Update create_diffusion_dataset.py
Browse files- create_diffusion_dataset.py +34 -10
create_diffusion_dataset.py
CHANGED
|
@@ -36,6 +36,8 @@ class ProteinNetworkConversationDataset:
|
|
| 36 |
|
| 37 |
interactions = []
|
| 38 |
protein_set = set()
|
|
|
|
|
|
|
| 39 |
|
| 40 |
for idx, row in self.df.iterrows():
|
| 41 |
try:
|
|
@@ -45,7 +47,19 @@ class ProteinNetworkConversationDataset:
|
|
| 45 |
|
| 46 |
if protein_a in ['-', 'nan', ''] or protein_b in ['-', 'nan', '']:
|
| 47 |
continue
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
|
|
|
| 49 |
protein_set.add(protein_a)
|
| 50 |
protein_set.add(protein_b)
|
| 51 |
|
|
@@ -63,7 +77,7 @@ class ProteinNetworkConversationDataset:
|
|
| 63 |
except Exception:
|
| 64 |
continue
|
| 65 |
|
| 66 |
-
print(f"Extracted {len(interactions)} valid interactions")
|
| 67 |
print(f"Found {len(protein_set)} unique proteins")
|
| 68 |
|
| 69 |
return interactions, sorted(list(protein_set))
|
|
@@ -331,24 +345,29 @@ class ProteinNetworkConversationDataset:
|
|
| 331 |
# Sort for consistency
|
| 332 |
proteins = sorted(proteins)
|
| 333 |
|
| 334 |
-
# Group interactions by type
|
| 335 |
-
interactions_by_type = defaultdict(
|
| 336 |
for interaction in interactions:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 337 |
int_type = interaction.get('interaction_type', 'physical')
|
| 338 |
# Ensure consistent ordering
|
| 339 |
p1, p2 = sorted([interaction['protein_a'], interaction['protein_b']])
|
| 340 |
-
interactions_by_type[int_type].
|
| 341 |
|
| 342 |
result = f"PROTEINS: {', '.join(proteins)}\n\n"
|
| 343 |
|
| 344 |
for int_type, edges in interactions_by_type.items():
|
| 345 |
if edges:
|
| 346 |
result += f"{int_type.upper()} INTERACTIONS:\n"
|
| 347 |
-
for edge in sorted(edges):
|
| 348 |
result += f" {edge}\n"
|
| 349 |
result += "\n"
|
| 350 |
|
| 351 |
-
|
|
|
|
| 352 |
return result.strip()
|
| 353 |
|
| 354 |
def format_interactions_as_text(self, interactions):
|
|
@@ -358,17 +377,22 @@ class ProteinNetworkConversationDataset:
|
|
| 358 |
if not interactions:
|
| 359 |
return "No interactions predicted."
|
| 360 |
|
| 361 |
-
|
|
|
|
| 362 |
for interaction in interactions:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 363 |
int_type = interaction.get('interaction_type', 'physical')
|
| 364 |
p1, p2 = sorted([interaction['protein_a'], interaction['protein_b']])
|
| 365 |
-
interactions_by_type[int_type].
|
| 366 |
|
| 367 |
result = ""
|
| 368 |
for int_type, edges in interactions_by_type.items():
|
| 369 |
if edges:
|
| 370 |
result += f"{int_type.upper()} INTERACTIONS:\n"
|
| 371 |
-
for edge in sorted(edges):
|
| 372 |
result += f" {edge}\n"
|
| 373 |
result += "\n"
|
| 374 |
|
|
@@ -433,4 +457,4 @@ if __name__ == "__main__":
|
|
| 433 |
|
| 434 |
print("\nTask distribution:")
|
| 435 |
for task, count in task_types.items():
|
| 436 |
-
print(f" {task}: {count}")
|
|
|
|
| 36 |
|
| 37 |
interactions = []
|
| 38 |
protein_set = set()
|
| 39 |
+
# Use set to track unique interactions and avoid duplicates
|
| 40 |
+
seen_interactions = set()
|
| 41 |
|
| 42 |
for idx, row in self.df.iterrows():
|
| 43 |
try:
|
|
|
|
| 47 |
|
| 48 |
if protein_a in ['-', 'nan', ''] or protein_b in ['-', 'nan', '']:
|
| 49 |
continue
|
| 50 |
+
|
| 51 |
+
# Skip self-interactions
|
| 52 |
+
if protein_a == protein_b:
|
| 53 |
+
continue
|
| 54 |
+
|
| 55 |
+
# Create canonical interaction key (sorted to avoid A-B vs B-A duplicates)
|
| 56 |
+
interaction_key = tuple(sorted([protein_a, protein_b]) + [interaction_type])
|
| 57 |
+
|
| 58 |
+
# Skip if we've already seen this exact interaction
|
| 59 |
+
if interaction_key in seen_interactions:
|
| 60 |
+
continue
|
| 61 |
|
| 62 |
+
seen_interactions.add(interaction_key)
|
| 63 |
protein_set.add(protein_a)
|
| 64 |
protein_set.add(protein_b)
|
| 65 |
|
|
|
|
| 77 |
except Exception:
|
| 78 |
continue
|
| 79 |
|
| 80 |
+
print(f"Extracted {len(interactions)} valid unique interactions")
|
| 81 |
print(f"Found {len(protein_set)} unique proteins")
|
| 82 |
|
| 83 |
return interactions, sorted(list(protein_set))
|
|
|
|
| 345 |
# Sort for consistency
|
| 346 |
proteins = sorted(proteins)
|
| 347 |
|
| 348 |
+
# Group interactions by type and use sets to avoid duplicates
|
| 349 |
+
interactions_by_type = defaultdict(set)
|
| 350 |
for interaction in interactions:
|
| 351 |
+
# Skip self-interactions
|
| 352 |
+
if interaction['protein_a'] == interaction['protein_b']:
|
| 353 |
+
continue
|
| 354 |
+
|
| 355 |
int_type = interaction.get('interaction_type', 'physical')
|
| 356 |
# Ensure consistent ordering
|
| 357 |
p1, p2 = sorted([interaction['protein_a'], interaction['protein_b']])
|
| 358 |
+
interactions_by_type[int_type].add(f"{p1}--{p2}")
|
| 359 |
|
| 360 |
result = f"PROTEINS: {', '.join(proteins)}\n\n"
|
| 361 |
|
| 362 |
for int_type, edges in interactions_by_type.items():
|
| 363 |
if edges:
|
| 364 |
result += f"{int_type.upper()} INTERACTIONS:\n"
|
| 365 |
+
for edge in sorted(edges): # Sort the set for consistent output
|
| 366 |
result += f" {edge}\n"
|
| 367 |
result += "\n"
|
| 368 |
|
| 369 |
+
total_interactions = sum(len(edges) for edges in interactions_by_type.values())
|
| 370 |
+
result += f"NETWORK SUMMARY: {len(proteins)} proteins, {total_interactions} unique interactions"
|
| 371 |
return result.strip()
|
| 372 |
|
| 373 |
def format_interactions_as_text(self, interactions):
|
|
|
|
| 377 |
if not interactions:
|
| 378 |
return "No interactions predicted."
|
| 379 |
|
| 380 |
+
# Group interactions by type and use sets to avoid duplicates
|
| 381 |
+
interactions_by_type = defaultdict(set)
|
| 382 |
for interaction in interactions:
|
| 383 |
+
# Skip self-interactions
|
| 384 |
+
if interaction['protein_a'] == interaction['protein_b']:
|
| 385 |
+
continue
|
| 386 |
+
|
| 387 |
int_type = interaction.get('interaction_type', 'physical')
|
| 388 |
p1, p2 = sorted([interaction['protein_a'], interaction['protein_b']])
|
| 389 |
+
interactions_by_type[int_type].add(f"{p1}--{p2}")
|
| 390 |
|
| 391 |
result = ""
|
| 392 |
for int_type, edges in interactions_by_type.items():
|
| 393 |
if edges:
|
| 394 |
result += f"{int_type.upper()} INTERACTIONS:\n"
|
| 395 |
+
for edge in sorted(edges): # Sort the set for consistent output
|
| 396 |
result += f" {edge}\n"
|
| 397 |
result += "\n"
|
| 398 |
|
|
|
|
| 457 |
|
| 458 |
print("\nTask distribution:")
|
| 459 |
for task, count in task_types.items():
|
| 460 |
+
print(f" {task}: {count}")
|