
coords = adata.obsm["spatial"] * sf
if color_key in adata.var_names:
var_idx = adata.var_names.get_loc(color_key)
raw = adata.X[:, var_idx]
color_values = raw.toarray().flatten() if hasattr(raw, "toarray") else np.asarray(raw).flatten()
is_categorical = False
elif color_key in adata.obs.columns:
color_values = adata.obs[color_key].values
is_categorical = not pd.api.types.is_numeric_dtype(adata.obs[color_key])
else:
color_values = np.full(len(coords), "N/A")
is_categorical = True
df = pd.DataFrame({
"x": coords[:, 0],
"y": coords[:, 1],
"color": color_values.astype(str) if is_categorical else color_values,
"domain": adata.obs["domain"].values if "domain" in adata.obs.columns else "N/A",
"spot_id": adata.obs_names.tolist()
})
last_key = st.session_state.get(f"{plot_name}_last_key")
if last_key != color_key:
st.session_state.pop(f"{plot_name}_relayout", None)
st.session_state[f"{plot_name}_last_key"] = color_key
plot_state = st.session_state.get(plot_name, {})
relayout = None
if isinstance(plot_state, dict):
relayout = plot_state.get("relayout_data") or plot_state.get("relayout")
elif hasattr(plot_state, "selection"):
relayout = getattr(plot_state, "relayout_data", None)
zoom_ratio = 1.0
has_zoom = relayout and isinstance(relayout, dict) and "xaxis.range[0]" in relayout
if has_zoom:
try:
xr = [relayout["xaxis.range[0]"], relayout["xaxis.range[1]"]]
zoom_ratio = abs(xr[1] - xr[0]) / img.shape[1]
except (IndexError, KeyError, ZeroDivisionError):
zoom_ratio = 1.0
fig = go.Figure()
fig.add_layout_image(
dict(
source=Image.fromarray((img * 255).astype(np.uint8)),
xref="x", yref="y",
x=0, y=0,
sizex=img.shape[1], sizey=img.shape[0],
sizing="stretch", layer="below"
)
)
if is_categorical:
palette = px.colors.qualitative.T10
unique_vals = sorted(df["color"].astype(str).unique())
for i, val in enumerate(unique_vals):
sub = df[df["color"].astype(str) == val]
fig.add_trace(go.Scattergl(
x=sub["x"],
y=sub["y"],
customdata=np.stack((sub["spot_id"], sub["domain"]), axis=-1),
mode="markers",
name=str(val),
marker=dict(
size=spot_size,
color=palette[i % len(palette)],
line=dict(width=0.5, color='white')
),
hovertemplate=(
"<b>Domain: %{customdata[1]}</b><br>"
"<span style='font-size:0.8rem;'>ID: %{customdata[0]}</span>"
"<extra></extra>"
)
))
else:
fig.add_trace(go.Scattergl(
x=df["x"], y=df["y"],
customdata=np.stack((df["spot_id"], df["domain"]), axis=-1),
mode="markers",
marker=dict(
size=spot_size,
color=df["color"],
colorscale="Jet",
showscale=True,
colorbar=dict(
thickness=8,
len=0.75,
xref="paper",
yref="paper",
tickfont=dict(size=10),
outlinewidth=0,
),
line=dict(width=0.3, color='white')
),
hovertemplate=(
"<b>Domain: %{customdata[1]}</b><br>"
f"<b>Flux:</b> %{{marker.color:.3e}}<br>"
"<span style='font-size:0.8rem;'>ID: %{customdata[0]}</span>"
"<extra></extra>"
)
))
# Enforce square axes aligned to tissue image
fig.update_xaxes(
visible=False,
range=[0, img.shape[1]],
scaleanchor="y",
scaleratio=1,
)
fig.update_yaxes(
visible=False,
range=[img.shape[0], 0],
scaleanchor="x",
scaleratio=1,
constrain="domain",
)
fig.update_layout(
title=dict(
text=title if title else "",
x=0.5,
y=0.98,
xanchor="center",
yanchor="top",
font=dict(size=16)
) if title else None,
margin=dict(l=0, r=0, t=40 if title else 0, b=0),
legend=dict(
orientation="v",
yanchor="top",
y=0.99,
xanchor="left",
x=0.01,
bgcolor="rgba(255,255,255,0.6)"
),
paper_bgcolor='rgba(0,0,0,0)',
plot_bgcolor='rgba(0,0,0,0)',
dragmode="pan",
uirevision="constant"
)
plot_event = st.plotly_chart(
fig,
use_container_width=False,
config={'scrollZoom': True},
key=plot_name,
on_select="rerun"
)
if plot_event and hasattr(plot_event, "get"):
relayout = plot_event.get("relayout_data") or plot_event.get("selection", {}).get("relayout_data")
if relayout:
st.session_state[f"{plot_name}_relayout"] = relayout
return True
except Exception as e:
st.error(f"Error rendering interactive plot: {e}")
return False
def display_formatted_table(df: pd.DataFrame, title: Optional[str] = None):
"""Display a dataframe with scientific notation for small float values."""
if title:
st.markdown(f"##### <i class='fas fa-table'></i> {title}", unsafe_allow_html=True)
config = {}
if not df.empty:
for col in df.select_dtypes(include=['float']).columns:
if 'p_val' in col.lower() or 'pvalue' in col.lower() or df[col].abs().max() < 1e-2:
config[col] = st.column_config.NumberColumn(format="%.2e")
else:
config[col] = st.column_config.NumberColumn(format="%.4f")
st.dataframe(df, width='stretch', column_config=config)
def add_significance_brackets(ax, df, domain_order, y_col="flux"):
"""
Add pairwise significance brackets above a boxen/box plot.
Uses Mann-Whitney U test with FDR-BH correction across all pairs.
Only significant pairs (p_adj < 0.05) are annotated.
"""
pairs = list(combinations(domain_order, 2))
pvalues = []
valid_pairs = []
for d1, d2 in pairs:
g1 = df.loc[df["domain"] == d1, y_col].dropna()
g2 = df.loc[df["domain"] == d2, y_col].dropna()
if len(g1) < 3 or len(g2) < 3:
continue
_, p = mannwhitneyu(g1, g2, alternative="two-sided")
pvalues.append(p)
valid_pairs.append((d1, d2))
if not valid_pairs:
return
if _HAS_STATSMODELS:
_, p_adj, _, _ = multipletests(pvalues, method="fdr_bh")
else:
p_adj = np.array(pvalues)
y_max = df[y_col].max()
y_range = df[y_col].max() - df[y_col].min()
step = y_range * 0.08
bracket_y = y_max + step
for (d1, d2), p in zip(valid_pairs, p_adj):
star = p_to_star(p)
if star == "ns":
continue
x1 = domain_order.index(d1)
x2 = domain_order.index(d2)
mid = (x1 + x2) / 2
ax.plot([x1, x1, x2, x2], [bracket_y, bracket_y + step * 0.3, bracket_y + step * 0.3, bracket_y],
lw=1.2, c="black")
ax.text(mid, bracket_y + step * 0.35, star, ha="center", va="bottom", fontsize=9)
bracket_y += step * 0.9 # stack brackets upward
def create_plotly_tme_plot(adata, interaction_type_df, interaction_score_df, selected_rxn_id, selected_display_name, percentile_threshold=95):
coords_df = pd.DataFrame(adata.obsm["spatial"], index=adata.obs.index, columns=['x', 'y'])
y_max = coords_df['y'].max()
coords_df['y_plot'] = y_max - coords_df['y']
coords_df['domain'] = adata.obs['domain'] if 'domain' in adata.obs.columns else "N/A"
if percentile_threshold > 0:
thresh = interaction_score_df['Interaction score'].quantile(percentile_threshold / 100)
scores = interaction_score_df[interaction_score_df['Interaction score'] >= thresh]
else:
scores = interaction_score_df
rxn_mask = interaction_type_df['Reaction'].str.replace(r'_(b|f)$', '', regex=True) == selected_rxn_id
rxn_data = interaction_type_df[rxn_mask]
merged = pd.merge(rxn_data, scores, on=['Source', 'Target'])
if merged.empty:
return None
fig = go.Figure()
fig.add_trace(go.Scattergl(
x=coords_df['x'], y=coords_df['y_plot'],
mode='markers',
marker=dict(size=4, color='#bdbdbd', opacity=0.5), # All spots in background
name='Tissue Background',
customdata=np.stack((coords_df.index, coords_df['domain']), axis=-1),
hovertemplate="<b>Spot ID: %{customdata[0]}</b><br>Domain: %{customdata[1]}<extra></extra>",
showlegend=False
))
types = merged['Interaction type'].unique()
colors = px.colors.qualitative.T10
for i, t in enumerate(types):
sub = merged[merged['Interaction type'] == t]
s_coords = coords_df.loc[sub['Source'], ['x', 'y_plot']].values
t_coords = coords_df.loc[sub['Target'], ['x', 'y_plot']].values
n = len(sub)
edge_x = np.full(n * 3, np.nan)
edge_y = np.full(n * 3, np.nan)
edge_x[0::3] = s_coords[:, 0]; edge_x[1::3] = t_coords[:, 0]
edge_y[0::3] = s_coords[:, 1]; edge_y[1::3] = t_coords[:, 1]
fig.add_trace(go.Scattergl(
x=edge_x, y=edge_y,
mode='lines',
line=dict(width=3, color=INTERACTION_COLORS.get(t, "#607d8b")),
name=str(t),
hoverinfo='none', # Hover is handled by midpoints
connectgaps=False
))
# Midpoints for robust hover in the middle of lines
mid_x = (s_coords[:, 0] + t_coords[:, 0]) / 2
mid_y = (s_coords[:, 1] + t_coords[:, 1]) / 2
fig.add_trace(go.Scattergl(
x=mid_x, y=mid_y,
mode='markers',
marker=dict(size=12, opacity=0), # Large invisible target
name=str(t),
hovertemplate=f"<b>Interaction: {t}</b><br>Score: %{{customdata:.4f}}<extra></extra>",
customdata=sub['Interaction score'].values,
showlegend=False
))
active_spots = sorted(list(set(merged['Source']).union(set(merged['Target']))))
active_df = coords_df.loc[active_spots]
fig.add_trace(go.Scattergl(
x=active_df['x'], y=active_df['y_plot'],
mode='markers',
marker=dict(size=5, color='#424242', opacity=0.9, line=dict(width=1, color='white')),
name='Interacting Spots',
customdata=np.stack((active_df.index, active_df['domain']), axis=-1),
hovertemplate="<b>Spot ID: %{customdata[0]}</b><br>Domain: %{customdata[1]}<extra></extra>",
showlegend=True
))
fig.update_layout(
title=dict(
text=f"Metabolic Interactions: {selected_display_name}",
),
xaxis=dict(visible=False), yaxis=dict(visible=False, scaleanchor="x"),
plot_bgcolor='#fcfcfc', paper_bgcolor='white',
width=850, height=850, margin=dict(l=10, r=10, t=60, b=10),
legend=dict(orientation="h", y=1.02, x=0, xanchor="left", title="Interaction Type:"),
hovermode='closest',
hoverdistance=30 # Makes it easier to hover on lines
)
return fig
def create_plotly_comm_plot(interaction_scores, adata, percentile_threshold=80):
"""
Optimized Communication Strength plot using WebGL and vectorized coordinates.
"""
coords_df = pd.DataFrame(adata.obsm["spatial"], index=adata.obs.index, columns=['x', 'y'])
y_max = coords_df['y'].max()
coords_df['y_plot'] = y_max - coords_df['y']
coords_df['domain'] = adata.obs['domain'] if 'domain' in adata.obs.columns else "N/A"
if percentile_threshold > 0:
thresh = interaction_scores['Interaction score'].quantile(percentile_threshold / 100)
interaction_scores = interaction_scores[interaction_scores['Interaction score'] >= thresh]
valid = interaction_scores[
(interaction_scores['Source'].isin(coords_df.index)) &
(interaction_scores['Target'].isin(coords_df.index))
]
if valid.empty: return None
fig = go.Figure()
# Background
fig.add_trace(go.Scattergl(
x=coords_df['x'], y=coords_df['y_plot'],
mode='markers',
marker=dict(size=4, color='#bdbdbd', opacity=0.3), # All spots in background
name='Tissue Background',
customdata=np.stack((coords_df.index, coords_df['domain']), axis=-1),
hovertemplate="<b>Spot ID: %{customdata[0]}</b><br>Domain: %{customdata[1]}<extra></extra>",
showlegend=False
))
# Binned Edges (Vectorized)
n_bins = 5
valid = valid.copy()
valid['bin'] = pd.qcut(valid['Interaction score'], n_bins, labels=False, duplicates='drop')
for b in range(n_bins):
sub = valid[valid['bin'] == b]
if sub.empty: continue
s_coords = coords_df.loc[sub['Source'], ['x', 'y_plot']].values
t_coords = coords_df.loc[sub['Target'], ['x', 'y_plot']].values
n = len(sub)
edge_x = np.full(n * 3, np.nan)
edge_y = np.full(n * 3, np.nan)
edge_x[0::3] = s_coords[:, 0]; edge_x[1::3] = t_coords[:, 0]
edge_y[0::3] = s_coords[:, 1]; edge_y[1::3] = t_coords[:, 1]
fig.add_trace(go.Scattergl(
x=edge_x, y=edge_y,
mode='lines',
line=dict(width=0.5 + b*1.5, color=px.colors.sample_colorscale("Viridis", b/(n_bins-1))[0]),
name=f"Level {b+1}", hoverinfo='none'
))
fig.update_layout(
title="Cell-Cell Metabolic Communication Strengths",
xaxis=dict(visible=False), yaxis=dict(visible=False, scaleanchor="x"),
plot_bgcolor='#fcfcfc', width=850, height=850,
legend=dict(title="Score Bin:", orientation="v", x=1.02, y=1)
)
return fig
|